Data-driven Multi-scalar Analysis of Spatial Morphology and Cross-scale Interactions in Traditional Settlements: A Case Study of Fulin Village, China

Data-driven Multi-scalar Analysis of Spatial Morphology and Cross-scale Interactions in Traditional Settlements: A Case Study of Fulin Village, China | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Data-driven Multi-scalar Analysis of Spatial Morphology and Cross-scale Interactions in Traditional Settlements: A Case Study of Fulin Village, China Wencan Huang, Liangliang Wang, Yixin Wang, Jie Han This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9296889/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The spatial morphology of traditional settlements results from long-term interactions between environmental constraints and socio-spatial processes, exhibiting pronounced multi-level and cross-scale characteristics. However, existing studies often remain at single-scale descriptions and lack systematic analysis of inter-scalar relationships. This study develops a multi-level analytical framework integrating settlement, parcel, and building scales, using Fulin Village in Quanzhou, China, as a case study. Machine learning and image segmentation are employed to identify spatial units and examine their structural relationships. The results show that Fulin Village presents a water-oriented clustered pattern with differentiated zones dominated by distinct building types. More importantly, clear cross-scale interactions are identified: the settlement-level structure is transmitted through parcel subdivision and organization and manifested at the building level as specific spatial forms and typological distributions. The parcel level acts as an intermediary that translates and organizes spatial structure, while the aggregation and differentiation of building types feed back to reinforce parcel organization and settlement zoning. This process reflects the coupling of environmental constraints shaping overall patterns, socio-spatial mechanisms operating through parcel organization, and building-level morphological expression across scales. This study conceptualizes traditional settlement morphology as a cross-scale mechanism of “macro-level constraints – intermediary translation – micro-level representation.” By integrating data-driven methods with morphological analysis, it provides a structured analytical approach for understanding and managing traditional settlements. Traditional Settlements Spatial Morphology Multi-Scalar Analysis Machine Learning Image Segmentation Cross-scale Mechanisms Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 1. Introduction Traditional villages in China represent important carriers of regional cultural heritage [ 1 ] , reflecting long-term interactions between human activities, natural environments, and socio-cultural systems. In the Chinese context, traditional villages generally refer to settlements that have preserved historical spatial patterns, architectural forms, and cultural practices over a long period, often with relatively intact layouts and strong continuity of local traditions [ 2 , 3 ] . Their spatial morphology embodies not only settlement patterns and architectural organization, but also the underlying logic of land use, social structure, and environmental adaptation [ 4 , 5 ] . However, under the rapid processes of urbanization and rural revitalization in China, traditional villages are increasingly facing pressures from modernization [ 6 ] . Modernization in this context refers to the transformation of rural areas driven by economic development, infrastructure expansion, and changes in construction practices, often characterized by the replacement of traditional building materials and forms with standardized, modern constructions [ 7 ] . Such processes frequently lead to the fragmentation of traditional spatial structures, the loss of morphological coherence, and the gradual disappearance of historically evolved settlement patterns. As a result, the preservation of traditional village spatial forms has become a critical issue in both academic research and planning practice. Existing studies on traditional village morphology have largely relied on qualitative interpretation and manual mapping, focusing on typological classification, spatial structure description, and historical evolution [ 4 , 5 ] . While these approaches have provided valuable insights into the cultural and spatial characteristics of traditional settlements, they are often limited by subjectivity, low efficiency, and insufficient quantitative support. In recent years, quantitative methods based on geographic information systems (GIS) [ 8 ] , spatial metrics [ 9 ] , and remote sensing [ 2 , 10 ] have been increasingly applied to analyze settlement patterns and spatial structures [ 11 ] . However, systematic data-driven approaches that integrate high-resolution spatial data and automated extraction techniques remain relatively limited, particularly at the fine-grained level of village morphology. With the rapid development of unmanned aerial vehicle (UAV) photogrammetry [ 10 , 12 ] and machine learning techniques [ 13 ] , new opportunities have emerged for the accurate acquisition and analysis of spatial morphological information. UAV-based data can provide high-resolution, up-to-date spatial information [ 10 ] , while machine learning and image segmentation methods enable efficient extraction of land use, building types, and spatial structures from complex visual data. These technologies have been widely applied in urban studies and land use classification; however, their application in the morphological analysis of traditional villages is still in an exploratory stage, especially in terms of integrating multi-scale spatial features and supporting heritage conservation practices. Despite these advancements, several challenges remain. First, many current studies lack clearly defined analytical frameworks that connect data extraction methods with spatial morphology interpretation [ 14 ] . Second, the reproducibility of machine learning-based approaches is often limited due to inadequate reporting of methodological details [ 13 ] . Third, the link between quantitative spatial analysis and heritage conservation strategies has not been fully examined, leading to a gap between technical analysis and practical application [ 15 ] . To address these issues, this study proposes a data-driven approach to spatial morphology analysis of traditional villages by integrating UAV-based data acquisition, object-based image analysis (OBIA) [ 16 ] , and machine learning techniques. Taking Fulin Village in Quanzhou, China, as a case study, the research aims to systematically extract spatial morphological elements and to examine their multi-level organization and cross-scale interactions. Specifically, this study addresses the following research questions: (1) How can UAV-based image segmentation and machine learning methods be effectively applied to extract spatial morphological elements of traditional villages? (2) What are the quantitative spatial characteristics and structural patterns of traditional village morphology across multiple scales (settlement, parcel, and building)? (3) How do cross-scale interactions among settlement, parcel, and building levels operate, and what underlying mechanisms govern their spatial coupling? By constructing a multi-dimensional spatial dataset and conducting quantitative analysis, this study seeks to improve the accuracy and efficiency of traditional village morphology recognition, and to provide a methodological framework for integrating digital technologies into heritage conservation and rural planning practices. 2. Literature Review 2.1 Spatial morphology of traditional settlements The concept of spatial morphology originates from human geography [ 17 ] and has been progressively extended across multiple disciplines, including geography [ 18 , 19 ] , architecture [ 20 – 22 ] , urban and rural planning [ 23 – 26 ] , and spatial political economy [ 27 – 29 ] . In the context of traditional settlements, spatial morphology refers to the physical organization of space formed under the combined influence of natural conditions and socio-cultural factors, encompassing building layouts, road networks, land-use structures, and their interactions and evolutionary processes [ 18 , 19 , 30 ] . Existing studies indicate that research on traditional settlement morphology in China has developed into a relatively systematic body of knowledge through interdisciplinary integration, covering spatial distribution patterns, morphological composition, evolution processes, and cultural implications [ 31 ] . In terms of methodological development, studies on spatial morphology have gradually shifted from qualitative description to quantitative analysis [ 32 ] . Early research primarily relied on field surveys and graphical interpretation, focusing on vernacular dwelling types, construction techniques, and regional cultural characteristics [ 33 – 36 ] . With the expansion of research scale, the focus has moved from individual buildings to the overall spatial structure of settlements [19]. In recent years, the development of quantitative methods has enabled researchers to employ various indicators to characterize spatial morphology, such as building size, orientation, and spacing [ 37 ] , two-dimensional morphological metrics to quantify overall form [ 11 , 38 ] , and space syntax to construct topological models of spatial organization [ 39 ] . Parametric modeling approaches have also been introduced to simulate and reconstruct settlement morphology [ 40 ] . In addition, dynamic simulation methods, including system dynamics, cellular automata (CA), and agent-based models (ABM), have been applied to explore settlement evolution and land-use change processes [ 41 , 42 ] , often in combination with remote sensing data for spatial prediction and historical reconstruction [ 43 , 44 ] . Despite these advances, existing studies remain largely focused on morphological feature extraction, with limited integration of spatial structure, function, and cultural interpretation. 2.2 Multi-scalar and hierarchical spatial analysis Traditional settlements are complex spatial systems characterized by significant multi-scalar and hierarchical structures. At the macro level, settlement location and overall spatial patterns are strongly constrained by natural environmental conditions and regional contexts [ 18 , 45 ] . At the meso level, parcel subdivision and spatial organization reflect social structures, institutional arrangements, and property relations [ 23 , 28 ] . At the micro level, building typologies and spatial configurations embody local cultural traditions and construction practices [ 33 , 34 ] . Recent studies have increasingly incorporated hierarchical perspectives into spatial analysis. For example, space syntax has been widely used to investigate spatial cognition and organizational logic based on road network structures [ 24 ] , while quantitative urban morphology research has proposed multi-indicator systems to describe hierarchical spatial structures [ 25 , 26 ] . Furthermore, institutional and political-economic approaches have been applied to explain the evolution of spatial forms through mechanisms such as transaction costs and property rights allocation [ 28 , 29 ] . However, current multi-scalar studies still exhibit notable limitations. On the one hand, different spatial scales are often analyzed in isolation, lacking an integrated analytical framework. On the other hand, the transmission mechanisms across scales remain unclear, making it difficult to explain how macro-level spatial patterns are translated into micro-level morphological characteristics. Therefore, there is a need to develop an integrated multi-scalar analytical approach to reveal the hierarchical organization of traditional settlement morphology. 2.3 Data-driven approaches and machine learning With the advancement of remote sensing and geospatial technologies, data-driven approaches have become increasingly important in the study of traditional settlement morphology. In particular, machine learning and deep learning techniques have been widely applied to spatial element extraction and pattern recognition. At the macro level, studies commonly combine remote sensing imagery with geospatial data to identify settlement patterns using machine learning methods [ 46 , 47 ] . For instance, machine learning has been employed to classify village morphology and spatial scales [ 48 ] , clustering algorithms have been used to analyze landscape texture patterns [ 48 ] , and deep learning techniques have been applied to recognize spatial textures in historical settlements [ 49 ] . In addition, some studies focus on environmental drivers of spatial morphology, analyzing influencing factors and formation mechanisms [ 50 , 51 ] , while tools such as the Geodetector model have been used to examine spatial distribution patterns and their determinants [ 52 ] . At the micro level, machine learning techniques are widely used for building detection and morphological extraction. High-resolution remote sensing images and UAV data enable the application of semantic segmentation and multi-task learning models for building footprint extraction and spatial feature identification [ 53 – 55 ] . Other studies have explored building typology classification and spatial relationships based on plan data [ 56 , 57 ] . At the façade level, deep learning models have been applied to component recognition, architectural style classification, and damage detection [ 58 – 63 ] , often integrating three-dimensional data for detailed analysis and condition assessment [ 64 ] . Additionally, oblique photogrammetry combined with deep learning has been used to achieve multi-scale integration and automatic classification of building characteristics [ 65 ] . Overall, data-driven methods have significantly improved the efficiency and accuracy of spatial morphology analysis. However, most studies remain task-specific or scale-specific, lacking systematic integration with comprehensive spatial morphological frameworks. 2.4 Research gaps and objectives Despite substantial progress in spatial morphology analysis and data-driven approaches, several critical limitations remain. First, existing studies are predominantly conducted at a single spatial scale, with insufficient attention to the hierarchical organization of settlement space. In particular, the interactions among settlement, parcel, and building levels are rarely examined in an integrated manner, limiting the understanding of cross-scale spatial processes [ 18 , 39 ] . Second, although machine learning and image-based techniques have been widely applied to spatial recognition and feature extraction, their role remains largely technical. The integration of data-driven methods with morphological analysis and mechanism interpretation is still limited, resulting in a gap between quantitative extraction and theoretical explanation [ 45 , 52 ] . Third, the intermediary role of parcel-level spatial units has been largely overlooked. As a critical link between macro-level settlement structure and micro-level building form, parcels play a key role in mediating spatial organization. However, their function in transmitting, transforming, and regulating spatial morphology across scales has not been sufficiently explored [ 31 , 66 ] . As a result, there is still a lack of a unified and operational analytical framework capable of capturing multi-scale spatial characteristics and explaining their underlying generative mechanisms. The main contributions of this study are threefold: (1) proposing a multi-level spatial morphology framework that strengthens the structural linkage across different spatial scales; (2) integrating machine learning techniques to achieve automated identification and quantitative analysis of spatial elements; (3) revealing the underlying mechanisms of traditional settlement morphology from a multi-scalar perspective. 3. Study Area Fulin Village is located in Quanzhou City, Fujian Province, in southeastern China, a region characterized by a subtropical monsoon climate and a long history of settlement development. The village is situated within a hilly coastal landscape, where natural topography and water systems have significantly influenced the formation of its spatial structure. As of the latest survey, Fulin Village covers an area of approximately 99.13 hectares and includes approximately 2200 households. The built environment consists of both traditional and modern constructions, among which traditional buildings account for approximately 56% of the total building stock. These traditional structures are mainly composed of Minnan-style residences, such as "Dacuo" and "Fanzai buildings" [ 67 ] which are characterized by enclosed courtyards, axial layouts, and strong spatial organization. The spatial morphology of Fulin Village exhibits a relatively well-preserved settlement pattern, including clustered building layouts, hierarchical road networks, and distinct functional zoning. At the same time, the village has experienced varying degrees of transformation due to recent rural development, leading to the coexistence of traditional and modern spatial elements. Fulin Village was selected as the case study for this research for the following reasons. First, it retains a representative spatial morphology typical of traditional villages in southern Fujian, making it suitable for morphological analysis. Second, the village shows clear spatial heterogeneity due to ongoing modernization processes, providing an appropriate context to examine the interaction between traditional and contemporary spatial structures. Third, high-resolution UAV data are available for the study area, enabling detailed spatial analysis based on image segmentation and machine learning techniques. To facilitate spatial analysis, the study area boundary was not defined strictly according to administrative limits, but instead delineated based on the extent of traditional settlement morphology and its functional relevance. Specifically, the study focuses on the historic village area with a high concentration of traditional buildings, together with an approximately 200-meter southward extension into adjacent farmland. This extended area corresponds to historically documented agricultural zones identified through field investigation, which are closely associated with the production–living spatial structure of the village. The total study area defined in this research is approximately 47.71 hectares (Fig. 1 ). 4. Methodology This study proposes a data-driven framework for the spatial morphology analysis of traditional villages by integrating UAV-based data acquisition [ 10 , 12 ] , object-based image analysis (OBIA) [ 16 ] , and machine learning techniques. Compared with conventional qualitative approaches, which rely on manual interpretation and typological description, the proposed method enables automated extraction, quantitative analysis, and multi-scale segmentation and characterization of spatial morphological elements. The overall workflow of the study consists of three main steps:(1) UAV data acquisition and preprocessing; (2) Multi-scale spatial morphology segmentation and feature extraction; (3) Quantitative spatial analysis. The workflow is illustrated in Fig. 2 . 4.1 UAV Data Acquisition and Preprocessing High-resolution spatial data of Fulin Village were acquired using unmanned aerial vehicle (UAV) oblique photogrammetry, which enables detailed three-dimensional reconstruction of complex settlement environments [ 12 ] . The UAV survey was conducted using a multi-rotor UAV platform equipped with a high-resolution digital camera [ 10 ] . The flight was carried out at an average altitude of approximately 80m, with a ground sampling distance (GSD) of approximately 5 cm, ensuring sufficient spatial resolution for fine-scale morphological analysis. To ensure data integrity and image quality, the flight mission was designed with a 70% overlap rate both longitudinally and laterally. Multi-angle oblique images were collected to capture building facades and complex spatial structures. The total number of images acquired was approximately 2485, covering the entire built-up area of the village. The collected images were processed using photogrammetric software (Context Capture ) to generate high-resolution orthophotos and digital surface models (DSM). The processing workflow included image alignment, sparse point cloud generation, dense point cloud reconstruction, mesh generation, and texture mapping. Subsequently, preprocessing was conducted to improve data quality and ensure consistency for further analysis. This included geometric correction, coordinate system unification (WGS84), and noise reduction. The resulting orthophoto and DSM datasets provided the fundamental spatial data for image segmentation and machine learning-based analysis. 4.2 Multi-scale spatial morphology segmentation and feature extraction To accurately extract spatial morphological elements at different hierarchical levels, a multi-scale segmentation and classification approach was adopted in this study. Specifically, the method includes: (1) Settlement-scale segmentation based on orthophoto and DSM data, focusing on the extraction of overall spatial patterns and land-use structure. (2) Parcel-scale extraction and classification , which serves as an intermediate layer linking settlement structure and building form. Given the lack of explicit parcel boundaries, parcels were delineated based on the spatial aggregation of building types. A density surface was generated from building clusters, and equal-density contour lines were used to approximate parcel boundaries. This approach defines parcels as emergent spatial units and supports the analysis of their mediating role in cross-scale interactions. (3) Building-scale feature extraction based on three-dimensional models, focusing on the identification of architectural elements such as roofs and facades. 4.2.1 Settlement-level Segmentation Based on OBIA At the settlement scale, spatial morphology was extracted through object-based image analysis (OBIA) using high-resolution orthophotos and digital surface models (DSM) [ 16 ] . Multi-resolution segmentation was performed in eCognition 9.0 software to partition the study area into homogeneous spatial objects representing land use and settlement texture. The segmentation process considered both spectral and elevation information, integrating RGB bands from orthophotos and height information from the DSM. The segmentation parameters were set as follows: scale parameter = 50, shape factor = 0.3, and compactness = 0.5. These parameters were determined through iterative testing to balance segmentation detail and object integrity. Based on the segmented objects, a supervised classification approach was implemented. Training samples were manually labeled according to major land use and spatial morphology categories, including building areas, roads, vegetation, water, and open spaces. A Random Forest classifier was applied to classify the segmented objects using a combination of spectral, textural (e.g., RGB), and geometric features (e.g., area, shape index). This process enabled the extraction of the overall spatial structure and planar morphological patterns of the village, providing a basis for analyzing settlement texture and land use organization. 4.2.2 Building-level Feature Extraction Based on Deep Learning At the façade morphology dimension, this study develops a deep learning–based workflow for building feature extraction and typological classification, using a three-dimensional reconstruction model generated from UAV photogrammetry as the primary data source [ 10 ] . This approach enables detailed analysis of roof structures, façade compositions, and architectural components (Fig. 4 ). First, during the data preprocessing stage, traditional building imagery was manually annotated using Label Studio, focusing on key morphological features such as roof types and courtyard (patio) configurations to construct the initial training dataset. On this basis, the Segment Anything Model (SAM) was employed to perform large-scale segmentation of building data in Fulin Village, extracting the outlines of roofs and courtyard spaces. By integrating morphological parameters—including roof height, number of pitched roofs, aspect ratio, and number of courtyards—an initial classification of major building typologies was achieved, including traditional courtyard-based “Dacuo”, multi-storey “Fanzai buildings”, and arcade-style buildings. Second, for façade element extraction, individual building models were isolated from the 3D dataset based on spatial geometric corner detection. Orthographic projection was then applied to generate distortion-free façade images [ 68 ] . Based on these images, key façade elements—such as doors, windows, and decorative components—were manually annotated to construct a standardized façade dataset [ 69 ] . After normalization and noise reduction, a convolutional neural network (CNN)-based semantic segmentation model was trained using supervised learning to improve the recognition accuracy of complex façade elements. During model training, a labeled dataset consisting of 64 building samples was used, with 70% allocated for training and 30% for validation. In addition, segmentation masks generated by SAM were incorporated to assist model optimization, enhancing robustness under complex geometric conditions. Model performance was evaluated using accuracy and Intersection over Union (IoU) metrics to ensure the reliability and consistency of feature extraction results. Based on the segmentation outputs, extracted elements were further aggregated and interpreted in combination with field survey data. Building morphology was then characterized across multiple dimensions, including roof form, façade composition, and spatial configuration. Representative samples were selected for standardized reconstruction, leading to the development of a systematic morphological atlas of building types. This atlas establishes a structured analytical workflow of “image segmentation–feature extraction–typological classification,” providing a consistent data foundation for subsequent quantitative analysis and cross-type comparison. Finally, a dual-dimensional analytical framework integrating both plan and façade perspectives was established at the building level. By combining horizontal spatial structure (plan-based features) and vertical interface characteristics (façade elements), this framework enables a more comprehensive representation of building morphology and supports multi-scalar spatial analysis. 4.3 Quantitative Spatial Analysis To systematically examine the spatial morphology of traditional villages and their multi-scalar structural relationships, this study establishes a three-level analytical framework encompassing the settlement, parcel, and building scales. Within this framework, spatial characteristics are quantitatively analyzed across three dimensions—size, shape, and configuration—through a set of unified indicators, enabling the standardized representation and typological interpretation of morphological features. 4.3.1 Definition of Morphological Indicators To ensure methodological consistency and comparability across different spatial scales, a set of core morphological indicators is defined and consistently applied throughout the analysis. (1) Aspect Ratio (AR) The aspect ratio is used to characterize the directional tendency and elongation of spatial units, calculated based on the minimum bounding rectangle [ 70 ] $$\:AR=\frac{L}{W}$$ where L and W represent the length of the longer and shorter sides of the bounding rectangle, respectively. This indicator is applicable at the settlement, parcel, and building levels, and is used to distinguish between elongated and compact spatial forms. (2) Shape Index (SI) The shape index measures the complexity and regularity of spatial unit boundaries [ 71 ] $$\:SI=\frac{P}{2\sqrt{\pi\:A}}$$ where P denotes the perimeter and A the area of the spatial unit. Values approaching 1 indicate more regular shapes, while higher values reflect increased boundary complexity and irregularity. This indicator is applicable across multiple spatial scales. (3) Land Use Proportion (LUP) The land use proportion describes the composition of different functional land uses within the study area [ 72 ] $$\:LU{P}_{i}=\frac{{A}_{i}}{{A}_{total}}$$ where \(\:{A}_{i}\) represents the area of land use type i , and \(\:{A}_{total}\) is the total area of the study region. This indicator is used to analyze the functional structure of the settlement. (4) Spatial Aggregation (Kernel Density) Spatial aggregation is assessed using kernel density estimation (KDE) to measure the concentration of building distribution. By analyzing the spatial density of building centroids, this method identifies core areas and overall distribution patterns within the settlement [ 73 ] $$\:f\left(x\right)=\frac{1}{nh}\sum\:_{i=1}^{n}K\left(\frac{x-{x}_{i}}{h}\right)$$ where f(x) represents the estimated density at location x , \(\:{x}_{i}\) denotes the location of each building centroid, n is the number of samples, h is the bandwidth, and K is the kernel function. The resulting density surface provides an intuitive representation of spatial clustering and dispersion patterns. 4.3.2 Multi-scalar Spatial Morphology Analysis Building upon the unified indicator system, spatial morphology is analyzed across three hierarchical levels—settlement, parcel, and building. Considering the differences in functional attributes and morphological characteristics across scales, a consistent analytical framework based on size, shape, and configuration is adopted, with scale-specific indicators selected to capture the key features at each level. (1) Settlement Level The settlement level focuses on overall spatial patterns and land-use structure. Size dimension : Land use proportion (LUP) is used to quantify the composition of different functional spaces and to characterize the overall land-use structure of the settlement. Shape dimension : Aspect ratio and shape index are applied to describe the overall form and boundary complexity of the settlement. Configuration dimension : Kernel density analysis is used to identify spatial clustering patterns and core areas of building distribution, revealing the overall spatial structure. Through the integration of these indicators, the settlement-level analysis captures the general spatial organization and morphological characteristics of the village. (2) Parcel Level The parcel level serves as an intermediate scale linking the overall settlement structure with individual buildings, reflecting patterns of spatial subdivision and organization. Size dimension : Parcel area is analyzed to examine the distribution of spatial unit sizes. Shape dimension : Aspect ratio and shape index are used to characterize the regularity and directional properties of parcel forms. Configuration dimension : Kernel density of buildings within parcel is used to assess internal spatial organization and identify different patterns of land use and spatial arrangement. This level of analysis helps reveal the internal logic of spatial subdivision and organization within the traditional village. (3) Building Level The building level focuses on the morphological characteristics of individual units and their spatial expression. Size dimension : The building footprint area is used to analyze size distribution. Shape dimension : Aspect ratio and shape index are applied to describe plan form characteristics. Configuration dimension : Façade element extraction and their compositional relationships are analyzed to identify patterns of façade organization and overall spatial configuration. This level provides a quantitative basis for understanding individual building characteristics and their role in shaping the overall spatial morphology. 4.4 Methodological Comparison and Applicability Compared with traditional qualitative methods, the proposed approach improves efficiency and reduces subjectivity by enabling automated extraction and quantitative analysis of spatial elements. While qualitative methods are effective in interpreting cultural meanings, they are limited in handling large-scale spatial data. Although the proposed method is tested in Fulin Village, similar approaches have been widely applied in remote sensing classification and urban morphology analysis, demonstrating their adaptability to different spatial contexts. However, the effectiveness of the model may vary depending on image quality, settlement complexity, and regional characteristics, which should be considered in future applications. 5. Results of quantitative analysis of spatial patterns of traditional settlements 5.1 Settlement-level Spatial Morphological Characteristics In the study of traditional village morphology, the settlement level reflects the overall spatial structure and pattern [ 74 ] , serving as the macro-scale foundation for multi-scalar analysis. To reveal the overall spatial organization of Fulin Village, this section examines its morphological characteristics from three aspects: land-use composition, boundary form, and spatial configuration. 5.1.1 Land-use Composition Based on the statistical results of land use proportion (LUP) (Fig. 5 ), the spatial composition of Fulin Village is dominated by built-up land and ecological land. Specifically, built-up land covers approximately 25.3 hectares, including buildings, roads, and public spaces. Cultivated land accounts for approximately 8.05 hectares, green space for 12.3 hectares, and water bodies for 2.06 hectares. In terms of spatial distribution, both built-up land and cultivated land are primarily arranged along the water system, which forms a clear spatial boundary within the settlement. The road network is largely aligned along one side of the water system, while internal roads exhibit a relatively dispersed pattern. Overall, the land-use structure presents an interwoven spatial pattern of production, living, and ecological spaces. 5.1.2 Overall Settlement Form According to the morphological indicator analysis (Fig. 6), the overall form of Fulin Village is characterized by irregularity and relative compactness. The shape index (SI) is 2.24, indicating a boundary that is more complex than that of regular geometric forms [ 71 ] . In terms of aspect ratio, the settlement exhibits an AR value of 1.22, suggesting a relatively balanced configuration. The overall morphology is therefore predominantly compact and cluster-like. At the same time, localized extensions can be observed along the direction of the water system, resulting in a composite pattern characterized by a dominant clustered form with partial linear extensions. 5.1.3 Spatial Configuration and Aggregation Patterns Kernel density analysis (Fig. 7) reveals a clear spatial clustering pattern in the distribution of buildings [ 73 ] . High-density areas are concentrated in the central part of the village, with a secondary cluster located in the southwestern area, while peripheral zones exhibit relatively lower density. Overall, building density shows a decreasing gradient from the center toward the outskirts, indicating a single-core dominant spatial structure. Analysis of nearest-neighbor distances (Fig. 8) further supports this pattern. Building spacing ranges from a minimum of 4.08 m to a maximum of 58.7 m, with an average distance of 14.93 m. The central area is characterized by relatively small inter-building distances, while spacing gradually increases toward the boundary, reflecting a transition from compact to more dispersed spatial arrangements. 5.1.4 Summary Based on the analysis of land-use composition, overall form, and spatial configuration, the settlement-level characteristics of Fulin Village can be summarized as follows: (1) The land-use structure is dominated by built-up and ecological land, with spatial distribution closely associated with the water system; (2) The overall morphology is predominantly compact, with a complex boundary and localized directional extensions; (3) The spatial structure exhibits a single-core clustering pattern, with building density decreasing from the center toward the periphery. These results indicate that Fulin Village presents clear spatial differentiation and clustering characteristics at the settlement level. However, such macro-scale patterns are not formed independently, but are supported by spatial organization at the intermediate scale. Therefore, further analysis at the parcel level is required to examine variations in size, shape, and configuration across different spatial units, in order to better understand the formation logic of the overall settlement structure. 5.2 Parcel-level Spatial Morphological Characteristics As an intermediate scale linking the overall settlement structure and individual buildings, parcel represent the fundamental units organizing village spatial structure [ 75 ] . The spatial differentiation and clustering observed at the settlement level are essentially the result of the spatial combination and evolution of heterogeneous parcel units [ 72 ] . Variations in parcel size and morphology, when aggregated spatially, further shape the functional zoning and structural hierarchy of the settlement. Based on the distribution of building typologies and their spatial clustering patterns, combined with parcel-scale morphological indicators, this section identifies spatial zoning patterns and examines parcel-level characteristics in Fulin Village. 5.2.1 Spatial Zoning and Parcel Size Characteristics Based on the automatically identified building typologies (Fig. 9 ), spatial zoning at the parcel level was delineated through an integrated assessment of dominant building types, the spatial extent of high kernel density areas, and parcel size distribution patterns. The results indicate that the study area can be classified into three representative zones: (1) a core residential zone dominated by Dacuo buildings; (2) a street-oriented commercial zone characterized by arcade buildings; (3) a peripheral residential zone dominated by Fanzai buildings. The outer expansion area, primarily composed of modern self-built houses, is excluded from this analysis. In terms of parcel size, significant variation is observed across different zones. The core Dacuo zone is characterized by parcel of moderate size, reflecting relatively stable courtyard-based spatial units. In contrast, parcel in the arcade building zone are generally smaller and more concentrated, indicating higher land-use intensity along street frontages. The Fanzai buildings zone exhibits larger parcel sizes, suggesting a more spacious spatial organization. From a spatial perspective, these zones display distinct clustering patterns. The Dacuo zone is concentrated in the central area of the village, the arcade zone extends along primary road corridors, and the Fanzai buildings zone is mainly distributed toward the settlement periphery. 5.2.2 Parcel Morphological Characteristics Parcel morphology shows clear differentiation across zones based on aspect ratio and shape index analysis (Fig. 10 ). Parcel in the core Dacuo zone are predominantly regular and rectangular, indicating a high degree of geometric order. In the arcade zone, parcel exhibit relatively low aspect ratios, resulting in compact and regular forms. In contrast, parcel in the Fanzai buildings zone display more elongated and irregular configurations, reflecting increased morphological variability. Overall, parcel morphology demonstrates a transition from regular and standardized forms in the core area to more diverse and irregular forms toward the periphery. 5.2.3 Building Density and Spatial Configuration Building density reflects the internal spatial organization of parcel [ 76 ] . Kernel density analysis indicates clear differences in spatial configuration among the identified zones. The core Dacuo zone exhibits high density with strongly clustered building distributions. The arcade zone is characterized by high building coverage and continuous linear arrangements along streets, resulting in a compact spatial configuration. The Fanzai buildings zone presents an intermediate density level, with a mixed pattern of clustering and dispersion. In contrast, peripheral areas show relatively low density and more dispersed distributions. Overall, building distribution within parcel exhibit a gradient transition from high-density clustering to more dispersed arrangements. 5.2.4 Summary Based on spatial zoning and multi-dimensional analysis at the parcel level, the following characteristics can be identified: (1) Parcel space exhibits clear zoning patterns, with distinct clusters corresponding to different spatial functions and variations in parcel size; (2) Parcel morphology transitions from regular rectangular forms in the core area to more diverse and irregular forms toward the periphery; (3) Spatial configuration shows a gradient change from high-density clustering to relatively dispersed arrangements. These results indicate that significant differences exist among parcel zones in terms of size, morphology, and building density, forming a spatial organization pattern that transitions from courtyard-based layouts to street-oriented compact forms and further to more loosely structured configurations. This pattern provides the basis for further analysis at the building level, where variations in scale, proportion, and morphology can be examined in greater detail. 5.3 Building-level Spatial Morphological Characteristics As the fundamental components of parcel organization, individual buildings play a decisive role in shaping internal spatial structure and land-use patterns. Variations in building type, size, and morphology, through their spatial arrangement within parcel, further influence parcel-level configuration and collectively contribute to the overall settlement form. The differentiated organizational patterns identified at the parcel level are essentially derived from the spatial distribution and combination of different building typologies [ 77 ] . Therefore, this section examines building-level spatial morphology in Fulin Village from both plan-based metrics and façade characteristics. 5.3.1 Size and Typological Characteristics Based on building typology classification, traditional buildings in Fulin Village mainly include five-bay courtyard houses (Dacuo), three-bay courtyard houses, Fanzai buildings, and arcade buildings. The dataset comprises 32 five-bay Dacuo, 10 three-bay Dacuo, approximately 61 Fanzai buildings, and 11 arcade buildings. In terms of size (Fig. 11 , left), clear differences can be observed among building types. Five-bay Dacuo generally range from 250–450 m², representing the largest building type. Three-bay Dacuo are primarily distributed between 100–180 m², indicating a relatively smaller scale. Fanzai buildings and multi-storey courtyard buildings are concentrated within 150–230 m², while arcade buildings typically range from 150–260 m², representing a moderate scale. Overall, building types exhibit a clear hierarchical differentiation in footprint area. 5.3.2 Plan Morphological Characteristics In terms of plan morphology (Fig. 11 , right), aspect ratio varies across building types. Five-bay Dacuo show an average aspect ratio of 1.21 (± 0.2), indicating near-rectangular forms. Three-bay Dacuo exhibit a higher value of 1.43 (± 0.4), reflecting a moderate directional tendency. Fanzai buildings and multi-storey courtyard buildings have an average aspect ratio of 1.31 (± 0.6), suggesting relatively balanced forms with greater variability. In contrast, arcade buildings present a significantly higher aspect ratio of 2.10 (± 2.0), indicating pronounced linear elongation. Overall, building plans are predominantly rectangular, with a transition in aspect ratio from relatively balanced to more elongated forms across different building types. 5.3.3 Façade Typology and Morphological Variation Building on the analysis of plan size and proportion, façade characteristics were further examined to construct a typological atlas of building morphology [ 68 , 78 ] (Fig. 12 ). Dacuo buildings are generally characterized by single-storey, axially symmetrical compositions with a tripartite vertical structure consisting of base, wall body, and roof. Roof forms commonly include curved ridge types, while façade elements such as lattice windows, wooden doors, and stone or carved decorative components are frequently observed. These buildings exhibit high symmetry, relatively low opening ratios, and enclosed spatial expressions. Fanzai buildings demonstrate noticeable variation in façade composition. The transition from single-storey to multi-storey forms introduces additional elements such as balconies and colonnades. Western-influenced features—including classical columns, pediments, arched openings, and metal railings—are incorporated, resulting in increased façade openness and greater diversity in decorative expression. Arcade buildings further emphasize continuity along the street interface. Their façades are characterized by open ground floors and continuous colonnades, forming linear and interconnected street-front spaces. Based on façade elements, opening ratios, and compositional characteristics, building façades in Fulin Village can be categorized into three typological groups: traditional, hybrid, and street-oriented. These categories reflect observable variation in façade composition across building types. 5.3.4 Summary Based on the analysis of building size, plan morphology, and façade configuration, the following characteristics can be identified at the building level: (1) Building types are diverse, with clear differentiation in size among different typologies; (2) Plan forms are predominantly rectangular, with variations in aspect ratio across building types; (3) Façade configurations differ in terms of element composition and opening patterns. These results indicate that building-level differences in size, shape, and configuration are significant and form the basis for spatial variation across higher levels. The spatial distribution and morphological differentiation of building types contribute to the organization of parcel and reinforce the spatial structure observed at the settlement level. 5.4 Summary of Results The integrated analysis across the settlement, parcel, and building levels reveals a clear multi-scalar organizational logic in the spatial morphology of Fulin Village. At the settlement level, patterns of spatial clustering and functional differentiation are observed, which correspond to variations in parcel size and organizational forms. At the parcel level, morphological characteristics are closely associated with the composition and distribution of building types. At the building level, differences in size, proportion, and configuration further contribute to spatial variation across higher levels. Taken together, these results indicate that spatial morphology in Fulin Village is characterized by hierarchical differentiation across scales, forming an interconnected system linking structure, organization, and form. The multi-level analytical framework developed in this study provides a structured approach for the identification, classification, and analysis of traditional village morphology, and establishes a consistent basis for further discussion on spatial patterns and their implications. 6. Discussion 6.1 Natural controls on spatial sequencing The distance gradient analysis and land-use statistics (Fig. 5 ) reveal a clear spatial differentiation between farmland and residential land in relation to the water system. Farmland is strongly concentrated in close proximity to water, whereas residential buildings are predominantly distributed within intermediate distance ranges. Together, these patterns form a sequential spatial structure extending outward from the water system. In addition, the ratio between built-up land and farmland (approximately 3.2:1) indicates a relatively balanced configuration between living and production spaces at the settlement scale. Rather than reflecting direct environmental determinism, this pattern suggests that natural factors operate indirectly by shaping locational choices across different land-use types. Specifically, farmland exhibits a strong dependence on water accessibility, as irrigation efficiency decreases with distance. In contrast, residential areas are positioned to balance accessibility to water resources with the need to avoid environmental risks, particularly flooding. This divergence implies that different spatial elements respond to the same environmental gradient through differentiated optimization strategies, resulting in a structured spatial sequence. Further evidence can be observed in the northern part of the village, where portions of farmland have transitioned into green space. This shift suggests that when water accessibility declines or irrigation efficiency becomes insufficient, land use may gradually shift from production-oriented to ecological functions. Such transformations highlight the long-term and dynamic influence of natural conditions on spatial organization. Overall, the spatial configuration of Fulin Village can be interpreted as an outcome of a “distance–risk trade-off” mechanism, in which water-related accessibility and environmental constraints jointly shape the allocation of land uses. While previous studies have described traditional settlements as “water-oriented”, this study advances the understanding by quantitatively demonstrating how such relationships manifest as measurable spatial gradients. 6.2 Social structure and spatial organization The spatial distribution of buildings in Fulin Village exhibits a clear core–periphery pattern, characterized by high-density clustering in the central area and a gradual decrease toward the periphery. This central zone corresponds closely with the distribution of traditional Dacuo buildings, while Fanzai buildings and arcaded houses are primarily located in outer areas. Quantitative results further show that the central area not only has the highest kernel density values but also features more regular and compact parcel configurations. In contrast, peripheral areas display greater variability in parcel size and more irregular morphological characteristics. These differences suggest that central spaces are subject to stronger spatial control, whereas peripheral zones reflect more flexible and adaptive development processes. This pattern can be attributed to the influence of social structure, particularly the role of clan-based organization in traditional settlements. By occupying central locations, dominant social groups establish a stable spatial framework that persists over time. Subsequent development tends to occur through infill within the existing structure or expansion along its edges, reinforcing the original spatial hierarchy. From a temporal perspective, the observed spatial configuration reflects strong path dependency. The persistence of central dominance and peripheral expansion indicates that historical spatial arrangements continue to constrain and guide later development. As a result, spatial organization in Fulin Village can be understood as a “core control–peripheral growth” model. This finding aligns with existing research emphasizing the role of social organization in shaping settlement structure. However, by integrating kernel density analysis with parcel-level morphological metrics, this study provides a more explicit and quantifiable account of how social structure is translated into spatial form, both in terms of distribution patterns and morphological characteristics. 6.3 Building typology-driven spatial differentiation and density restructuring The spatial distribution of building typologies (Fig. 10 ) reveals clear clustering patterns, forming distinct functional zones within the village. Traditional Dacuo are concentrated in the central area, arcade buildings are located along key transportation nodes near the village entrance and water system, while Fanzai buildings occupy transitional zones between these two areas. Parcel-level analysis further demonstrates systematic differences among these zones in terms of size, morphology, and spatial configuration. Arcade zones are characterized by the smallest parcel sizes and the highest building coverage, indicating intensive land use. In contrast, Fanzai zones exhibit relatively larger parcel and more flexible spatial arrangements, while Dacuo zones are associated with moderate parcel sizes and more regular morphological forms. These findings suggest that building typology is not only a morphological outcome but also an active factor shaping parcel organization. This process can be interpreted as a hierarchical transformation from typology to spatial structure: building types, through spatial clustering, generate distinct zone configurations, which in turn influence the overall settlement pattern. For instance, the continuous high-density arrangement of arcade buildings reinforces linear street interfaces, whereas the relatively dispersed distribution of Fanzai buildings results in more open spatial configurations. While previous studies have emphasized function-driven spatial differentiation, this study extends the discussion by demonstrating how such differentiation can be identified through data-driven approaches, combining automated typology classification with density-based spatial analysis. 6.4 Morphological transformation from ritual order to functional adaptation The analysis of building morphology reveals significant variation across typologies in terms of plan proportions, spatial organization, and façade composition. Traditional Dacuo are characterized by enclosed courtyard layouts, symmetrical organization, and relatively stable aspect ratios. In contrast, Fanzai buildings and arcade structures exhibit greater variability in both plan proportion and façade openness. Quantitatively, Dacuo buildings show limited variation in aspect ratio, indicating strong formal constraints, whereas arcade buildings display elongated forms and continuous street-facing interfaces, reflecting increased adaptation to external spatial conditions. These differences suggest that morphological variation is not random but closely associated with shifts in spatial organization. This process can be understood as a transition from ritual-based spatial order to function-oriented adaptation. Under clan-based social structures, building forms emphasize internal hierarchy and enclosure. With the intensification of commercial activities and changing social conditions, building morphology increasingly responds to functional requirements such as ventilation, lighting, and accessibility. This interpretation is consistent with previous findings describing a shift from inward-oriented to outward-facing spatial forms. However, by employing quantitative morphological indicators and typological classification, this study provides a measurable basis for understanding this transformation. 6.5 Reconfiguration of local building traditions and external influences Façade typology analysis indicates a high degree of consistency in overall proportions and compositional frameworks across different building types, while variation is primarily expressed through localized elements such as openings, balconies, and external corridors. This pattern of “overall stability with localized variation” is consistently observed across the dataset. Further examination of façade elements shows that external influences are mainly incorporated at the interface level, without fundamentally altering the underlying structural logic. For example, Fanzai buildings and arcade buildings increase façade openness and introduce additional spatial elements to enhance usability, while maintaining basic compositional relationships inherited from traditional architecture. This suggests that the integration of external elements is not a process of simple addition, but rather a selective adaptation constrained by existing construction systems. The process can be interpreted as a form of functional filtering and structural adaptation, whereby new elements are incorporated only when they improve spatial performance without disrupting the established framework. This finding aligns with previous studies highlighting the adaptive capacity of local building traditions. By quantifying façade elements and compositional characteristics, this study provides a more structured account of how such adaptation occurs at the morphological level. 6.6 Summary The above analysis demonstrates that the spatial morphology of Fulin Village is not determined by a single factor, but emerges from the interaction of multiple processes across different scales. Natural factors shape spatial sequences through distance-related constraints, social structures influence spatial organization through hierarchical occupation patterns, and building typologies contribute to morphological differentiation through their spatial distribution and configuration. Overall, this process can be conceptualized as a multi-scalar interaction framework linking environmental constraints, social organization, and typological transformation. The relationships across scales are established through quantifiable indicators, resulting in a structured rather than random spatial evolution. The identified cross-scale mechanism provides insights for heritage conservation by highlighting the importance of maintaining not only individual buildings but also the intermediary spatial organization that links settlement structure and architectural form. Compared with conventional manual mapping approaches, the integration of image segmentation and machine learning enables more consistent and scalable identification of building types. This data-driven framework not only enhances the analytical rigor of morphological studies but also provides a transferable approach for the analysis and interpretation of traditional settlements. 7. Conclusion This study takes Fulin Village in Quanzhou as a case study and develops a multi-scalar quantitative framework for spatial morphological analysis at the settlement, parcel, and building levels. By integrating machine learning and image segmentation techniques, the study systematically examines the spatial characteristics of traditional settlements and their underlying formation processes. (1) Methodological integration and technical innovation This study introduces image segmentation into the analysis of traditional settlement morphology and integrates it with a system of morphological indicators and spatial analytical methods. The proposed framework enables a multi-level quantitative representation from settlement patterns to parcel structures and building forms. Compared with conventional approaches relying on manual interpretation, this method allows for efficient identification and structured analysis of spatial elements based on large-scale datasets, providing an operational and transferable analytical workflow for traditional settlement studies. (2) Identification of multi-scalar spatial morphological characteristics The results indicate that Fulin Village exhibits a water-oriented spatial structure, characterized by central clustering and the coexistence of multiple differentiated zones. At the parcel level, spatial morphology is marked by irregularity and heterogeneity, with distinct typological combinations. At the building level, diverse morphological types are observed, including enclosed, detached, and regularized forms. These characteristics demonstrate strong interconnections across scales, reflecting both the hierarchical organization and overall coherence of spatial structure. (3) Theoretical interpretation of spatial formation processes At the interpretative level, the spatial morphology of Fulin Village can be understood through a multi-scalar formation framework of “environmental constraints – social drivers – hierarchical transmission.” Natural conditions, particularly water systems and topography, impose fundamental constraints on settlement location, parcel boundaries, and building layouts. Social structures, including clan organization, property relations, and economic activities, shape the logic of spatial organization. The parcel level functions as a critical intermediary, translating macro-scale spatial patterns into micro-scale morphological configurations. This framework provides a systematic explanation of how spatial morphology evolves from overall structure to localized forms. (4) Theoretical contributions and methodological extension This study advances beyond conventional single-scale descriptive approaches by integrating multi-scalar morphological analysis with interpretative perspectives, thereby enhancing the explanatory capacity of spatial morphology research. Moreover, the incorporation of digital technologies offers a new methodological pathway for the analysis of traditional settlements, expanding the application scope of digital approaches in architectural heritage studies. (5) Practical implications and future directions The findings provide a scientific basis for the conservation and regeneration of traditional settlements. By identifying spatial characteristics and formation processes across multiple scales, the study supports zoning-based conservation strategies, morphological control, and spatial optimization. This approach facilitates the preservation of spatial structure and cultural significance from a structural perspective, avoiding superficial or purely formal replication. Future research can further incorporate temporal datasets and comparative case studies to deepen the understanding of spatial evolution patterns across different regions. In addition, the integration of multi-source data and advanced computational methods offers promising potential for further development in the spatial analysis of cultural heritage. Declarations Conflicts of Interest: The authors declare no conflicts of interest. Author Contribution W.H. and L.W. designed the research and developed the methodology. W.H. performed data analysis and wrote the main manuscript text. Y.W. assisted with data processing and visualization, and prepared Figures 1–3. J.H. supervised the research and contributed to the conceptual framework and revision of the manuscript. All authors reviewed and approved the final manuscript. Acknowledgement The author gratefully acknowledges Stefano and Letizia for their insightful comments and constructive suggestions, which have greatly improved the quality of this manuscript. Data Availability The raw data supporting the conclusions of this article will be made available by the authors on request. References FANG, Q., & LI, Z. (2022). Cultural ecology cognition and heritage value of huizhou traditional villages [J]. Heliyon , 8 (12), e12627. LIU, F., & XU, W. (2025). Spatial Pattern of Traditional Villages in Remote Mountainous Areas and Their Development Potential Assessment: The Case of Enshi. China [J] Sustainability , 17 (3), 1138. WU, C., CHEN, M., ZHOU, L., et al. (2020). Identifying the Spatiotemporal Patterns of Traditional Villages in China: A Multiscale Perspective [J]. Land , 9 (11), 449. ZHAO, Y., & BIAN, G. (2026). In situ characteristics and patterns of settlement spatial differentiation along the Grand Canal’s Tianjin Section [J]. npj Heritage Science , 14 (1), 128. JIA, W. (2024). Morphological typology pedigree of centripetal spatial schema in traditional Chinese settlements [J]. Frontiers of Architectural Research , 13 (6), 1198–1214. WANG, Z., ZHU, J., & WU, Z. (2024). Study on the spatial distribution characteristics of traditional villages and their response to the water network system in the lower yangtze river basin [J]. Scientific Reports , 14 (1), 22586. LIU, Y., HE, S., WU, F., et al. (2010). Urban villages under China's rapid urbanization: Unregulated assets and transitional neighbourhoods [J]. Habitat International , 34 (2), 135–144. PENG, P., ZHANG, Y., ZHAO, J., et al. (2025). A study on the spatial distribution and driving factors of traditional villages–a case study of the Beijing-Tianjin-Hebei region in China [J]. Scientific Reports , 15 (1), 28516. LIU, W., & XUE, Y. (2023). Spatial distribution analysis and driving factors of traditional villages in Henan province: a comprehensive approach via geospatial techniques and statistical models [J]. Heritage Science , 11 (1), 185. SHI, Y., WANG, W., ZHANG, J., et al. (2025). UAV remote sensing and deep learning for assessing and optimizing architectural texture in traditional villages [J]. npj Heritage Science , 13 (1), 325. ZHANG, P., & GHOSH, D. (2023). Spatial measures and methods in sustainable urban morphology: A systematic review [J]. Landscape and Urban Planning , 237 , 104776. LIU, C., CAO, Y., YANG, C., et al. (2020). Pattern identification and analysis for the traditional village using low altitude UAV-borne remote sensing: multifeatured geospatial data to support rural landscape investigation, documentation and management [J]. Journal of Cultural Heritage , 44 , 185–195. LI, Z., LI, D., LAI, W., et al. (2025). A deep learning-based method for identifying traditional villages across cultural types [J]. npj Heritage Science , 13 (1), 671. ZHANG, J., ZHU, Z., DENG, J. (2025). Multi-Modal Feature Fusion for Spatial Morphology Analysis of Traditional Villages via Hierarchical Graph Neural Networks [J]. ArXiv, abs/2510.27208. CAMPIANI, A., LINGLE, A., & LERCARI, N. (2019). Spatial analysis and heritage conservation: Leveraging 3-D data and GIS for monitoring earthen architecture [J]. Journal of Cultural Heritage , 39 , 166–176. TANG, Z., & LI, M. (2020). Mapping Tea Plantations from VHR Images Using OBIA and Convolutional Neural Networks [J]. Remote Sensing , 12 (18), 2935. DADAO L. A Preliminary Study on Regional Analysis in Human Geography [J]. Acta Geographica Sinica , (1984). 51(4). YANBO, Q., & SHUWEN, W. (2019). Spatial morphology of rural settlements based on site and functional characteristics [J]. Resources Science , 41 (6), 1035–1047. PEILIN L. (2014). Traditional settlement cultural landscape gene [M] . Commercial. YIGANG P. (1992). Landscape analysis of traditional villages settlement [M] . China Architecture & Building. QING C. (2019). A perspective of the traditional settlements from ancient to now commemorating the ninetieth anniversary of the society for research in Chinese architecture [J]. Architectural Journal , 2019 (12), 14–19. JIE, Z., & SONGNAN, W. (2010). Quantitative analysis of the morphologies of traditional Chinese villages [J]. World Architecture , (2010 (1), 118–121. GUOGIN, J. I. N. D. (2015). Z. Policy-induced contemporary village form:a field survey from Wuxi city [J] (Vol. 2015, pp. 65–71). Urban Planning Forum. 2. WEI, T., & HONGYE, C. (2013). Spatial form and spatial cognition of traditional village insyntactical view: A case study of Xiaozhou Village,Guangzhou [J]. Acta Geographica Sinica , 68 (02), 209–218. BINGYAO L. Quantitative Methods and Evaluation of Urban Spatial Form [J]. Urban Planning Forum , (1998). (03): 42–45. KAI G. Urban morphology: an introduction and evaluation of the theories and the methods [J]. City Planning Review , (2001). (12): 36–42. YONG J. Spatial Political Economy Paradigm of Urban Design —— Review of Designing Cities by A. R. Cuthbert [J]. Urban Planning International, (2005). (06): 73–77. HONG, J. (2015). Transaction costs, property rights allocation and the evolution of urban spatial form: an analysis based on the perspective of new institutional economics [J]. Urban Planning Forum , (06): 63–69. LINGLI B. Spatial transaction costs: the evolution of the spatial form of industrial organization [J]. Academic Research , (2010). (05): 80–83. LI, H., & SONG, W. (2020). Pattern of spatial evolution of rural settlements in the Jizhou District of China during 1962–2030 [J]. Applied Geography , 122 , 102247. LUO, X. (2021). Knowledge Map on the Study of Chinese Rural Settlement Form Based on CSSCI [J]. Economic Geography , 41 (04), 148–157. OLIVEIRA, V. (2025). PORTA S. Quantitative and qualitative analysis in urban morphology: systematic legacy and latest developments [J]. Proceedings of the Institution of Civil Engineers - Urban Design and Planning, 178(2): 75–87. XIANKUI L. The Cultural Character of Ancient Bashu Architecture [J]. Architectural Journal , (1995). (03): 48–53. YILI W (2002). Diaojiaolou - a treasure of Tujia folk houses [J]. China Engineering Consulting , (10): 48–49. JUN W (2006). Form study on Xi'an traditional dwellings in the ancient city zone of Xi'an [D]. ZURUN Y (2001). Analysis on the spatial structure of traditional settlement environment [J]. Architectural Journal , (12): 21–24. JUN W. (2009). Spatial Concepts in Traditional Settlement Structure [M] . China Architecture & Building. XINCHENG P. (2013). Research on the quantitative method of the overall morphology of traditional rural settlements in two-dimensional plane [M] . Southeast University. YING, S., & DAWEI, X. (2015). A Space Syntax Analysis of Hakka Historical Villages in Meizhou [J]. Urban Development Studies , 22 (05), 63–70. DANDONG, G., LEI, T., NING, W., et al. (2017). Parametric analysis and reconstruction method of village's road morphology [J]. Journal of Zhejiang University(Engineering Science) , 51 (02), 279–286. F W. Simulating urban encroachment on rural land with fuzzy-logiccontrolled cellular automata in a geographical information system [J]. Journal of environmental management , (1998). 53(4): 293–308. R W, G E. Cellular automata and fractal urban form:a cellular modelling approach to the evolution of urban land-use patterns [J]. Environment&planning , (1993). 25(8): 1175–1199. TIAN, G., QIAO, Z., & GAO, X. (2014). Rural settlement land dynamic modes and policy implications in Beijing metropolitan region. China [J] Habitat International , 44 , 237–246. YING, L., XIAOBIN, J., MIAOYI, L., et al. (2014). A constrained cellular automata model for reconstructing historical arable land in Jiangsu province [J]. Geographical Research , 33 (12), 2239–2250. YAWEI, K., ZIYU, Z., ZENGDA, G. (2023). Study on the Causes of Spatial Morphology of RuralSettlements in the Yellow River Delta under the Influenceof Basin Environment [J]. Architectural Journal , (S2): 70–75. ESCH, T., MARCONCINI, M., MARMANIS, D., et al. (2014). Dimensioning urbanization – An advanced procedure for characterizing human settlement properties and patterns using spatial network analysis [J]. Applied Geography , 55 , 212–228. MOU, L., & LI, H. (2024). Identification and Spatial Characterization of suburban areas in Chengdu [J]. Applied Geography , 172 , 103428. ZUO, L., & TAO, X. (2021). Research on the spatial morphology of natural villages based on machine learning identification; proceedings of the 2020/2021 China Urban Planning Annual Conference and 2021 China Urban Planning Academic Season, Chengdu, China, F, [C]. ZHE, L., XINYI, L., JIAYING, S., et al. (2023). Quantitative Research of Landscape Texture for Polders inXuanwu Plain Based on Morphological Index:A Case Studyof Yongfeng Polder,Gucheng Lake [J]. Chinese Landscape Architecture , 39 (01), 41–46. PENG, F., JING, X., ZHIQI, Z., et al. (2022). The Method of Space-Dynamic Coupling Mechanism ofRural Settlements Based on Machine Learning: TakingLiyang City, Jiangsu Province as an Example [J]. Journal of Human Settlements in West China , 37 (04), 1–9. WU, S., DI, B., USTIN S, L, et al. (2022). Classification and detection of dominant factors in geospatial patterns of traditional settlements in China [J]. Journal of Geographical Sciences , 32 (5), 873–891. CHEN, W., YANG, L., WU, J., et al. (2023). Spatio-temporal characteristics and influencing factors of traditional villages in the Yangtze River Basin: a Geodetector model [J]. Heritage Science , 11 , 111. TENGYUN, H., PENGFEI, X., YANAN, W., et al. (2023). Research on Building Footprints Extraction Methods based on Different Deep Learning Models [J]. Remote Sensing Technology and Application , 38 (04), 892–902. WANG, M., XU, W., CAO, G., et al. (2024). Identification of rural courtyards’ utilization status using deep learning and machine learning methods on unmanned aerial vehicle images in north China [J]. Building Simulation , 17 , 799–818. PANPAN, Z., SHUAIPENG, L., LIQIANG, Z., et al. (2021). Multitask Learning-based Building Extraction from High-Resolution Remote Sensing Images [J]. Journal of Geo-information Science , 23 (03), 514–523. WANG, Q., YUAN, Q., LI, Z., et al. (2025). Study on the spatial relationships of traditional regional dwellings in Huizhou district [J]. npj heritage science , 238 , 534. WANG, Y., & LI, X. (2024). The AI’s Cognition of the Cultural Characteristics Underlying Chinese Courtyard Dwelling A visual perception-based approach for Siheyuan plan classification using machine learning and convolutional neural networks [Z]. Data-Driven Intelligence Proceedings of the 42nd Conference on Education and Research in Computer Aided Architectural Design in Europe. Nicosia. : 643–652. CHEN, Y., WANG, T., CHEN, G. (2025). A balanced multiscale object detection network for building façade elements detection [J]. Advanced Engineering Informatics , : 103289. RAN, X. (2019). HAISHAN X. A Saliency-model Based Quantitative Study and Evaluation on the Perception of Visual Attributes of Building Facades [J]. New Architecture , (06): 26–31. SUN, M., ZHANG, F., DUARTE, F., et al. (2022). Understanding architecture age and style through deep learning [J]. Cities , 128 , 103787. JI-WEI Z, DE-ER L, PENG, Y., et al. (2020). Window extraction method in dual depth image of point cloud of building facade [J] (Vol. 50, pp. 875–881). Laser & Infrared. 07. ZHANG, L., CHEN, Y., ZHENG, L., et al. (2024). Investigating the Surface Damage to Fuzhou’s Ancient Houses (Gu-Cuo) Using a Non-Destructive Testing Method Constructed via Machine Learning [J]. Coatings , 14 (11), 1466. YANG, S., CHEN, Y., ZHENG, L., et al. (2025). Investigating and Identifying the Surface Damage of Traditional Ancient Town Residence Roofs in Western Zhejiang Based on YOLOv8 Technology [J]. Coatings , 15 (2), 205. TING, Y., & SHAN, Z. (2016). A Study on the Application of Digital Technologies of Architectural Surveying in the Renovation of Historic Districts Practice Based in the Pearl River Delta [J]. Architectural Journal , (12): 22–27. MENG, C., SONG, Y., JI, J. (2022). Automatic classification of rural building characteristics using deep learning methods on oblique photography [J]. Building Simulation , (40): 1078. XIAOWEI, H., JIE, Z., WEN, L., et al. (2024). The Technical Exploration and Practice of Deep LearningRecognition in Traditional Textures of HistoricalSettlements [J] (Vol. 42, pp. 87–94). Development of Small Cities & Towns. 07. MIU, T. (2026). Characteristics and periodization of self-built houses in the Quanzhou coastal area based on archaeological typology method (late 19th century to early 21st century) [J]. Frontiers of Architectural Research , 15 (2), 522–539. ZHANG, X., & LI, S. (2024). Classification of architectural styles in Chinese traditional settlements using remote sensing images and building facade pictures [J]. Journal of Geographical Sciences , 34 (12), 2457–2476. JIANG, Y., & LI, N. (2023). Parametric Reconstruction of Traditional Village Morphology Based on the Space Gene Perspective—The Case Study of Xiaoxi Village in Western Hunan. China [J] Sustainability , 15 (3), 2088. ZHANG, C., PING, X., FAN, Q., et al. (2024). Measurement of 2D and 3D fractal features of urban morphology from an architectural view and its influencing factors [J]. Fractal and Fractional , 8 (3), 138. YADAV, M. J. (2025). LAVATE M. Morphometric Analysis and Spatial Pattern of Settlements in the Sina River Basin: A Case Study of Solapur District, Maharashtra [J]. IIP: International Multidisciplinary Research Journal , 2(Issue-IV (October-December)): 11–11. KWINTA, A., & GNIADEK, J. (2017). The description of parcel geometry and its application in terms of land consolidation planning [J]. Computers and Electronics in Agriculture , 136 , 117–124. LUCCHI E, D'ALONZO V, EXNER, D. (2019 [C]). A density-based spatial cluster analysis supporting the building stock analysis in historical towns; proceedings of the Building Simulation 2019, F, IBPSA. DUAN, Y., CHEN, S., ZHANG, L., et al. (2023). Spatial distribution characteristic and type classification of rural settlements: A case study of Weibei plain. China [J] Sustainability , 15 (11), 8736. ZHANG, J., FAN, B., LI, H., et al. (2023). Research on the shape classification method of rural homesteads based on parcel scale—Taking Yangdun village as an example [J]. Remote Sensing , 15 (19), 4763. VIALARD, A., & CARPENTER, A. (2015). Building density of parcels and block-faces from a syntactical, morphological and planning perspective; proceedings of the Proceedings of the 10th international space syntax symposium, F, [C]. LI, J., & LI, C. (2024). Characterizing urban spatial structure through built form typologies: A new framework using clustering ensembles [J]. Land Use Policy , 141 , 107166. DE SMET F, T. E. L. L. E. R. J. (2016). Characterising the morphology of suburban settlements: A method based on a semi-automatic classification of building clusters [J]. Landscape Research , 41 (1), 113–130. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9296889","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":624221642,"identity":"2819e846-f62e-4fbc-8d0d-2e92230859c6","order_by":0,"name":"Wencan Huang","email":"","orcid":"","institution":"Xiamen University","correspondingAuthor":false,"prefix":"","firstName":"Wencan","middleName":"","lastName":"Huang","suffix":""},{"id":624221643,"identity":"737e5ff6-55a7-4cc4-9bbd-ef407a58aa90","order_by":1,"name":"Liangliang Wang","email":"","orcid":"","institution":"Xiamen University","correspondingAuthor":false,"prefix":"","firstName":"Liangliang","middleName":"","lastName":"Wang","suffix":""},{"id":624221644,"identity":"23e31812-1374-4ca3-83e9-b547770d844d","order_by":2,"name":"Yixin Wang","email":"","orcid":"","institution":"Xiamen University","correspondingAuthor":false,"prefix":"","firstName":"Yixin","middleName":"","lastName":"Wang","suffix":""},{"id":624221645,"identity":"3dd9af64-fbd7-4d8b-ad78-2728d48b7a96","order_by":3,"name":"Jie Han","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAy0lEQVRIiWNgGAWjYBAC++P9H4z//mOr52dvIFbPmQMGBTxsfAmSPQeI1XIjweADD5tcgsGNBCJ1MPYcSNwgwWOWx3Dz8cYbDDU20QS1MLM3HDYwkEgrZpydVmzBcCwtt4GQFjaeg20GCQbHGJulc8wkGBsOE9bCI5HM/uNAwn/GNskzRGqRkEhjMGw4wJbYA/QQcVoMeM4wGDM2sBlL8AD9kkCMXwzYe8Ba5OyPH95440ONDWEtKNolEkhRDtFCqo5RMApGwSgYGQAAz6U9vaU0s2gAAAAASUVORK5CYII=","orcid":"","institution":"Xiamen University","correspondingAuthor":true,"prefix":"","firstName":"Jie","middleName":"","lastName":"Han","suffix":""}],"badges":[],"createdAt":"2026-04-02 01:54:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9296889/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9296889/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107319748,"identity":"8f390e47-4436-4419-94c2-8828acb50a6f","added_by":"auto","created_at":"2026-04-20 10:20:34","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2409150,"visible":true,"origin":"","legend":"\u003cp\u003eCase location and research scope\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/de6cec6c137798cdd30a51d9.png"},{"id":107485116,"identity":"2bdec457-5dc6-4405-a8e1-29f62f91ffa8","added_by":"auto","created_at":"2026-04-22 02:33:36","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":713162,"visible":true,"origin":"","legend":"\u003cp\u003eTechnical lines of research\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/35d73d315cc835118404e7df.png"},{"id":107319749,"identity":"51646270-30bd-4256-b9fb-1a7dcbe86cf0","added_by":"auto","created_at":"2026-04-20 10:20:34","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":258943,"visible":true,"origin":"","legend":"\u003cp\u003eOBlA image segmentation technical route\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/6f6e07d14e754bb478d2ca50.png"},{"id":107485459,"identity":"d49b3437-de49-4d70-bef8-76026434b5ce","added_by":"auto","created_at":"2026-04-22 02:34:58","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":452913,"visible":true,"origin":"","legend":"\u003cp\u003eThe process of extracting, building facade and roof elements for deep learning modeling\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/604ffb98fc58248fff3285e7.png"},{"id":107319751,"identity":"a8ce06f4-b025-468a-bf9c-bb612e985c68","added_by":"auto","created_at":"2026-04-20 10:20:34","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1704231,"visible":true,"origin":"","legend":"\u003cp\u003eLand-use composition of Fulin Village at the settlement level\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/c4a3629d2d9ee112dc2d6e5a.png"},{"id":107484054,"identity":"5f7d7219-8e0c-46a6-a4c0-ba4698b1886e","added_by":"auto","created_at":"2026-04-22 02:30:35","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":82327,"visible":true,"origin":"","legend":"\u003cp\u003eBoundary scale of FulinVillage settlement\u003c/p\u003e","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/35146f2f2ef2c67eee5e2822.jpg"},{"id":107319753,"identity":"9cdd2527-b83b-40ca-948f-503055bcaa4a","added_by":"auto","created_at":"2026-04-20 10:20:34","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":57360,"visible":true,"origin":"","legend":"\u003cp\u003eNuclear density of FulinVillage settlement\u003c/p\u003e","description":"","filename":"7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/55c7bdf46bf7b7d14ce4a68b.jpg"},{"id":107319755,"identity":"d2b0a84f-f088-4bde-9575-d5c4a6cc0855","added_by":"auto","created_at":"2026-04-20 10:20:34","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":89991,"visible":true,"origin":"","legend":"\u003cp\u003eNeighborhood Distance Analysis of FulinVillage settlement\u003c/p\u003e","description":"","filename":"8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/59ba41f8fde8f0577d7e0045.jpg"},{"id":107486301,"identity":"335ecad7-bb91-4df3-9e22-6b166ffbc711","added_by":"auto","created_at":"2026-04-22 02:38:03","extension":"jpeg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":596178,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of automatically labeled traditional building types with the traditional building database\u003c/p\u003e","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/ec7bb8ed9ec4ccccee853149.jpeg"},{"id":107486114,"identity":"ab405d54-76be-4931-a885-74edfd50e9c5","added_by":"auto","created_at":"2026-04-22 02:37:27","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":621701,"visible":true,"origin":"","legend":"\u003cp\u003eMorphological characteristics of parcel in high-density zones of different building types\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/d1812a7da169dab9398d6418.png"},{"id":107485072,"identity":"3afff730-db51-4cce-97ee-aa774587f562","added_by":"auto","created_at":"2026-04-22 02:33:35","extension":"jpeg","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":164765,"visible":true,"origin":"","legend":"\u003cp\u003eMorphological statistics of different traditional building types in Fulin village\u003c/p\u003e","description":"","filename":"floatimage9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/fb2f1e31783cdc58e5ba8f45.jpeg"},{"id":107319759,"identity":"f801e86b-cbe3-4b7f-9d0f-0a905ee7771f","added_by":"auto","created_at":"2026-04-20 10:20:34","extension":"jpeg","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":756875,"visible":true,"origin":"","legend":"\u003cp\u003eTypological atlas of representative building façades in Fulin Village\u003c/p\u003e","description":"","filename":"floatimage10.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/d3759b87a9e38f8403d7c0a9.jpeg"},{"id":107488416,"identity":"2a8a998d-9090-4979-9eb4-927b8682ff13","added_by":"auto","created_at":"2026-04-22 02:44:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":9742490,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9296889/v1/0890361c-8365-4deb-ab47-59dec3205069.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Data-driven Multi-scalar Analysis of Spatial Morphology and Cross-scale Interactions in Traditional Settlements: A Case Study of Fulin Village, China","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eTraditional villages in China represent important carriers of regional cultural heritage\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/sup\u003e, reflecting long-term interactions between human activities, natural environments, and socio-cultural systems. In the Chinese context, traditional villages generally refer to settlements that have preserved historical spatial patterns, architectural forms, and cultural practices over a long period, often with relatively intact layouts and strong continuity of local traditions\u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]\u003c/sup\u003e. Their spatial morphology embodies not only settlement patterns and architectural organization, but also the underlying logic of land use, social structure, and environmental adaptation\u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHowever, under the rapid processes of urbanization and rural revitalization in China, traditional villages are increasingly facing pressures from modernization\u003csup\u003e[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]\u003c/sup\u003e. Modernization in this context refers to the transformation of rural areas driven by economic development, infrastructure expansion, and changes in construction practices, often characterized by the replacement of traditional building materials and forms with standardized, modern constructions\u003csup\u003e[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e. Such processes frequently lead to the fragmentation of traditional spatial structures, the loss of morphological coherence, and the gradual disappearance of historically evolved settlement patterns. As a result, the preservation of traditional village spatial forms has become a critical issue in both academic research and planning practice.\u003c/p\u003e \u003cp\u003eExisting studies on traditional village morphology have largely relied on qualitative interpretation and manual mapping, focusing on typological classification, spatial structure description, and historical evolution\u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/sup\u003e. While these approaches have provided valuable insights into the cultural and spatial characteristics of traditional settlements, they are often limited by subjectivity, low efficiency, and insufficient quantitative support. In recent years, quantitative methods based on geographic information systems (GIS)\u003csup\u003e[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/sup\u003e, spatial metrics\u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/sup\u003e, and remote sensing\u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e have been increasingly applied to analyze settlement patterns and spatial structures\u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e. However, systematic data-driven approaches that integrate high-resolution spatial data and automated extraction techniques remain relatively limited, particularly at the fine-grained level of village morphology.\u003c/p\u003e \u003cp\u003eWith the rapid development of unmanned aerial vehicle (UAV) photogrammetry\u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e and machine learning techniques\u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/sup\u003e, new opportunities have emerged for the accurate acquisition and analysis of spatial morphological information. UAV-based data can provide high-resolution, up-to-date spatial information\u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e, while machine learning and image segmentation methods enable efficient extraction of land use, building types, and spatial structures from complex visual data. These technologies have been widely applied in urban studies and land use classification; however, their application in the morphological analysis of traditional villages is still in an exploratory stage, especially in terms of integrating multi-scale spatial features and supporting heritage conservation practices.\u003c/p\u003e \u003cp\u003eDespite these advancements, several challenges remain. First, many current studies lack clearly defined analytical frameworks that connect data extraction methods with spatial morphology interpretation\u003csup\u003e[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e. Second, the reproducibility of machine learning-based approaches is often limited due to inadequate reporting of methodological details\u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/sup\u003e. Third, the link between quantitative spatial analysis and heritage conservation strategies has not been fully examined, leading to a gap between technical analysis and practical application\u003csup\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eTo address these issues, this study proposes a data-driven approach to spatial morphology analysis of traditional villages by integrating UAV-based data acquisition, object-based image analysis (OBIA) \u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e, and machine learning techniques. Taking Fulin Village in Quanzhou, China, as a case study, the research aims to systematically extract spatial morphological elements and to examine their multi-level organization and cross-scale interactions.\u003c/p\u003e \u003cp\u003eSpecifically, this study addresses the following research questions:\u003c/p\u003e \u003cp\u003e(1) How can UAV-based image segmentation and machine learning methods be effectively applied to extract spatial morphological elements of traditional villages?\u003c/p\u003e \u003cp\u003e(2) What are the quantitative spatial characteristics and structural patterns of traditional village morphology across multiple scales (settlement, parcel, and building)?\u003c/p\u003e \u003cp\u003e(3) How do cross-scale interactions among settlement, parcel, and building levels operate, and what underlying mechanisms govern their spatial coupling?\u003c/p\u003e \u003cp\u003eBy constructing a multi-dimensional spatial dataset and conducting quantitative analysis, this study seeks to improve the accuracy and efficiency of traditional village morphology recognition, and to provide a methodological framework for integrating digital technologies into heritage conservation and rural planning practices.\u003c/p\u003e"},{"header":"2. Literature Review","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Spatial morphology of traditional settlements\u003c/h2\u003e \u003cp\u003eThe concept of spatial morphology originates from human geography\u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e and has been progressively extended across multiple disciplines, including geography\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]\u003c/sup\u003e, architecture\u003csup\u003e[\u003cspan additionalcitationids=\"CR21\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/sup\u003e, urban and rural planning\u003csup\u003e[\u003cspan additionalcitationids=\"CR24 CR25\" citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]\u003c/sup\u003e, and spatial political economy\u003csup\u003e[\u003cspan additionalcitationids=\"CR28\" citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]\u003c/sup\u003e. In the context of traditional settlements, spatial morphology refers to the physical organization of space formed under the combined influence of natural conditions and socio-cultural factors, encompassing building layouts, road networks, land-use structures, and their interactions and evolutionary processes \u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]\u003c/sup\u003e. Existing studies indicate that research on traditional settlement morphology in China has developed into a relatively systematic body of knowledge through interdisciplinary integration, covering spatial distribution patterns, morphological composition, evolution processes, and cultural implications\u003csup\u003e[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn terms of methodological development, studies on spatial morphology have gradually shifted from qualitative description to quantitative analysis\u003csup\u003e[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]\u003c/sup\u003e. Early research primarily relied on field surveys and graphical interpretation, focusing on vernacular dwelling types, construction techniques, and regional cultural characteristics\u003csup\u003e[\u003cspan additionalcitationids=\"CR34 CR35\" citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]\u003c/sup\u003e. With the expansion of research scale, the focus has moved from individual buildings to the overall spatial structure of settlements [19]. In recent years, the development of quantitative methods has enabled researchers to employ various indicators to characterize spatial morphology, such as building size, orientation, and spacing\u003csup\u003e[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]\u003c/sup\u003e, two-dimensional morphological metrics to quantify overall form\u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/sup\u003e, and space syntax to construct topological models of spatial organization\u003csup\u003e[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]\u003c/sup\u003e. Parametric modeling approaches have also been introduced to simulate and reconstruct settlement morphology\u003csup\u003e[\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn addition, dynamic simulation methods, including system dynamics, cellular automata (CA), and agent-based models (ABM), have been applied to explore settlement evolution and land-use change processes\u003csup\u003e[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]\u003c/sup\u003e, often in combination with remote sensing data for spatial prediction and historical reconstruction\u003csup\u003e[\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eDespite these advances, existing studies remain largely focused on morphological feature extraction, with limited integration of spatial structure, function, and cultural interpretation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Multi-scalar and hierarchical spatial analysis\u003c/h2\u003e \u003cp\u003eTraditional settlements are complex spatial systems characterized by significant multi-scalar and hierarchical structures. At the macro level, settlement location and overall spatial patterns are strongly constrained by natural environmental conditions and regional contexts \u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]\u003c/sup\u003e. At the meso level, parcel subdivision and spatial organization reflect social structures, institutional arrangements, and property relations\u003csup\u003e[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e. At the micro level, building typologies and spatial configurations embody local cultural traditions and construction practices\u003csup\u003e[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eRecent studies have increasingly incorporated hierarchical perspectives into spatial analysis. For example, space syntax has been widely used to investigate spatial cognition and organizational logic based on road network structures\u003csup\u003e[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e, while quantitative urban morphology research has proposed multi-indicator systems to describe hierarchical spatial structures \u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]\u003c/sup\u003e. Furthermore, institutional and political-economic approaches have been applied to explain the evolution of spatial forms through mechanisms such as transaction costs and property rights allocation \u003csup\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHowever, current multi-scalar studies still exhibit notable limitations. On the one hand, different spatial scales are often analyzed in isolation, lacking an integrated analytical framework. On the other hand, the transmission mechanisms across scales remain unclear, making it difficult to explain how macro-level spatial patterns are translated into micro-level morphological characteristics. Therefore, there is a need to develop an integrated multi-scalar analytical approach to reveal the hierarchical organization of traditional settlement morphology.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Data-driven approaches and machine learning\u003c/h2\u003e \u003cp\u003eWith the advancement of remote sensing and geospatial technologies, data-driven approaches have become increasingly important in the study of traditional settlement morphology. In particular, machine learning and deep learning techniques have been widely applied to spatial element extraction and pattern recognition.\u003c/p\u003e \u003cp\u003eAt the macro level, studies commonly combine remote sensing imagery with geospatial data to identify settlement patterns using machine learning methods\u003csup\u003e[\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]\u003c/sup\u003e. For instance, machine learning has been employed to classify village morphology and spatial scales\u003csup\u003e[\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]\u003c/sup\u003e, clustering algorithms have been used to analyze landscape texture patterns\u003csup\u003e[\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]\u003c/sup\u003e, and deep learning techniques have been applied to recognize spatial textures in historical settlements\u003csup\u003e[\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]\u003c/sup\u003e. In addition, some studies focus on environmental drivers of spatial morphology, analyzing influencing factors and formation mechanisms\u003csup\u003e[\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]\u003c/sup\u003e, while tools such as the Geodetector model have been used to examine spatial distribution patterns and their determinants\u003csup\u003e[\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAt the micro level, machine learning techniques are widely used for building detection and morphological extraction. High-resolution remote sensing images and UAV data enable the application of semantic segmentation and multi-task learning models for building footprint extraction and spatial feature identification\u003csup\u003e[\u003cspan additionalcitationids=\"CR54\" citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e]\u003c/sup\u003e. Other studies have explored building typology classification and spatial relationships based on plan data\u003csup\u003e[\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e, \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e]\u003c/sup\u003e. At the fa\u0026ccedil;ade level, deep learning models have been applied to component recognition, architectural style classification, and damage detection\u003csup\u003e[\u003cspan additionalcitationids=\"CR59 CR60 CR61 CR62\" citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e]\u003c/sup\u003e, often integrating three-dimensional data for detailed analysis and condition assessment\u003csup\u003e[\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e]\u003c/sup\u003e. Additionally, oblique photogrammetry combined with deep learning has been used to achieve multi-scale integration and automatic classification of building characteristics \u003csup\u003e[\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e]\u003c/sup\u003e .\u003c/p\u003e \u003cp\u003eOverall, data-driven methods have significantly improved the efficiency and accuracy of spatial morphology analysis. However, most studies remain task-specific or scale-specific, lacking systematic integration with comprehensive spatial morphological frameworks.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Research gaps and objectives\u003c/h2\u003e \u003cp\u003eDespite substantial progress in spatial morphology analysis and data-driven approaches, several critical limitations remain.\u003c/p\u003e \u003cp\u003eFirst, existing studies are predominantly conducted at a single spatial scale, with insufficient attention to the hierarchical organization of settlement space. In particular, the interactions among settlement, parcel, and building levels are rarely examined in an integrated manner, limiting the understanding of cross-scale spatial processes\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eSecond, although machine learning and image-based techniques have been widely applied to spatial recognition and feature extraction, their role remains largely technical. The integration of data-driven methods with morphological analysis and mechanism interpretation is still limited, resulting in a gap between quantitative extraction and theoretical explanation \u003csup\u003e[\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThird, the intermediary role of parcel-level spatial units has been largely overlooked. As a critical link between macro-level settlement structure and micro-level building form, parcels play a key role in mediating spatial organization. However, their function in transmitting, transforming, and regulating spatial morphology across scales has not been sufficiently explored\u003csup\u003e[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAs a result, there is still a lack of a unified and operational analytical framework capable of capturing multi-scale spatial characteristics and explaining their underlying generative mechanisms.\u003c/p\u003e \u003cp\u003eThe main contributions of this study are threefold:\u003c/p\u003e \u003cp\u003e(1) proposing a multi-level spatial morphology framework that strengthens the structural linkage across different spatial scales;\u003c/p\u003e \u003cp\u003e(2) integrating machine learning techniques to achieve automated identification and quantitative analysis of spatial elements;\u003c/p\u003e \u003cp\u003e(3) revealing the underlying mechanisms of traditional settlement morphology from a multi-scalar perspective.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Study Area","content":"\u003cp\u003eFulin Village is located in Quanzhou City, Fujian Province, in southeastern China, a region characterized by a subtropical monsoon climate and a long history of settlement development. The village is situated within a hilly coastal landscape, where natural topography and water systems have significantly influenced the formation of its spatial structure.\u003c/p\u003e \u003cp\u003eAs of the latest survey, Fulin Village covers an area of approximately 99.13 hectares and includes approximately 2200 households. The built environment consists of both traditional and modern constructions, among which traditional buildings account for approximately 56% of the total building stock. These traditional structures are mainly composed of Minnan-style residences, such as \"Dacuo\" and \"Fanzai buildings\"\u003csup\u003e[\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e]\u003c/sup\u003e which are characterized by enclosed courtyards, axial layouts, and strong spatial organization.\u003c/p\u003e \u003cp\u003eThe spatial morphology of Fulin Village exhibits a relatively well-preserved settlement pattern, including clustered building layouts, hierarchical road networks, and distinct functional zoning. At the same time, the village has experienced varying degrees of transformation due to recent rural development, leading to the coexistence of traditional and modern spatial elements.\u003c/p\u003e \u003cp\u003eFulin Village was selected as the case study for this research for the following reasons. First, it retains a representative spatial morphology typical of traditional villages in southern Fujian, making it suitable for morphological analysis. Second, the village shows clear spatial heterogeneity due to ongoing modernization processes, providing an appropriate context to examine the interaction between traditional and contemporary spatial structures. Third, high-resolution UAV data are available for the study area, enabling detailed spatial analysis based on image segmentation and machine learning techniques.\u003c/p\u003e \u003cp\u003eTo facilitate spatial analysis, the study area boundary was not defined strictly according to administrative limits, but instead delineated based on the extent of traditional settlement morphology and its functional relevance. Specifically, the study focuses on the historic village area with a high concentration of traditional buildings, together with an approximately 200-meter southward extension into adjacent farmland. This extended area corresponds to historically documented agricultural zones identified through field investigation, which are closely associated with the production\u0026ndash;living spatial structure of the village. The total study area defined in this research is approximately 47.71 hectares (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"4. Methodology","content":"\u003cp\u003eThis study proposes a data-driven framework for the spatial morphology analysis of traditional villages by integrating UAV-based data acquisition\u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e, object-based image analysis (OBIA)\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e, and machine learning techniques. Compared with conventional qualitative approaches, which rely on manual interpretation and typological description, the proposed method enables automated extraction, quantitative analysis, and multi-scale segmentation and characterization of spatial morphological elements.\u003c/p\u003e \u003cp\u003eThe overall workflow of the study consists of three main steps:(1) UAV data acquisition and preprocessing; (2) Multi-scale spatial morphology segmentation and feature extraction; (3) Quantitative spatial analysis. The workflow is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e4.1 UAV Data Acquisition and Preprocessing\u003c/h2\u003e \u003cp\u003eHigh-resolution spatial data of Fulin Village were acquired using unmanned aerial vehicle (UAV) oblique photogrammetry, which enables detailed three-dimensional reconstruction of complex settlement environments\u003csup\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e. The UAV survey was conducted using a multi-rotor UAV platform equipped with a high-resolution digital camera\u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e. The flight was carried out at an average altitude of approximately 80m, with a ground sampling distance (GSD) of approximately 5 cm, ensuring sufficient spatial resolution for fine-scale morphological analysis.\u003c/p\u003e \u003cp\u003eTo ensure data integrity and image quality, the flight mission was designed with a 70% overlap rate both longitudinally and laterally. Multi-angle oblique images were collected to capture building facades and complex spatial structures. The total number of images acquired was approximately 2485, covering the entire built-up area of the village.\u003c/p\u003e \u003cp\u003eThe collected images were processed using photogrammetric software (Context Capture ) to generate high-resolution orthophotos and digital surface models (DSM). The processing workflow included image alignment, sparse point cloud generation, dense point cloud reconstruction, mesh generation, and texture mapping.\u003c/p\u003e \u003cp\u003eSubsequently, preprocessing was conducted to improve data quality and ensure consistency for further analysis. This included geometric correction, coordinate system unification (WGS84), and noise reduction. The resulting orthophoto and DSM datasets provided the fundamental spatial data for image segmentation and machine learning-based analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Multi-scale spatial morphology segmentation and feature extraction\u003c/h2\u003e \u003cp\u003eTo accurately extract spatial morphological elements at different hierarchical levels, a multi-scale segmentation and classification approach was adopted in this study. Specifically, the method includes:\u003c/p\u003e \u003cp\u003e(1) \u003cb\u003eSettlement-scale segmentation\u003c/b\u003e based on orthophoto and DSM data, focusing on the extraction of overall spatial patterns and land-use structure.\u003c/p\u003e \u003cp\u003e(2) \u003cb\u003eParcel-scale extraction and classification\u003c/b\u003e, which serves as an intermediate layer linking settlement structure and building form. Given the lack of explicit parcel boundaries, parcels were delineated based on the spatial aggregation of building types. A density surface was generated from building clusters, and equal-density contour lines were used to approximate parcel boundaries. This approach defines parcels as emergent spatial units and supports the analysis of their mediating role in cross-scale interactions.\u003c/p\u003e \u003cp\u003e(3) \u003cb\u003eBuilding-scale feature extraction\u003c/b\u003e based on three-dimensional models, focusing on the identification of architectural elements such as roofs and facades.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e4.2.1 Settlement-level Segmentation Based on OBIA\u003c/h2\u003e \u003cp\u003eAt the settlement scale, spatial morphology was extracted through object-based image analysis (OBIA) using high-resolution orthophotos and digital surface models (DSM)\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e. Multi-resolution segmentation was performed in eCognition 9.0 software to partition the study area into homogeneous spatial objects representing land use and settlement texture.\u003c/p\u003e \u003cp\u003eThe segmentation process considered both spectral and elevation information, integrating RGB bands from orthophotos and height information from the DSM. The segmentation parameters were set as follows: scale parameter\u0026thinsp;=\u0026thinsp;50, shape factor\u0026thinsp;=\u0026thinsp;0.3, and compactness\u0026thinsp;=\u0026thinsp;0.5. These parameters were determined through iterative testing to balance segmentation detail and object integrity.\u003c/p\u003e \u003cp\u003eBased on the segmented objects, a supervised classification approach was implemented. Training samples were manually labeled according to major land use and spatial morphology categories, including building areas, roads, vegetation, water, and open spaces. A Random Forest classifier was applied to classify the segmented objects using a combination of spectral, textural (e.g., RGB), and geometric features (e.g., area, shape index).\u003c/p\u003e \u003cp\u003eThis process enabled the extraction of the overall spatial structure and planar morphological patterns of the village, providing a basis for analyzing settlement texture and land use organization.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section3\"\u003e \u003ch2\u003e4.2.2 Building-level Feature Extraction Based on Deep Learning\u003c/h2\u003e \u003cp\u003eAt the fa\u0026ccedil;ade morphology dimension, this study develops a deep learning\u0026ndash;based workflow for building feature extraction and typological classification, using a three-dimensional reconstruction model generated from UAV photogrammetry as the primary data source\u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e. This approach enables detailed analysis of roof structures, fa\u0026ccedil;ade compositions, and architectural components (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFirst, during the data preprocessing stage, traditional building imagery was manually annotated using Label Studio, focusing on key morphological features such as roof types and courtyard (patio) configurations to construct the initial training dataset. On this basis, the Segment Anything Model (SAM) was employed to perform large-scale segmentation of building data in Fulin Village, extracting the outlines of roofs and courtyard spaces. By integrating morphological parameters\u0026mdash;including roof height, number of pitched roofs, aspect ratio, and number of courtyards\u0026mdash;an initial classification of major building typologies was achieved, including traditional courtyard-based \u0026ldquo;Dacuo\u0026rdquo;, multi-storey \u0026ldquo;Fanzai buildings\u0026rdquo;, and arcade-style buildings.\u003c/p\u003e \u003cp\u003eSecond, for fa\u0026ccedil;ade element extraction, individual building models were isolated from the 3D dataset based on spatial geometric corner detection. Orthographic projection was then applied to generate distortion-free fa\u0026ccedil;ade images\u003csup\u003e[\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e]\u003c/sup\u003e. Based on these images, key fa\u0026ccedil;ade elements\u0026mdash;such as doors, windows, and decorative components\u0026mdash;were manually annotated to construct a standardized fa\u0026ccedil;ade dataset\u003csup\u003e[\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e]\u003c/sup\u003e. After normalization and noise reduction, a convolutional neural network (CNN)-based semantic segmentation model was trained using supervised learning to improve the recognition accuracy of complex fa\u0026ccedil;ade elements.\u003c/p\u003e \u003cp\u003eDuring model training, a labeled dataset consisting of 64 building samples was used, with 70% allocated for training and 30% for validation. In addition, segmentation masks generated by SAM were incorporated to assist model optimization, enhancing robustness under complex geometric conditions. Model performance was evaluated using accuracy and Intersection over Union (IoU) metrics to ensure the reliability and consistency of feature extraction results.\u003c/p\u003e \u003cp\u003eBased on the segmentation outputs, extracted elements were further aggregated and interpreted in combination with field survey data. Building morphology was then characterized across multiple dimensions, including roof form, fa\u0026ccedil;ade composition, and spatial configuration. Representative samples were selected for standardized reconstruction, leading to the development of a systematic morphological atlas of building types. This atlas establishes a structured analytical workflow of \u0026ldquo;image segmentation\u0026ndash;feature extraction\u0026ndash;typological classification,\u0026rdquo; providing a consistent data foundation for subsequent quantitative analysis and cross-type comparison.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFinally, a dual-dimensional analytical framework integrating both plan and fa\u0026ccedil;ade perspectives was established at the building level. By combining horizontal spatial structure (plan-based features) and vertical interface characteristics (fa\u0026ccedil;ade elements), this framework enables a more comprehensive representation of building morphology and supports multi-scalar spatial analysis.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Quantitative Spatial Analysis\u003c/h2\u003e \u003cp\u003eTo systematically examine the spatial morphology of traditional villages and their multi-scalar structural relationships, this study establishes a three-level analytical framework encompassing the settlement, parcel, and building scales. Within this framework, spatial characteristics are quantitatively analyzed across three dimensions\u0026mdash;size, shape, and configuration\u0026mdash;through a set of unified indicators, enabling the standardized representation and typological interpretation of morphological features.\u003c/p\u003e \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e \u003ch2\u003e4.3.1 Definition of Morphological Indicators\u003c/h2\u003e \u003cp\u003eTo ensure methodological consistency and comparability across different spatial scales, a set of core morphological indicators is defined and consistently applied throughout the analysis.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003e(1) Aspect Ratio (AR)\u003c/strong\u003e \u003cp\u003eThe aspect ratio is used to characterize the directional tendency and elongation of spatial units, calculated based on the minimum bounding rectangle\u003csup\u003e[\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e]\u003c/sup\u003e\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv id=\"Equa\" class=\"Equation\"\u003e \u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:AR=\\frac{L}{W}$$\u003c/div\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003ewhere \u003cem\u003eL\u003c/em\u003e and \u003cem\u003eW\u003c/em\u003e represent the length of the longer and shorter sides of the bounding rectangle, respectively. This indicator is applicable at the settlement, parcel, and building levels, and is used to distinguish between elongated and compact spatial forms.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003e(2) Shape Index (SI)\u003c/strong\u003e \u003cp\u003eThe shape index measures the complexity and regularity of spatial unit boundaries\u003csup\u003e[\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e]\u003c/sup\u003e\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv id=\"Equb\" class=\"Equation\"\u003e \u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:SI=\\frac{P}{2\\sqrt{\\pi\\:A}}$$\u003c/div\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003ewhere \u003cem\u003eP\u003c/em\u003e denotes the perimeter and \u003cem\u003eA\u003c/em\u003e the area of the spatial unit. Values approaching 1 indicate more regular shapes, while higher values reflect increased boundary complexity and irregularity. This indicator is applicable across multiple spatial scales.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003e(3) Land Use Proportion (LUP)\u003c/strong\u003e \u003cp\u003eThe land use proportion describes the composition of different functional land uses within the study area\u003csup\u003e[\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e]\u003c/sup\u003e\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv id=\"Equc\" class=\"Equation\"\u003e \u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e\n$$\\:LU{P}_{i}=\\frac{{A}_{i}}{{A}_{total}}$$\u003c/div\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{A}_{i}\\)\u003c/span\u003e\u003c/span\u003e represents the area of land use type \u003cem\u003ei\u003c/em\u003e, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{A}_{total}\\)\u003c/span\u003e\u003c/span\u003e is the total area of the study region. This indicator is used to analyze the functional structure of the settlement.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003e(4) Spatial Aggregation (Kernel Density)\u003c/strong\u003e \u003cp\u003eSpatial aggregation is assessed using kernel density estimation (KDE) to measure the concentration of building distribution. By analyzing the spatial density of building centroids, this method identifies core areas and overall distribution patterns within the settlement\u003csup\u003e[\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e]\u003c/sup\u003e\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv id=\"Equd\" class=\"Equation\"\u003e \u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equd\" name=\"EquationSource\"\u003e\n$$\\:f\\left(x\\right)=\\frac{1}{nh}\\sum\\:_{i=1}^{n}K\\left(\\frac{x-{x}_{i}}{h}\\right)$$\u003c/div\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003ewhere \u003cem\u003ef(x)\u003c/em\u003e represents the estimated density at location \u003cem\u003ex\u003c/em\u003e, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{x}_{i}\\)\u003c/span\u003e\u003c/span\u003e denotes the location of each building centroid, \u003cem\u003en\u003c/em\u003e is the number of samples, \u003cem\u003eh\u003c/em\u003e is the bandwidth, and \u003cem\u003eK\u003c/em\u003e is the kernel function. The resulting density surface provides an intuitive representation of spatial clustering and dispersion patterns.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section3\"\u003e \u003ch2\u003e4.3.2 Multi-scalar Spatial Morphology Analysis\u003c/h2\u003e \u003cp\u003eBuilding upon the unified indicator system, spatial morphology is analyzed across three hierarchical levels\u0026mdash;settlement, parcel, and building. Considering the differences in functional attributes and morphological characteristics across scales, a consistent analytical framework based on size, shape, and configuration is adopted, with scale-specific indicators selected to capture the key features at each level.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003e(1) Settlement Level\u003c/strong\u003e \u003cp\u003eThe settlement level focuses on overall spatial patterns and land-use structure.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eSize dimension\u003c/b\u003e: Land use proportion (LUP) is used to quantify the composition of different functional spaces and to characterize the overall land-use structure of the settlement.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eShape dimension\u003c/b\u003e: Aspect ratio and shape index are applied to describe the overall form and boundary complexity of the settlement.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eConfiguration dimension\u003c/b\u003e: Kernel density analysis is used to identify spatial clustering patterns and core areas of building distribution, revealing the overall spatial structure.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThrough the integration of these indicators, the settlement-level analysis captures the general spatial organization and morphological characteristics of the village.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003e(2) Parcel Level\u003c/strong\u003e \u003cp\u003eThe parcel level serves as an intermediate scale linking the overall settlement structure with individual buildings, reflecting patterns of spatial subdivision and organization.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eSize dimension\u003c/b\u003e: Parcel area is analyzed to examine the distribution of spatial unit sizes.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eShape dimension\u003c/b\u003e: Aspect ratio and shape index are used to characterize the regularity and directional properties of parcel forms.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eConfiguration dimension\u003c/b\u003e: Kernel density of buildings within parcel is used to assess internal spatial organization and identify different patterns of land use and spatial arrangement.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThis level of analysis helps reveal the internal logic of spatial subdivision and organization within the traditional village.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003e(3) Building Level\u003c/strong\u003e \u003cp\u003eThe building level focuses on the morphological characteristics of individual units and their spatial expression.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eSize dimension\u003c/b\u003e: The building footprint area is used to analyze size distribution.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eShape dimension\u003c/b\u003e: Aspect ratio and shape index are applied to describe plan form characteristics.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eConfiguration dimension\u003c/b\u003e: Fa\u0026ccedil;ade element extraction and their compositional relationships are analyzed to identify patterns of fa\u0026ccedil;ade organization and overall spatial configuration.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThis level provides a quantitative basis for understanding individual building characteristics and their role in shaping the overall spatial morphology.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Methodological Comparison and Applicability\u003c/h2\u003e \u003cp\u003eCompared with traditional qualitative methods, the proposed approach improves efficiency and reduces subjectivity by enabling automated extraction and quantitative analysis of spatial elements. While qualitative methods are effective in interpreting cultural meanings, they are limited in handling large-scale spatial data.\u003c/p\u003e \u003cp\u003eAlthough the proposed method is tested in Fulin Village, similar approaches have been widely applied in remote sensing classification and urban morphology analysis, demonstrating their adaptability to different spatial contexts. However, the effectiveness of the model may vary depending on image quality, settlement complexity, and regional characteristics, which should be considered in future applications.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Results of quantitative analysis of spatial patterns of traditional settlements","content":"\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e5.1 Settlement-level Spatial Morphological Characteristics\u003c/h2\u003e \u003cp\u003eIn the study of traditional village morphology, the settlement level reflects the overall spatial structure and pattern\u003csup\u003e[\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e]\u003c/sup\u003e, serving as the macro-scale foundation for multi-scalar analysis. To reveal the overall spatial organization of Fulin Village, this section examines its morphological characteristics from three aspects: land-use composition, boundary form, and spatial configuration.\u003c/p\u003e \u003cdiv id=\"Sec19\" class=\"Section3\"\u003e \u003ch2\u003e5.1.1 Land-use Composition\u003c/h2\u003e \u003cp\u003eBased on the statistical results of land use proportion (LUP) (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e), the spatial composition of Fulin Village is dominated by built-up land and ecological land. Specifically, built-up land covers approximately 25.3 hectares, including buildings, roads, and public spaces. Cultivated land accounts for approximately 8.05 hectares, green space for 12.3 hectares, and water bodies for 2.06 hectares.\u003c/p\u003e \u003cp\u003eIn terms of spatial distribution, both built-up land and cultivated land are primarily arranged along the water system, which forms a clear spatial boundary within the settlement. The road network is largely aligned along one side of the water system, while internal roads exhibit a relatively dispersed pattern. Overall, the land-use structure presents an interwoven spatial pattern of production, living, and ecological spaces.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section3\"\u003e \u003ch2\u003e5.1.2 Overall Settlement Form\u003c/h2\u003e \u003cp\u003eAccording to the morphological indicator analysis (Fig.\u0026nbsp;6), the overall form of Fulin Village is characterized by irregularity and relative compactness. The shape index (SI) is 2.24, indicating a boundary that is more complex than that of regular geometric forms\u003csup\u003e[\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn terms of aspect ratio, the settlement exhibits an AR value of 1.22, suggesting a relatively balanced configuration. The overall morphology is therefore predominantly compact and cluster-like. At the same time, localized extensions can be observed along the direction of the water system, resulting in a composite pattern characterized by a dominant clustered form with partial linear extensions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section3\"\u003e \u003ch2\u003e5.1.3 Spatial Configuration and Aggregation Patterns\u003c/h2\u003e \u003cp\u003eKernel density analysis (Fig.\u0026nbsp;7) reveals a clear spatial clustering pattern in the distribution of buildings\u003csup\u003e[\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e]\u003c/sup\u003e. High-density areas are concentrated in the central part of the village, with a secondary cluster located in the southwestern area, while peripheral zones exhibit relatively lower density. Overall, building density shows a decreasing gradient from the center toward the outskirts, indicating a single-core dominant spatial structure.\u003c/p\u003e \u003cp\u003eAnalysis of nearest-neighbor distances (Fig.\u0026nbsp;8) further supports this pattern. Building spacing ranges from a minimum of 4.08 m to a maximum of 58.7 m, with an average distance of 14.93 m. The central area is characterized by relatively small inter-building distances, while spacing gradually increases toward the boundary, reflecting a transition from compact to more dispersed spatial arrangements.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section3\"\u003e \u003ch2\u003e5.1.4 Summary\u003c/h2\u003e \u003cp\u003eBased on the analysis of land-use composition, overall form, and spatial configuration, the settlement-level characteristics of Fulin Village can be summarized as follows:\u003c/p\u003e \u003cp\u003e(1) The land-use structure is dominated by built-up and ecological land, with spatial distribution closely associated with the water system;\u003c/p\u003e \u003cp\u003e(2) The overall morphology is predominantly compact, with a complex boundary and localized directional extensions;\u003c/p\u003e \u003cp\u003e(3) The spatial structure exhibits a single-core clustering pattern, with building density decreasing from the center toward the periphery.\u003c/p\u003e \u003cp\u003eThese results indicate that Fulin Village presents clear spatial differentiation and clustering characteristics at the settlement level. However, such macro-scale patterns are not formed independently, but are supported by spatial organization at the intermediate scale. Therefore, further analysis at the parcel level is required to examine variations in size, shape, and configuration across different spatial units, in order to better understand the formation logic of the overall settlement structure.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section2\"\u003e \u003ch2\u003e5.2 Parcel-level Spatial Morphological Characteristics\u003c/h2\u003e \u003cp\u003eAs an intermediate scale linking the overall settlement structure and individual buildings, parcel represent the fundamental units organizing village spatial structure\u003csup\u003e[\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e]\u003c/sup\u003e. The spatial differentiation and clustering observed at the settlement level are essentially the result of the spatial combination and evolution of heterogeneous parcel units\u003csup\u003e[\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e]\u003c/sup\u003e. Variations in parcel size and morphology, when aggregated spatially, further shape the functional zoning and structural hierarchy of the settlement.\u003c/p\u003e \u003cp\u003eBased on the distribution of building typologies and their spatial clustering patterns, combined with parcel-scale morphological indicators, this section identifies spatial zoning patterns and examines parcel-level characteristics in Fulin Village.\u003c/p\u003e \u003cdiv id=\"Sec24\" class=\"Section3\"\u003e \u003ch2\u003e5.2.1 Spatial Zoning and Parcel Size Characteristics\u003c/h2\u003e \u003cp\u003eBased on the automatically identified building typologies (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e9\u003c/span\u003e), spatial zoning at the parcel level was delineated through an integrated assessment of dominant building types, the spatial extent of high kernel density areas, and parcel size distribution patterns. The results indicate that the study area can be classified into three representative zones:\u003c/p\u003e \u003cp\u003e(1) a core residential zone dominated by Dacuo buildings;\u003c/p\u003e \u003cp\u003e(2) a street-oriented commercial zone characterized by arcade buildings;\u003c/p\u003e \u003cp\u003e(3) a peripheral residential zone dominated by Fanzai buildings.\u003c/p\u003e \u003cp\u003eThe outer expansion area, primarily composed of modern self-built houses, is excluded from this analysis.\u003c/p\u003e \u003cp\u003eIn terms of parcel size, significant variation is observed across different zones. The core Dacuo zone is characterized by parcel of moderate size, reflecting relatively stable courtyard-based spatial units. In contrast, parcel in the arcade building zone are generally smaller and more concentrated, indicating higher land-use intensity along street frontages. The Fanzai buildings zone exhibits larger parcel sizes, suggesting a more spacious spatial organization.\u003c/p\u003e \u003cp\u003eFrom a spatial perspective, these zones display distinct clustering patterns. The Dacuo zone is concentrated in the central area of the village, the arcade zone extends along primary road corridors, and the Fanzai buildings zone is mainly distributed toward the settlement periphery.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003e5.2.2 Parcel Morphological Characteristics\u003c/h2\u003e \u003cp\u003eParcel morphology shows clear differentiation across zones based on aspect ratio and shape index analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e10\u003c/span\u003e). Parcel in the core Dacuo zone are predominantly regular and rectangular, indicating a high degree of geometric order. In the arcade zone, parcel exhibit relatively low aspect ratios, resulting in compact and regular forms. In contrast, parcel in the Fanzai buildings zone display more elongated and irregular configurations, reflecting increased morphological variability.\u003c/p\u003e \u003cp\u003eOverall, parcel morphology demonstrates a transition from regular and standardized forms in the core area to more diverse and irregular forms toward the periphery.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003e5.2.3 Building Density and Spatial Configuration\u003c/h2\u003e \u003cp\u003eBuilding density reflects the internal spatial organization of parcel\u003csup\u003e[\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e]\u003c/sup\u003e. Kernel density analysis indicates clear differences in spatial configuration among the identified zones. The core Dacuo zone exhibits high density with strongly clustered building distributions. The arcade zone is characterized by high building coverage and continuous linear arrangements along streets, resulting in a compact spatial configuration. The Fanzai buildings zone presents an intermediate density level, with a mixed pattern of clustering and dispersion. In contrast, peripheral areas show relatively low density and more dispersed distributions.\u003c/p\u003e \u003cp\u003eOverall, building distribution within parcel exhibit a gradient transition from high-density clustering to more dispersed arrangements.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003e5.2.4 Summary\u003c/h2\u003e \u003cp\u003eBased on spatial zoning and multi-dimensional analysis at the parcel level, the following characteristics can be identified:\u003c/p\u003e \u003cp\u003e(1) Parcel space exhibits clear zoning patterns, with distinct clusters corresponding to different spatial functions and variations in parcel size;\u003c/p\u003e \u003cp\u003e(2) Parcel morphology transitions from regular rectangular forms in the core area to more diverse and irregular forms toward the periphery;\u003c/p\u003e \u003cp\u003e(3) Spatial configuration shows a gradient change from high-density clustering to relatively dispersed arrangements.\u003c/p\u003e \u003cp\u003eThese results indicate that significant differences exist among parcel zones in terms of size, morphology, and building density, forming a spatial organization pattern that transitions from courtyard-based layouts to street-oriented compact forms and further to more loosely structured configurations. This pattern provides the basis for further analysis at the building level, where variations in scale, proportion, and morphology can be examined in greater detail.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003e5.3 Building-level Spatial Morphological Characteristics\u003c/h2\u003e \u003cp\u003eAs the fundamental components of parcel organization, individual buildings play a decisive role in shaping internal spatial structure and land-use patterns. Variations in building type, size, and morphology, through their spatial arrangement within parcel, further influence parcel-level configuration and collectively contribute to the overall settlement form.\u003c/p\u003e \u003cp\u003eThe differentiated organizational patterns identified at the parcel level are essentially derived from the spatial distribution and combination of different building typologies\u003csup\u003e[\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e]\u003c/sup\u003e. Therefore, this section examines building-level spatial morphology in Fulin Village from both plan-based metrics and fa\u0026ccedil;ade characteristics.\u003c/p\u003e \u003cdiv id=\"Sec29\" class=\"Section3\"\u003e \u003ch2\u003e5.3.1 Size and Typological Characteristics\u003c/h2\u003e \u003cp\u003eBased on building typology classification, traditional buildings in Fulin Village mainly include five-bay courtyard houses (Dacuo), three-bay courtyard houses, Fanzai buildings, and arcade buildings. The dataset comprises 32 five-bay Dacuo, 10 three-bay Dacuo, approximately 61 Fanzai buildings, and 11 arcade buildings.\u003c/p\u003e \u003cp\u003eIn terms of size (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e11\u003c/span\u003e, left), clear differences can be observed among building types. Five-bay Dacuo generally range from 250\u0026ndash;450 m\u0026sup2;, representing the largest building type. Three-bay Dacuo are primarily distributed between 100\u0026ndash;180 m\u0026sup2;, indicating a relatively smaller scale. Fanzai buildings and multi-storey courtyard buildings are concentrated within 150\u0026ndash;230 m\u0026sup2;, while arcade buildings typically range from 150\u0026ndash;260 m\u0026sup2;, representing a moderate scale. Overall, building types exhibit a clear hierarchical differentiation in footprint area.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec30\" class=\"Section3\"\u003e \u003ch2\u003e5.3.2 Plan Morphological Characteristics\u003c/h2\u003e \u003cp\u003eIn terms of plan morphology (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e11\u003c/span\u003e, right), aspect ratio varies across building types. Five-bay Dacuo show an average aspect ratio of 1.21 (\u0026plusmn;\u0026thinsp;0.2), indicating near-rectangular forms. Three-bay Dacuo exhibit a higher value of 1.43 (\u0026plusmn;\u0026thinsp;0.4), reflecting a moderate directional tendency. Fanzai buildings and multi-storey courtyard buildings have an average aspect ratio of 1.31 (\u0026plusmn;\u0026thinsp;0.6), suggesting relatively balanced forms with greater variability. In contrast, arcade buildings present a significantly higher aspect ratio of 2.10 (\u0026plusmn;\u0026thinsp;2.0), indicating pronounced linear elongation.\u003c/p\u003e \u003cp\u003eOverall, building plans are predominantly rectangular, with a transition in aspect ratio from relatively balanced to more elongated forms across different building types.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec31\" class=\"Section3\"\u003e \u003ch2\u003e5.3.3 Fa\u0026ccedil;ade Typology and Morphological Variation\u003c/h2\u003e \u003cp\u003eBuilding on the analysis of plan size and proportion, fa\u0026ccedil;ade characteristics were further examined to construct a typological atlas of building morphology\u003csup\u003e[\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e, \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e78\u003c/span\u003e]\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e12\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDacuo buildings are generally characterized by single-storey, axially symmetrical compositions with a tripartite vertical structure consisting of base, wall body, and roof. Roof forms commonly include curved ridge types, while fa\u0026ccedil;ade elements such as lattice windows, wooden doors, and stone or carved decorative components are frequently observed. These buildings exhibit high symmetry, relatively low opening ratios, and enclosed spatial expressions.\u003c/p\u003e \u003cp\u003eFanzai buildings demonstrate noticeable variation in fa\u0026ccedil;ade composition. The transition from single-storey to multi-storey forms introduces additional elements such as balconies and colonnades. Western-influenced features\u0026mdash;including classical columns, pediments, arched openings, and metal railings\u0026mdash;are incorporated, resulting in increased fa\u0026ccedil;ade openness and greater diversity in decorative expression.\u003c/p\u003e \u003cp\u003eArcade buildings further emphasize continuity along the street interface. Their fa\u0026ccedil;ades are characterized by open ground floors and continuous colonnades, forming linear and interconnected street-front spaces.\u003c/p\u003e \u003cp\u003eBased on fa\u0026ccedil;ade elements, opening ratios, and compositional characteristics, building fa\u0026ccedil;ades in Fulin Village can be categorized into three typological groups: traditional, hybrid, and street-oriented. These categories reflect observable variation in fa\u0026ccedil;ade composition across building types.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec32\" class=\"Section3\"\u003e \u003ch2\u003e5.3.4 Summary\u003c/h2\u003e \u003cp\u003eBased on the analysis of building size, plan morphology, and fa\u0026ccedil;ade configuration, the following characteristics can be identified at the building level:\u003c/p\u003e \u003cp\u003e(1) Building types are diverse, with clear differentiation in size among different typologies;\u003c/p\u003e \u003cp\u003e(2) Plan forms are predominantly rectangular, with variations in aspect ratio across building types;\u003c/p\u003e \u003cp\u003e(3) Fa\u0026ccedil;ade configurations differ in terms of element composition and opening patterns.\u003c/p\u003e \u003cp\u003eThese results indicate that building-level differences in size, shape, and configuration are significant and form the basis for spatial variation across higher levels. The spatial distribution and morphological differentiation of building types contribute to the organization of parcel and reinforce the spatial structure observed at the settlement level.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec33\" class=\"Section2\"\u003e \u003ch2\u003e5.4 Summary of Results\u003c/h2\u003e \u003cp\u003eThe integrated analysis across the settlement, parcel, and building levels reveals a clear multi-scalar organizational logic in the spatial morphology of Fulin Village. At the settlement level, patterns of spatial clustering and functional differentiation are observed, which correspond to variations in parcel size and organizational forms. At the parcel level, morphological characteristics are closely associated with the composition and distribution of building types. At the building level, differences in size, proportion, and configuration further contribute to spatial variation across higher levels.\u003c/p\u003e \u003cp\u003eTaken together, these results indicate that spatial morphology in Fulin Village is characterized by hierarchical differentiation across scales, forming an interconnected system linking structure, organization, and form.\u003c/p\u003e \u003cp\u003eThe multi-level analytical framework developed in this study provides a structured approach for the identification, classification, and analysis of traditional village morphology, and establishes a consistent basis for further discussion on spatial patterns and their implications.\u003c/p\u003e \u003c/div\u003e"},{"header":"6. Discussion","content":"\u003cdiv id=\"Sec35\" class=\"Section2\"\u003e \u003ch2\u003e6.1 Natural controls on spatial sequencing\u003c/h2\u003e \u003cp\u003eThe distance gradient analysis and land-use statistics (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e) reveal a clear spatial differentiation between farmland and residential land in relation to the water system. Farmland is strongly concentrated in close proximity to water, whereas residential buildings are predominantly distributed within intermediate distance ranges. Together, these patterns form a sequential spatial structure extending outward from the water system.\u003c/p\u003e \u003cp\u003eIn addition, the ratio between built-up land and farmland (approximately 3.2:1) indicates a relatively balanced configuration between living and production spaces at the settlement scale. Rather than reflecting direct environmental determinism, this pattern suggests that natural factors operate indirectly by shaping locational choices across different land-use types.\u003c/p\u003e \u003cp\u003eSpecifically, farmland exhibits a strong dependence on water accessibility, as irrigation efficiency decreases with distance. In contrast, residential areas are positioned to balance accessibility to water resources with the need to avoid environmental risks, particularly flooding. This divergence implies that different spatial elements respond to the same environmental gradient through differentiated optimization strategies, resulting in a structured spatial sequence.\u003c/p\u003e \u003cp\u003eFurther evidence can be observed in the northern part of the village, where portions of farmland have transitioned into green space. This shift suggests that when water accessibility declines or irrigation efficiency becomes insufficient, land use may gradually shift from production-oriented to ecological functions. Such transformations highlight the long-term and dynamic influence of natural conditions on spatial organization.\u003c/p\u003e \u003cp\u003eOverall, the spatial configuration of Fulin Village can be interpreted as an outcome of a \u0026ldquo;distance\u0026ndash;risk trade-off\u0026rdquo; mechanism, in which water-related accessibility and environmental constraints jointly shape the allocation of land uses. While previous studies have described traditional settlements as \u0026ldquo;water-oriented\u0026rdquo;, this study advances the understanding by quantitatively demonstrating how such relationships manifest as measurable spatial gradients.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec36\" class=\"Section2\"\u003e \u003ch2\u003e6.2 Social structure and spatial organization\u003c/h2\u003e \u003cp\u003eThe spatial distribution of buildings in Fulin Village exhibits a clear core\u0026ndash;periphery pattern, characterized by high-density clustering in the central area and a gradual decrease toward the periphery. This central zone corresponds closely with the distribution of traditional \u003cem\u003eDacuo\u003c/em\u003e buildings, while \u003cem\u003eFanzai\u003c/em\u003e buildings and arcaded houses are primarily located in outer areas.\u003c/p\u003e \u003cp\u003eQuantitative results further show that the central area not only has the highest kernel density values but also features more regular and compact parcel configurations. In contrast, peripheral areas display greater variability in parcel size and more irregular morphological characteristics. These differences suggest that central spaces are subject to stronger spatial control, whereas peripheral zones reflect more flexible and adaptive development processes.\u003c/p\u003e \u003cp\u003eThis pattern can be attributed to the influence of social structure, particularly the role of clan-based organization in traditional settlements. By occupying central locations, dominant social groups establish a stable spatial framework that persists over time. Subsequent development tends to occur through infill within the existing structure or expansion along its edges, reinforcing the original spatial hierarchy.\u003c/p\u003e \u003cp\u003eFrom a temporal perspective, the observed spatial configuration reflects strong path dependency. The persistence of central dominance and peripheral expansion indicates that historical spatial arrangements continue to constrain and guide later development. As a result, spatial organization in Fulin Village can be understood as a \u0026ldquo;core control\u0026ndash;peripheral growth\u0026rdquo; model.\u003c/p\u003e \u003cp\u003eThis finding aligns with existing research emphasizing the role of social organization in shaping settlement structure. However, by integrating kernel density analysis with parcel-level morphological metrics, this study provides a more explicit and quantifiable account of how social structure is translated into spatial form, both in terms of distribution patterns and morphological characteristics.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec37\" class=\"Section2\"\u003e \u003ch2\u003e6.3 Building typology-driven spatial differentiation and density restructuring\u003c/h2\u003e \u003cp\u003eThe spatial distribution of building typologies (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e10\u003c/span\u003e) reveals clear clustering patterns, forming distinct functional zones within the village. Traditional Dacuo are concentrated in the central area, arcade buildings are located along key transportation nodes near the village entrance and water system, while Fanzai buildings occupy transitional zones between these two areas.\u003c/p\u003e \u003cp\u003eParcel-level analysis further demonstrates systematic differences among these zones in terms of size, morphology, and spatial configuration. Arcade zones are characterized by the smallest parcel sizes and the highest building coverage, indicating intensive land use. In contrast, Fanzai zones exhibit relatively larger parcel and more flexible spatial arrangements, while Dacuo zones are associated with moderate parcel sizes and more regular morphological forms.\u003c/p\u003e \u003cp\u003eThese findings suggest that building typology is not only a morphological outcome but also an active factor shaping parcel organization. This process can be interpreted as a hierarchical transformation from typology to spatial structure: building types, through spatial clustering, generate distinct zone configurations, which in turn influence the overall settlement pattern. For instance, the continuous high-density arrangement of arcade buildings reinforces linear street interfaces, whereas the relatively dispersed distribution of Fanzai buildings results in more open spatial configurations.\u003c/p\u003e \u003cp\u003eWhile previous studies have emphasized function-driven spatial differentiation, this study extends the discussion by demonstrating how such differentiation can be identified through data-driven approaches, combining automated typology classification with density-based spatial analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec38\" class=\"Section2\"\u003e \u003ch2\u003e6.4 Morphological transformation from ritual order to functional adaptation\u003c/h2\u003e \u003cp\u003eThe analysis of building morphology reveals significant variation across typologies in terms of plan proportions, spatial organization, and fa\u0026ccedil;ade composition. Traditional Dacuo are characterized by enclosed courtyard layouts, symmetrical organization, and relatively stable aspect ratios. In contrast, Fanzai buildings and arcade structures exhibit greater variability in both plan proportion and fa\u0026ccedil;ade openness.\u003c/p\u003e \u003cp\u003eQuantitatively, Dacuo buildings show limited variation in aspect ratio, indicating strong formal constraints, whereas arcade buildings display elongated forms and continuous street-facing interfaces, reflecting increased adaptation to external spatial conditions. These differences suggest that morphological variation is not random but closely associated with shifts in spatial organization.\u003c/p\u003e \u003cp\u003eThis process can be understood as a transition from ritual-based spatial order to function-oriented adaptation. Under clan-based social structures, building forms emphasize internal hierarchy and enclosure. With the intensification of commercial activities and changing social conditions, building morphology increasingly responds to functional requirements such as ventilation, lighting, and accessibility.\u003c/p\u003e \u003cp\u003eThis interpretation is consistent with previous findings describing a shift from inward-oriented to outward-facing spatial forms. However, by employing quantitative morphological indicators and typological classification, this study provides a measurable basis for understanding this transformation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec39\" class=\"Section2\"\u003e \u003ch2\u003e6.5 Reconfiguration of local building traditions and external influences\u003c/h2\u003e \u003cp\u003eFa\u0026ccedil;ade typology analysis indicates a high degree of consistency in overall proportions and compositional frameworks across different building types, while variation is primarily expressed through localized elements such as openings, balconies, and external corridors. This pattern of \u0026ldquo;overall stability with localized variation\u0026rdquo; is consistently observed across the dataset.\u003c/p\u003e \u003cp\u003eFurther examination of fa\u0026ccedil;ade elements shows that external influences are mainly incorporated at the interface level, without fundamentally altering the underlying structural logic. For example, Fanzai buildings and arcade buildings increase fa\u0026ccedil;ade openness and introduce additional spatial elements to enhance usability, while maintaining basic compositional relationships inherited from traditional architecture.\u003c/p\u003e \u003cp\u003eThis suggests that the integration of external elements is not a process of simple addition, but rather a selective adaptation constrained by existing construction systems. The process can be interpreted as a form of functional filtering and structural adaptation, whereby new elements are incorporated only when they improve spatial performance without disrupting the established framework.\u003c/p\u003e \u003cp\u003eThis finding aligns with previous studies highlighting the adaptive capacity of local building traditions. By quantifying fa\u0026ccedil;ade elements and compositional characteristics, this study provides a more structured account of how such adaptation occurs at the morphological level.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec40\" class=\"Section2\"\u003e \u003ch2\u003e6.6 Summary\u003c/h2\u003e \u003cp\u003eThe above analysis demonstrates that the spatial morphology of Fulin Village is not determined by a single factor, but emerges from the interaction of multiple processes across different scales. Natural factors shape spatial sequences through distance-related constraints, social structures influence spatial organization through hierarchical occupation patterns, and building typologies contribute to morphological differentiation through their spatial distribution and configuration.\u003c/p\u003e \u003cp\u003eOverall, this process can be conceptualized as a multi-scalar interaction framework linking environmental constraints, social organization, and typological transformation. The relationships across scales are established through quantifiable indicators, resulting in a structured rather than random spatial evolution. The identified cross-scale mechanism provides insights for heritage conservation by highlighting the importance of maintaining not only individual buildings but also the intermediary spatial organization that links settlement structure and architectural form.\u003c/p\u003e \u003cp\u003eCompared with conventional manual mapping approaches, the integration of image segmentation and machine learning enables more consistent and scalable identification of building types. This data-driven framework not only enhances the analytical rigor of morphological studies but also provides a transferable approach for the analysis and interpretation of traditional settlements.\u003c/p\u003e \u003c/div\u003e"},{"header":"7. Conclusion","content":"\u003cp\u003eThis study takes Fulin Village in Quanzhou as a case study and develops a multi-scalar quantitative framework for spatial morphological analysis at the settlement, parcel, and building levels. By integrating machine learning and image segmentation techniques, the study systematically examines the spatial characteristics of traditional settlements and their underlying formation processes.\u003c/p\u003e \u003cp\u003e \u003cb\u003e(1) Methodological integration and technical innovation\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThis study introduces image segmentation into the analysis of traditional settlement morphology and integrates it with a system of morphological indicators and spatial analytical methods. The proposed framework enables a multi-level quantitative representation from settlement patterns to parcel structures and building forms. Compared with conventional approaches relying on manual interpretation, this method allows for efficient identification and structured analysis of spatial elements based on large-scale datasets, providing an operational and transferable analytical workflow for traditional settlement studies.\u003c/p\u003e \u003cp\u003e \u003cb\u003e(2) Identification of multi-scalar spatial morphological characteristics\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThe results indicate that Fulin Village exhibits a water-oriented spatial structure, characterized by central clustering and the coexistence of multiple differentiated zones. At the parcel level, spatial morphology is marked by irregularity and heterogeneity, with distinct typological combinations. At the building level, diverse morphological types are observed, including enclosed, detached, and regularized forms. These characteristics demonstrate strong interconnections across scales, reflecting both the hierarchical organization and overall coherence of spatial structure.\u003c/p\u003e \u003cp\u003e \u003cb\u003e(3) Theoretical interpretation of spatial formation processes\u003c/b\u003e \u003c/p\u003e \u003cp\u003eAt the interpretative level, the spatial morphology of Fulin Village can be understood through a multi-scalar formation framework of \u0026ldquo;environmental constraints \u0026ndash; social drivers \u0026ndash; hierarchical transmission.\u0026rdquo; Natural conditions, particularly water systems and topography, impose fundamental constraints on settlement location, parcel boundaries, and building layouts. Social structures, including clan organization, property relations, and economic activities, shape the logic of spatial organization. The parcel level functions as a critical intermediary, translating macro-scale spatial patterns into micro-scale morphological configurations. This framework provides a systematic explanation of how spatial morphology evolves from overall structure to localized forms.\u003c/p\u003e \u003cp\u003e \u003cb\u003e(4) Theoretical contributions and methodological extension\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThis study advances beyond conventional single-scale descriptive approaches by integrating multi-scalar morphological analysis with interpretative perspectives, thereby enhancing the explanatory capacity of spatial morphology research. Moreover, the incorporation of digital technologies offers a new methodological pathway for the analysis of traditional settlements, expanding the application scope of digital approaches in architectural heritage studies.\u003c/p\u003e \u003cp\u003e \u003cb\u003e(5) Practical implications and future directions\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThe findings provide a scientific basis for the conservation and regeneration of traditional settlements. By identifying spatial characteristics and formation processes across multiple scales, the study supports zoning-based conservation strategies, morphological control, and spatial optimization. This approach facilitates the preservation of spatial structure and cultural significance from a structural perspective, avoiding superficial or purely formal replication.\u003c/p\u003e \u003cp\u003eFuture research can further incorporate temporal datasets and comparative case studies to deepen the understanding of spatial evolution patterns across different regions. In addition, the integration of multi-source data and advanced computational methods offers promising potential for further development in the spatial analysis of cultural heritage.\u003c/p\u003e \u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eConflicts of Interest:\u003c/h2\u003e \u003cp\u003eThe authors declare no conflicts of interest.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eW.H. and L.W. designed the research and developed the methodology. W.H. performed data analysis and wrote the main manuscript text. Y.W. assisted with data processing and visualization, and prepared Figures 1\u0026ndash;3. J.H. supervised the research and contributed to the conceptual framework and revision of the manuscript. All authors reviewed and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThe author gratefully acknowledges Stefano and Letizia for their insightful comments and constructive suggestions, which have greatly improved the quality of this manuscript.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe raw data supporting the conclusions of this article will be made available by the authors on request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eFANG, Q., \u0026amp; LI, Z. (2022). Cultural ecology cognition and heritage value of huizhou traditional villages [J]. \u003cem\u003eHeliyon\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e(12), e12627.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLIU, F., \u0026amp; XU, W. (2025). Spatial Pattern of Traditional Villages in Remote Mountainous Areas and Their Development Potential Assessment: The Case of Enshi. \u003cem\u003eChina [J] Sustainability\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e(3), 1138.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWU, C., CHEN, M., ZHOU, L., et al. (2020). Identifying the Spatiotemporal Patterns of Traditional Villages in China: A Multiscale Perspective [J]. \u003cem\u003eLand\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(11), 449.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZHAO, Y., \u0026amp; BIAN, G. (2026). In situ characteristics and patterns of settlement spatial differentiation along the Grand Canal\u0026rsquo;s Tianjin Section [J]. \u003cem\u003enpj Heritage Science\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(1), 128.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJIA, W. (2024). Morphological typology pedigree of centripetal spatial schema in traditional Chinese settlements [J]. \u003cem\u003eFrontiers of Architectural Research\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(6), 1198\u0026ndash;1214.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWANG, Z., ZHU, J., \u0026amp; WU, Z. (2024). Study on the spatial distribution characteristics of traditional villages and their response to the water network system in the lower yangtze river basin [J]. \u003cem\u003eScientific Reports\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(1), 22586.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLIU, Y., HE, S., WU, F., et al. (2010). Urban villages under China's rapid urbanization: Unregulated assets and transitional neighbourhoods [J]. \u003cem\u003eHabitat International\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(2), 135\u0026ndash;144.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePENG, P., ZHANG, Y., ZHAO, J., et al. (2025). A study on the spatial distribution and driving factors of traditional villages\u0026ndash;a case study of the Beijing-Tianjin-Hebei region in China [J]. \u003cem\u003eScientific Reports\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(1), 28516.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLIU, W., \u0026amp; XUE, Y. (2023). Spatial distribution analysis and driving factors of traditional villages in Henan province: a comprehensive approach via geospatial techniques and statistical models [J]. \u003cem\u003eHeritage Science\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), 185.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSHI, Y., WANG, W., ZHANG, J., et al. (2025). UAV remote sensing and deep learning for assessing and optimizing architectural texture in traditional villages [J]. \u003cem\u003enpj Heritage Science\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(1), 325.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZHANG, P., \u0026amp; GHOSH, D. (2023). Spatial measures and methods in sustainable urban morphology: A systematic review [J]. \u003cem\u003eLandscape and Urban Planning\u003c/em\u003e, \u003cem\u003e237\u003c/em\u003e, 104776.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLIU, C., CAO, Y., YANG, C., et al. (2020). Pattern identification and analysis for the traditional village using low altitude UAV-borne remote sensing: multifeatured geospatial data to support rural landscape investigation, documentation and management [J]. \u003cem\u003eJournal of Cultural Heritage\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e, 185\u0026ndash;195.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLI, Z., LI, D., LAI, W., et al. (2025). A deep learning-based method for identifying traditional villages across cultural types [J]. \u003cem\u003enpj Heritage Science\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(1), 671.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZHANG, J., ZHU, Z., DENG, J. (2025). Multi-Modal Feature Fusion for Spatial Morphology Analysis of Traditional Villages via Hierarchical Graph Neural Networks [J]. ArXiv, abs/2510.27208.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCAMPIANI, A., LINGLE, A., \u0026amp; LERCARI, N. (2019). Spatial analysis and heritage conservation: Leveraging 3-D data and GIS for monitoring earthen architecture [J]. \u003cem\u003eJournal of Cultural Heritage\u003c/em\u003e, \u003cem\u003e39\u003c/em\u003e, 166\u0026ndash;176.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTANG, Z., \u0026amp; LI, M. (2020). Mapping Tea Plantations from VHR Images Using OBIA and Convolutional Neural Networks [J]. \u003cem\u003eRemote Sensing\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(18), 2935.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDADAO L. A Preliminary Study on Regional Analysis in Human Geography [J]. \u003cem\u003eActa Geographica Sinica\u003c/em\u003e, (1984). 51(4).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYANBO, Q., \u0026amp; SHUWEN, W. (2019). Spatial morphology of rural settlements based on site and functional characteristics [J]. \u003cem\u003eResources Science\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(6), 1035\u0026ndash;1047.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePEILIN L. (2014). \u003cem\u003eTraditional settlement cultural landscape gene [M]\u003c/em\u003e. Commercial.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYIGANG P. (1992). \u003cem\u003eLandscape analysis of traditional villages settlement [M]\u003c/em\u003e. China Architecture \u0026amp; Building.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQING C. (2019). A perspective of the traditional settlements from ancient to now commemorating the ninetieth anniversary of the society for research in Chinese architecture [J]. \u003cem\u003eArchitectural Journal\u003c/em\u003e, \u003cem\u003e2019\u003c/em\u003e(12), 14\u0026ndash;19.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJIE, Z., \u0026amp; SONGNAN, W. (2010). Quantitative analysis of the morphologies of traditional Chinese villages [J]. \u003cem\u003eWorld Architecture\u003c/em\u003e, \u003cem\u003e(2010\u003c/em\u003e(1), 118\u0026ndash;121.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGUOGIN, J. I. N. D. (2015). Z. \u003cem\u003ePolicy-induced contemporary village form:a field survey from Wuxi city [J]\u003c/em\u003e (Vol. 2015, pp. 65\u0026ndash;71). Urban Planning Forum. 2.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWEI, T., \u0026amp; HONGYE, C. (2013). Spatial form and spatial cognition of traditional village insyntactical view: A case study of Xiaozhou Village,Guangzhou [J]. \u003cem\u003eActa Geographica Sinica\u003c/em\u003e, \u003cem\u003e68\u003c/em\u003e(02), 209\u0026ndash;218.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBINGYAO L. Quantitative Methods and Evaluation of Urban Spatial Form [J]. \u003cem\u003eUrban Planning Forum\u003c/em\u003e, (1998). (03): 42\u0026ndash;45.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKAI G. Urban morphology: an introduction and evaluation of the theories and the methods [J]. \u003cem\u003eCity Planning Review\u003c/em\u003e, (2001). (12): 36\u0026ndash;42.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYONG J. Spatial Political Economy Paradigm of Urban Design \u0026mdash;\u0026mdash; Review of Designing Cities by A. R. Cuthbert [J]. Urban Planning International, (2005). (06): 73\u0026ndash;77.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHONG, J. (2015). Transaction costs, property rights allocation and the evolution of urban spatial form: an analysis based on the perspective of new institutional economics [J]. \u003cem\u003eUrban Planning Forum\u003c/em\u003e, (06): 63\u0026ndash;69.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLINGLI B. Spatial transaction costs: the evolution of the spatial form of industrial organization [J]. \u003cem\u003eAcademic Research\u003c/em\u003e, (2010). (05): 80\u0026ndash;83.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLI, H., \u0026amp; SONG, W. (2020). Pattern of spatial evolution of rural settlements in the Jizhou District of China during 1962\u0026ndash;2030 [J]. \u003cem\u003eApplied Geography\u003c/em\u003e, \u003cem\u003e122\u003c/em\u003e, 102247.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLUO, X. (2021). Knowledge Map on the Study of Chinese Rural Settlement Form Based on CSSCI [J]. \u003cem\u003eEconomic Geography\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(04), 148\u0026ndash;157.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOLIVEIRA, V. (2025). PORTA S. Quantitative and qualitative analysis in urban morphology: systematic legacy and latest developments [J]. Proceedings of the Institution of Civil Engineers - Urban Design and Planning, 178(2): 75\u0026ndash;87.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXIANKUI L. The Cultural Character of Ancient Bashu Architecture [J]. \u003cem\u003eArchitectural Journal\u003c/em\u003e, (1995). (03): 48\u0026ndash;53.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYILI W (2002). Diaojiaolou - a treasure of Tujia folk houses [J]. \u003cem\u003eChina Engineering Consulting\u003c/em\u003e, (10): 48\u0026ndash;49.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJUN W (2006). Form study on Xi'an traditional dwellings in the ancient city zone of Xi'an [D].\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZURUN Y (2001). Analysis on the spatial structure of traditional settlement environment [J]. \u003cem\u003eArchitectural Journal\u003c/em\u003e, (12): 21\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJUN W. (2009). \u003cem\u003eSpatial Concepts in Traditional Settlement Structure [M]\u003c/em\u003e. China Architecture \u0026amp; Building.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXINCHENG P. (2013). \u003cem\u003eResearch on the quantitative method of the overall morphology of traditional rural settlements in two-dimensional plane [M]\u003c/em\u003e. Southeast University.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYING, S., \u0026amp; DAWEI, X. (2015). A Space Syntax Analysis of Hakka Historical Villages in Meizhou [J]. \u003cem\u003eUrban Development Studies\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(05), 63\u0026ndash;70.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDANDONG, G., LEI, T., NING, W., et al. (2017). Parametric analysis and reconstruction method of village's road morphology [J]. \u003cem\u003eJournal of Zhejiang University(Engineering Science)\u003c/em\u003e, \u003cem\u003e51\u003c/em\u003e(02), 279\u0026ndash;286.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eF W. Simulating urban encroachment on rural land with fuzzy-logiccontrolled cellular automata in a geographical information system [J]. \u003cem\u003eJournal of environmental management\u003c/em\u003e, (1998). 53(4): 293\u0026ndash;308.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR W, G E. Cellular automata and fractal urban form:a cellular modelling approach to the evolution of urban land-use patterns [J]. \u003cem\u003eEnvironment\u0026amp;planning\u003c/em\u003e, (1993). 25(8): 1175\u0026ndash;1199.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTIAN, G., QIAO, Z., \u0026amp; GAO, X. (2014). Rural settlement land dynamic modes and policy implications in Beijing metropolitan region. \u003cem\u003eChina [J] Habitat International\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e, 237\u0026ndash;246.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYING, L., XIAOBIN, J., MIAOYI, L., et al. (2014). A constrained cellular automata model for reconstructing historical arable land in Jiangsu province [J]. \u003cem\u003eGeographical Research\u003c/em\u003e, \u003cem\u003e33\u003c/em\u003e(12), 2239\u0026ndash;2250.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYAWEI, K., ZIYU, Z., ZENGDA, G. (2023). Study on the Causes of Spatial Morphology of RuralSettlements in the Yellow River Delta under the Influenceof Basin Environment [J]. \u003cem\u003eArchitectural Journal\u003c/em\u003e, (S2): 70\u0026ndash;75.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eESCH, T., MARCONCINI, M., MARMANIS, D., et al. (2014). Dimensioning urbanization \u0026ndash; An advanced procedure for characterizing human settlement properties and patterns using spatial network analysis [J]. \u003cem\u003eApplied Geography\u003c/em\u003e, \u003cem\u003e55\u003c/em\u003e, 212\u0026ndash;228.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMOU, L., \u0026amp; LI, H. (2024). Identification and Spatial Characterization of suburban areas in Chengdu [J]. \u003cem\u003eApplied Geography\u003c/em\u003e, \u003cem\u003e172\u003c/em\u003e, 103428.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZUO, L., \u0026amp; TAO, X. (2021). Research on the spatial morphology of natural villages based on machine learning identification; proceedings of the 2020/2021 China Urban Planning Annual Conference and 2021 China Urban Planning Academic Season, Chengdu, China, F, [C].\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZHE, L., XINYI, L., JIAYING, S., et al. (2023). Quantitative Research of Landscape Texture for Polders inXuanwu Plain Based on Morphological Index:A Case Studyof Yongfeng Polder,Gucheng Lake [J]. \u003cem\u003eChinese Landscape Architecture\u003c/em\u003e, \u003cem\u003e39\u003c/em\u003e(01), 41\u0026ndash;46.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePENG, F., JING, X., ZHIQI, Z., et al. (2022). The Method of Space-Dynamic Coupling Mechanism ofRural Settlements Based on Machine Learning: TakingLiyang City, Jiangsu Province as an Example [J]. \u003cem\u003eJournal of Human Settlements in West China\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(04), 1\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWU, S., DI, B., USTIN S, L, et al. (2022). Classification and detection of dominant factors in geospatial patterns of traditional settlements in China [J]. \u003cem\u003eJournal of Geographical Sciences\u003c/em\u003e, \u003cem\u003e32\u003c/em\u003e(5), 873\u0026ndash;891.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCHEN, W., YANG, L., WU, J., et al. (2023). Spatio-temporal characteristics and influencing factors of traditional villages in the Yangtze River Basin: a Geodetector model [J]. \u003cem\u003eHeritage Science\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e, 111.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTENGYUN, H., PENGFEI, X., YANAN, W., et al. (2023). Research on Building Footprints Extraction Methods based on Different Deep Learning Models [J]. \u003cem\u003eRemote Sensing Technology and Application\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e(04), 892\u0026ndash;902.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWANG, M., XU, W., CAO, G., et al. (2024). Identification of rural courtyards\u0026rsquo; utilization status using deep learning and machine learning methods on unmanned aerial vehicle images in north China [J]. \u003cem\u003eBuilding Simulation\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e, 799\u0026ndash;818.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePANPAN, Z., SHUAIPENG, L., LIQIANG, Z., et al. (2021). Multitask Learning-based Building Extraction from High-Resolution Remote Sensing Images [J]. \u003cem\u003eJournal of Geo-information Science\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(03), 514\u0026ndash;523.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWANG, Q., YUAN, Q., LI, Z., et al. (2025). Study on the spatial relationships of traditional regional dwellings in Huizhou district [J]. \u003cem\u003enpj heritage science\u003c/em\u003e, \u003cem\u003e238\u003c/em\u003e, 534.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWANG, Y., \u0026amp; LI, X. (2024). The AI\u0026rsquo;s Cognition of the Cultural Characteristics Underlying Chinese Courtyard Dwelling A visual perception-based approach for Siheyuan plan classification using machine learning and convolutional neural networks [Z]. Data-Driven Intelligence Proceedings of the 42nd Conference on Education and Research in Computer Aided Architectural Design in Europe. Nicosia. : 643\u0026ndash;652.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCHEN, Y., WANG, T., CHEN, G. (2025). A balanced multiscale object detection network for building fa\u0026ccedil;ade elements detection [J]. \u003cem\u003eAdvanced Engineering Informatics\u003c/em\u003e, : 103289.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRAN, X. (2019). HAISHAN X. A Saliency-model Based Quantitative Study and Evaluation on the Perception of Visual Attributes of Building Facades [J]. \u003cem\u003eNew Architecture\u003c/em\u003e, (06): 26\u0026ndash;31.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSUN, M., ZHANG, F., DUARTE, F., et al. (2022). Understanding architecture age and style through deep learning [J]. \u003cem\u003eCities\u003c/em\u003e, \u003cem\u003e128\u003c/em\u003e, 103787.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJI-WEI Z, DE-ER L, PENG, Y., et al. (2020). \u003cem\u003eWindow extraction method in dual depth image of point cloud of building facade [J]\u003c/em\u003e (Vol. 50, pp. 875\u0026ndash;881). Laser \u0026amp; Infrared. 07.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZHANG, L., CHEN, Y., ZHENG, L., et al. (2024). Investigating the Surface Damage to Fuzhou\u0026rsquo;s Ancient Houses (Gu-Cuo) Using a Non-Destructive Testing Method Constructed via Machine Learning [J]. \u003cem\u003eCoatings\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(11), 1466.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYANG, S., CHEN, Y., ZHENG, L., et al. (2025). Investigating and Identifying the Surface Damage of Traditional Ancient Town Residence Roofs in Western Zhejiang Based on YOLOv8 Technology [J]. \u003cem\u003eCoatings\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(2), 205.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTING, Y., \u0026amp; SHAN, Z. (2016). A Study on the Application of Digital Technologies of Architectural Surveying in the Renovation of Historic Districts Practice Based in the Pearl River Delta [J]. \u003cem\u003eArchitectural Journal\u003c/em\u003e, (12): 22\u0026ndash;27.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMENG, C., SONG, Y., JI, J. (2022). Automatic classification of rural building characteristics using deep learning methods on oblique photography [J]. \u003cem\u003eBuilding Simulation\u003c/em\u003e, (40): 1078.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXIAOWEI, H., JIE, Z., WEN, L., et al. (2024). \u003cem\u003eThe Technical Exploration and Practice of Deep LearningRecognition in Traditional Textures of HistoricalSettlements [J]\u003c/em\u003e (Vol. 42, pp. 87\u0026ndash;94). Development of Small Cities \u0026amp; Towns. 07.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMIU, T. (2026). Characteristics and periodization of self-built houses in the Quanzhou coastal area based on archaeological typology method (late 19th century to early 21st century) [J]. \u003cem\u003eFrontiers of Architectural Research\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(2), 522\u0026ndash;539.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZHANG, X., \u0026amp; LI, S. (2024). Classification of architectural styles in Chinese traditional settlements using remote sensing images and building facade pictures [J]. \u003cem\u003eJournal of Geographical Sciences\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(12), 2457\u0026ndash;2476.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJIANG, Y., \u0026amp; LI, N. (2023). Parametric Reconstruction of Traditional Village Morphology Based on the Space Gene Perspective\u0026mdash;The Case Study of Xiaoxi Village in Western Hunan. \u003cem\u003eChina [J] Sustainability\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(3), 2088.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZHANG, C., PING, X., FAN, Q., et al. (2024). Measurement of 2D and 3D fractal features of urban morphology from an architectural view and its influencing factors [J]. \u003cem\u003eFractal and Fractional\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e(3), 138.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYADAV, M. J. (2025). LAVATE M. Morphometric Analysis and Spatial Pattern of Settlements in the Sina River Basin: A Case Study of Solapur District, Maharashtra [J]. \u003cem\u003eIIP: International Multidisciplinary Research Journal\u003c/em\u003e, 2(Issue-IV (October-December)): 11\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKWINTA, A., \u0026amp; GNIADEK, J. (2017). The description of parcel geometry and its application in terms of land consolidation planning [J]. \u003cem\u003eComputers and Electronics in Agriculture\u003c/em\u003e, \u003cem\u003e136\u003c/em\u003e, 117\u0026ndash;124.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLUCCHI E, D'ALONZO V, EXNER, D. (2019 [C]). A density-based spatial cluster analysis supporting the building stock analysis in historical towns; proceedings of the Building Simulation 2019, F, IBPSA.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDUAN, Y., CHEN, S., ZHANG, L., et al. (2023). Spatial distribution characteristic and type classification of rural settlements: A case study of Weibei plain. \u003cem\u003eChina [J] Sustainability\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(11), 8736.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZHANG, J., FAN, B., LI, H., et al. (2023). Research on the shape classification method of rural homesteads based on parcel scale\u0026mdash;Taking Yangdun village as an example [J]. \u003cem\u003eRemote Sensing\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(19), 4763.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVIALARD, A., \u0026amp; CARPENTER, A. (2015). Building density of parcels and block-faces from a syntactical, morphological and planning perspective; proceedings of the Proceedings of the 10th international space syntax symposium, F, [C].\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLI, J., \u0026amp; LI, C. (2024). Characterizing urban spatial structure through built form typologies: A new framework using clustering ensembles [J]. \u003cem\u003eLand Use Policy\u003c/em\u003e, \u003cem\u003e141\u003c/em\u003e, 107166.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDE SMET F, T. E. L. L. E. R. J. (2016). Characterising the morphology of suburban settlements: A method based on a semi-automatic classification of building clusters [J]. \u003cem\u003eLandscape Research\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(1), 113\u0026ndash;130.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Traditional Settlements, Spatial Morphology, Multi-Scalar Analysis, Machine Learning, Image Segmentation, Cross-scale Mechanisms","lastPublishedDoi":"10.21203/rs.3.rs-9296889/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9296889/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe spatial morphology of traditional settlements results from long-term interactions between environmental constraints and socio-spatial processes, exhibiting pronounced multi-level and cross-scale characteristics. However, existing studies often remain at single-scale descriptions and lack systematic analysis of inter-scalar relationships. This study develops a multi-level analytical framework integrating settlement, parcel, and building scales, using Fulin Village in Quanzhou, China, as a case study. Machine learning and image segmentation are employed to identify spatial units and examine their structural relationships.\u003c/p\u003e \u003cp\u003eThe results show that Fulin Village presents a water-oriented clustered pattern with differentiated zones dominated by distinct building types. More importantly, clear cross-scale interactions are identified: the settlement-level structure is transmitted through parcel subdivision and organization and manifested at the building level as specific spatial forms and typological distributions. The parcel level acts as an intermediary that translates and organizes spatial structure, while the aggregation and differentiation of building types feed back to reinforce parcel organization and settlement zoning. This process reflects the coupling of environmental constraints shaping overall patterns, socio-spatial mechanisms operating through parcel organization, and building-level morphological expression across scales.\u003c/p\u003e \u003cp\u003eThis study conceptualizes traditional settlement morphology as a cross-scale mechanism of \u0026ldquo;macro-level constraints \u0026ndash; intermediary translation \u0026ndash; micro-level representation.\u0026rdquo; By integrating data-driven methods with morphological analysis, it provides a structured analytical approach for understanding and managing traditional settlements.\u003c/p\u003e","manuscriptTitle":"Data-driven Multi-scalar Analysis of Spatial Morphology and Cross-scale Interactions in Traditional Settlements: A Case Study of Fulin Village, China","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-20 10:20:29","doi":"10.21203/rs.3.rs-9296889/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"0e774b13-1992-4b87-ac2f-f26e2a034d12","owner":[],"postedDate":"April 20th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-14T07:29:24+00:00","index":23,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-20T10:20:29+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-20 10:20:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9296889","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9296889","identity":"rs-9296889","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}