Application of Artificial Intelligence Algorithms for Earthquake Damage Mapping Using Very High-Resolution Satellite Imagery: Case Studies from Al Hoceima, Morocco, and Gaziantep, Turkey

Application of Artificial Intelligence Algorithms for Earthquake Damage Mapping Using Very High-Resolution Satellite Imagery: Case Studies from Al Hoceima, Morocco, and Gaziantep, Turkey | 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 Application of Artificial Intelligence Algorithms for Earthquake Damage Mapping Using Very High-Resolution Satellite Imagery: Case Studies from Al Hoceima, Morocco, and Gaziantep, Turkey Wardia Boutalkhoukhte, Ahmed Algouti This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7769060/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 Seismic events continue to challenge the resilience of urban systems, particularly in regions marked by tectonic complexity and high population density. In such contexts, timely and accurate damage assessment remains critical for effective emergency response. This study explores the potential of very high-resolution (VHR) satellite imagery integrated with artificial intelligence (AI) to enhance the spatial understanding of post-earthquake structural impacts. By focusing on two distinct seismic events, the 2004 Al Hoceima earthquake in Morocco and the 2023 Gaziantep earthquake in Turkey. The research underscores the utility of AI-supported remote sensing in overcoming the limitations of conventional ground-based methods. Emphasis is placed on the capacity of VHR imagery to capture subtle spatial variations that are often overlooked during manual assessments. The findings highlight the potential of integrating AI with satellite data as a scalable, transferable, and practical solution for disaster risk management. Broader implications are drawn for future applications in real-time decision making, urban resilience planning, and the development of data-driven humanitarian strategies. Earthquake Remote Sensing Deep Learning Classification Satellite Imagery Artificial Intelligence Algorithms 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 Figure 13 1 Introduction Earthquakes are among the most destructive natural hazards, capable of causing extensive human and economic losses and long-term societal disruptions (Chaudhary & Piracha 2021 ). The growing intersection between seismic activity and unregulated development has led to heightened structural vulnerability in North Africa and the Middle East. Traditional methods used to assess post-earthquake damage have historically relied on ground-based inspections. These approaches have faced significant limitations during large-scale events, although known for their accuracy at local scales. Restricted access to disaster zones, prolonged deployment times, and high operational costs have constrained their utility, especially in densely populated or remote areas (Subash Ghimire 2023 ). Remote sensing has emerged as a transformative tool for disaster assessment in response to these challenges. Very high-resolution satellite imagery has enabled broad, repeatable, and rapid evaluations of earthquake-induced destruction. The capacity to observe spatial damage patterns from orbit has supported improved situational awareness in environments where timely decision-making is critical (Karls and Braun 2019 ). The integration of artificial intelligence (AI) and deep learning (DL) into satellite-based assessment has further enhanced analytical capabilities. These technologies have enabled the automation of damage detection, offering scalable alternatives to manual interpretation. Existing studies often remain constrained by localised case studies that lack cross-regional validation and broader applicability. The present study builds upon this foundation by examining the applicability of AI-assisted damage assessment across distinct seismic environments using satellite-derived data. 2. Study Area and Data 2.1 Geographical and Tectonic Context Al Hoceima in northern Morocco and Gaziantep in southeastern Turkey have been selected based on their tectonic relevance and documented seismic activity. Both locations lie along active fault zones where plate interactions have intensified earthquake risks (van der Woerd et al. 2014 ). The placement within global tectonic frameworks is illustrated in Fig. 1 . Al Hoceima is located in the central Rif region, where the convergence of the African and Eurasian tectonic plates has created a seismically complex environment. The 2004 earthquake (Mw 6.3) led to significant structural damage and loss of life (ARAB Oussama 2021). The area's tectonic layout, which highlights structural deformation zones that justify its selection, is depicted in Fig. 2 . Gaziantep is located along the East Anatolian Fault, which remains one of Turkey's most active tectonic structures. The 2023 earthquake caused widespread destruction, which emphasised the city’s vulnerability to seismic shocks (OzturkArslan and Korkmaz 2023). Key structural features relevant to the region's seismic behaviour are presented in Fig. 3 . 2.2 Satellite and Ancillary Datasets High-resolution satellite imagery was employed to capture pre- and post-earthquake conditions in both sites. QuickBird imagery (50 cm) was used for Al Hoceima, while Airbus imagery (2 m) supported damage mapping in Gaziantep. Supplementary GIS layers such as road networks and building footprints were used to strengthen spatial analysis. Data Preprocessing involved geometric correction, radiometric normalisation, and derivation of NDVI and NDBI indices. Image co-registration was conducted within QGIS and ArcGIS Pro environments, ensuring spatial alignment. Ancillary layers, e.g., cadastral maps, road networks, and building footprints, were also integrated for spatial accuracy (Benchekroun and Chakir 2004; Gaziantep Municipality Report 2023). 3. Methodology A structured remote sensing framework was developed to evaluate post-earthquake structural damage in Al Hoceima (Morocco) and Gaziantep (Turkey). The approach combined very high-resolution (VHR) satellite imagery with AI-based classification models, allowing consistent detection across contrasting urban and seismic settings. Key methodological steps are illustrated in Figure 4. QuickBird multispectral imagery from April 2004 was utilised for Al Hoceima, while Airbus RGB imagery from 2023 was used for Gaziantep. Both datasets provided sub-meter to 2-meter resolution, which supported urban-scale damage delineation. Ancillary geospatial layers, including building footprints, cadastral maps, and road networks, were integrated to enhance spatial coherence. Ground truth data were established using field reports, municipal documentation, and manual digitisation of imagery. Training samples captured five thematic classes: intact structures, moderate and complete collapses, vegetation, and open spaces. Multiple AI classifiers were tested. Random Forest served as a traditional machine learning benchmark. Deep learning models, including U-Net and Mask R-CNN, were selected for their capabilities in semantic segmentation and object-level classification. 3.1 Random Forest (RF) The Random Forest algorithm, consisting of an ensemble of decision trees, was implemented using Python’s Scikit-learn library. Its robustness in handling high-dimensional multispectral data and its reduced susceptibility to overfitting made it a suitable baseline. Key hyperparameters such as the number of trees, maximum depth, and minimum samples per leaf were tuned through grid search validation (Zourarah et al. 2005). 3.2 U-Net Deep Learning Model U-Net, a convolutional neural network designed for semantic segmentation, was deployed for pixel-level classification of damage zones. Input layers included RGB bands and spectral indices. The architecture comprised an encoder-decoder structure with skip connections that improved localisation. The model was implemented using Keras and TensorFlow frameworks and trained using Google Colab GPU instances for 200 epochs. 3.3 Mask R-CNN Mask R-CNN was applied to detect and delineate individual damaged buildings. Its two-stage framework, Region Proposal Network (RPN) and classification enabled precise object extraction. Model performance was evaluated using confusion matrices and the following metrics: overall accuracy, precision, recall, F1-score, and Kappa coefficient. Ground truth polygons were used for testing. Accuracy results are summarised in Tables 1-6. Table 1: Confusion matrix for the SVM model Result of KNN Classification Reference Destroyed building Undamaged building Tree Field Bare soil Road Total reference Destroyed building 18 4 3 1 13 1 40 Undamaged building 1 40 0 0 0 0 41 Tree 0 1 14 1 1 0 17 Field 0 0 0 7 3 2 12 Bare soil 2 1 1 1 62 3 70 Road 2 1 0 0 6 12 21 Total classified 23 47 18 10 85 18 201 Précision 75,879397 Kappa 0,68261563 Table 2:Confusion matrix for the KNN model Result of the SVM classification Reference Destroyed building Undestroyed building Tree Field Bare soil Road Total reference Destroyed building 15 11 2 5 5 3 41 Undestroyed building 6 25 0 1 0 2 34 Tree 0 0 13 4 0 0 17 Field 1 1 2 5 4 1 14 Bare soil 1 2 1 3 58 6 72 Road 2 0 0 0 8 13 23 Total classified 25 39 18 18 76 25 200 Precision 65,60846561 Kappa 0,554051111 Table 3: Confusion matrix for the Decision Tree (DT) model Result of the DT classification Référence Destroyed building Undamaged building Tree Field Bare soil Road Total reference Destroyed building 26 4 1 1 3 1 39 Undamaged building 0 40 0 0 0 1 42 Tree 0 1 14 1 0 1 17 Field 0 1 1 7 3 2 12 Bare soil 2 1 0 3 59 7 70 Road 0 0 0 1 1 19 21 Total classified 26 47 16 13 69 30 200 Précision 83,2335025 Kappa 0,79143899 Table 4:Confusion matrix for the Random Forest (RF) model Result of the RF classification Reference Destroyed building Undestroyed building Tree Field Bare soil Road Total reference Destroyed building 15 5 2 1 16 0 39 Undestroyed building 0 40 0 0 0 0 40 Tree 0 2 11 2 1 0 16 Field 0 0 2 6 5 0 13 Bare soil 0 1 1 0 66 2 70 Road 0 1 0 0 10 11 22 Total classified 15 49 16 9 98 13 200 Précision 74,3718593 Kappa 0,655989424 Table 5: Confusion matrix for the Naive Bayes model Result of the Bayes classification Reference Destroyed building Undamaged building Tree Field Bare soil Road Total reference Destroyed building 27 2 2 1 5 2 39 Undamaged building 4 37 0 0 0 0 41 Tree 1 0 11 4 0 0 16 Field 0 0 1 10 0 2 13 Bare soil 6 1 1 3 48 10 69 Road 5 0 0 0 4 13 22 Total classified 43 40 15 18 57 27 200 Precision 72,5888325 Kappa 0,65341761 Table 6:Confusion matrix for the SharkRF model Result of the SHARKRF classification Reference Destroyed building Undestroyed building Tree Field Bare soil Road Total reference Destroyed building 7 14 3 0 15 0 39 Undestroyed building 1 33 0 0 7 0 41 Tree 3 0 11 1 1 0 16 Field 0 0 3 2 7 0 12 Bare soil 3 0 2 0 64 1 70 Road 0 1 0 0 10 11 22 Total classified 14 48 19 3 104 12 200 Précision 64,97461929 Kappa 0,527265772 Post-processing was executed in QGIS to remove classification noise and ensure topological integrity. Damage density hotspots were derived via kernel density estimation in Gaziantep. Model accuracy was assessed through metrics such as precision, recall, F1-score, and Kappa coefficient using reserved ground truth polygons. This dual-case design enabled a cross-contextual evaluation of model performance under varied urban geometries and seismic impact levels. 4. Results and Discussion 4.1 Evaluation Framework and Interpretation Strategy Post-earthquake damage detection models were tested in Al Hoceima and Gaziantep using very high-resolution imagery and artificial intelligence (AI). The evaluation employed a binary classification scheme, “Destruction” vs. “No Destruction”, to support operational crisis response. Performance metrics included overall accuracy and Cohen’s kappa coefficient. An 80/20 training-validation split ensured model generalizability. Confusion matrices revealed trends in misclassification, particularly false positives in nondamaged areas and false negatives in undetected damaged areas. False positives for non-damaged areas misclassified as damaged and false negatives for undetected actual damage were particularly critical in boundary zones between collapsed structures and open land (Labiak 2025 ). These diagnostic patterns guided model refinement. 4.2 Traditional Classifiers: MD and MLH Minimum Distance (MD) and Maximum Likelihood (MLH) were initially tested for baseline comparison. Both rely on pixel-level spectral similarity but ignore spatial context and object geometry, rendering them inadequate in post-earthquake debris environments (Rossi et al. 2020 ). Visual assessments (Figs. 5 and 6 ) showed substantial confusion between rubble and bare surfaces. MD achieved 47.36% accuracy with a kappa of 0.48, while MLH scored 46.25% and 0.42, respectively. These values suggest minimal reliability for field deployment. Spectral confusion, radiometric variability, and lack of spatial awareness accounted for their poor performance. Their operational limitations make them unsuitable for urban seismic mapping under crisis conditions. 4.3 Machine Learning Model Comparison Machine learning (ML) models showed improved but varied performance. Results are detailed in Tables 1 to 6 . SVM (Table 1 ) achieved 65.6% accuracy and a kappa of 0.55 KNN (Table 2 ) improved to 75.8% and 0.68 Decision Tree (DT) (Table 3 ) delivered the highest ML scores: 83.2% accuracy, 0.79 kappa. Random Forest (RF) (Table 4 ) followed at 74.3% and 0.65 Naive Bayes and SharkRF (Tables 5 and 6 ) showed lower reliability While DT showed strength in handling complex boundaries, performance dropped in heterogeneous urban zones. ML models lacked the spatial granularity required to differentiate structural forms in high-density neighbourhoods. This limited their applicability for emergency damage segmentation. (Fig. 7 – 11 ). 4.4 Deep Learning Results: Mask-RCNN Mask-RCNN outperformed all other models. As a region-based convolutional neural network, it captured both spectral and spatial patterns, detecting object boundaries with high precision. In Gaziantep, accuracy reached 90.48% with a kappa of 0.93. Al Hoceima returned 90.2% accuracy and 0.91 kappa (Figs. 12 and 13 ), indicating excellent agreement with ground truth (Alafandy et al. 2020 ). Success stemmed from the model’s object-based segmentation, spatial coherence, and architectural capacity to learn structural variation. These capabilities proved vital in recognising collapsed buildings amidst debris and overlapping features. Limitations included the requirement for annotated training data and vulnerability to atmospheric inconsistencies. Despite this, Mask-RCNN offered rapid post-training deployment, ideal for operational crisis mapping (AlAfandy et al. 2020 ). 4.5 Comparative Insights and Operational Implications A cross-model comparison confirmed the superiority of Mask-RCNN. Traditional models underperformed significantly. Compared with advanced machine learning algorithms such as DT and RF, or the deep learning model Mask-RCNN, traditional methods like MD and MLH are outperformed. While modern techniques reached accuracy levels above 80% and kappa values near 0.8, traditional approaches barely exceeded the 45% mark, with kappa coefficients under 0.5 (Alloghani et al. 2020). Only Mask-RCNN produced maps with sufficient fidelity for emergency planning (Tables 1 – 6 , Figs. 5 – 13 ). These differences are not only reflected in metrics but also the visual quality of the classification outputs. As seen by comparing Figs. 5 – 12 , which present the results for Al Hoceima and Turkey, deep learning delivers cleaner boundaries, reduced noise, and a closer match to the true spatial distribution of damage. Although MD and MLH require lower computational resources and are easier to implement, which may justify their use in resource-constrained contexts, their limitations in damage detection must be carefully considered before deployment in crisis scenarios. Deep learning approaches reduced processing time and subjective interpretation errors. Their real-time application capacity enhances field decision-making. Reliance on high-quality imagery and training data may constrain use in data-scarce environments (Saah et al. 2019 ). Hybrid frameworks combining open-source AI, cloud computing, and auxiliary datasets are recommended. Image fusion and AI-based super-resolution could mitigate reliance on costly VHR data (Shafique et al. 2022 ). 5 Conclusion Accurate identification of seismic damage has been recognised as essential for effective disaster response and recovery. Timely and spatially precise assessments are required so that emergency interventions can be prioritised and resources can be allocated where most needed. In the absence of reliable geospatial intelligence, critical delays in crisis response are often observed. The integration of very high-resolution (VHR) satellite imagery with artificial intelligence methods has been validated as a powerful approach to support post-earthquake damage mapping. By enabling detailed observation across large geographic extents, satellite-based remote sensing provides a scalable alternative where ground-based surveys are impractical or unsafe. This study confirmed that advanced object-based deep learning models, particularly Mask-RCNN, consistently outperformed traditional pixel-based classifiers in detecting damaged structures within complex urban environments. The cases of Al Hoceima and Gaziantep were selected due to their tectonic relevance and diverse urban characteristics. Through their analysis, the framework demonstrated strong model transferability and high classification accuracy, supported by kappa coefficients exceeding 0.9. These results emphasised the importance of integrating spectral and spatial information, especially where debris patterns and structural collapse are spatially heterogeneous. The study concludes that AI-driven methodologies provide significant advantages in terms of operational speed, reproducibility, and the reduction of human error. As satellite access improves and AI tools become increasingly open-source, future applications will benefit from enhanced model generalizability. Strengthening collaboration between scientific, governmental, and humanitarian sectors will remain vital in transforming these technological capabilities into real-world solutions for seismic resilience and disaster risk reduction. 6 Future Directions Several future research directions are proposed. The development of open and collaborative datasets is strongly recommended to improve the generalizability and robustness of deep learning models (Alloghani et al. 2020). Hybrid approaches should be explored, integrating deep learning with complementary data such as socioeconomic or environmental variables, to enhance accuracy and operational relevance in real-world scenarios (Shinde & Shah 2018). It is also advised to investigate methodologies that reduce reliance on VHR satellite images through the fusion of multiple satellite data sources or by implementing advanced AI techniques capable of enhancing the effective resolution of available imagery (Shravantandale 2023) Abbreviations AI Artificial Intelligence DL Deep Learning NDVI Normalized Difference Vegetation Index NDBI Naturalistic Developmental Behavioral Interventions VHR Very Highresolution RF Random Forest RPN Region Proposal Network MD Minimum Distance MLH Maximum Likelihood ML Machine learning DT Decision Tree RCNN Region-based Convolutional Neural Network SVM Support Vector Machine Declarations Competing Interests No conflicts of interest (student/supervisor relationship) Funding Not applicable (no external funding). Author Contributions All authors contributed to the study conception and design. Conceptualization, methodology, data acquisition and processing, writing – original draft, visualization by Wardia Boutalkhoukhte. Supervision, validation, methodology, writing – review & editing, critical revision by Ahmed Algouti. All authors read and approved the final manuscript. Acknowledgments We thank colleagues at the Faculty of Sciences Semlalia, Cadi Ayyad University, for constructive discussions and logistical support, and the agencies that provided satellite imagery and ancillary geospatial datasets used in this study. Data Availability Available upon request from the corresponding author. References AlAfandy, K.A., Hicham, H., Lazaar, M. and Achhab, M.A. (2020). Investment of Classic Deep CNNs and SVM for Classifying Remote Sensing Images. Advances in Science, Technology and Engineering Systems Journal , 5(5), pp.652–659. doi:https://doi.org/10.25046/aj050580. ARAB Oussama (2021). CONTRIBUTION UTILISANT L’ATTENUATION DES ONDES DE CODA ET LA LIQUEFACTION DES SOL DANS LA REGION DU RIF. Imist.ma . doi:http://toubkal.imist.ma/handle/123456789/33470. Chaudhary, M.T. and Piracha, A. (2021). Natural Disasters—Origins, Impacts, Management. Encyclopedia , [online] 1(4), pp.1101–1131. Available at: https://www.mdpi.com/2673-8392/1/4/84 Karls, E. and Braun, A. (2019). Radar satellite imagery for humanitarian response Bridging the gap between technology and application Dissertation der Mathematisch-Naturwissenschaftlichen Fakultät . Available at: https://publikationen.uni-tuebingen.de/xmlui/bitstream/handle/10900/91317/Braun2019%20Radar%20satellite%20imagery%20for%20humanitarian%20response%20UB.pdf?sequence=1 Labiak, R. (2025). A Method for detection and quantification of building damage using post-disaster LiDAR data . RIT Digital Institutional Repository. Available at: https://repository.rit.edu/theses/3023/ Ozturk, M., Arslan, M.H. and Korkmaz, H.H. (2023). Effect on RC buildings of 6 February 2023 Turkey earthquake doublets and new doctrines for seismic design. Engineering Failure Analysis , 153, p.107521. doi:https://doi.org/10.1016/j.engfailanal.2023.107521. Shafique, A., Cao, G., Khan, Z., Asad, M. and Aslam, M. (2022). Deep Learning-Based Change Detection in Remote Sensing Images: A Review. Remote Sensing , 14(4), p.871. doi:https://doi.org/10.3390/rs14040871. Subash Ghimire (2023). Data-driven-based models for seismic damage assessment of buildings : using machine learning at a regional scale and in-situ structural monitoring at building scale. Hal.science . doi:https://theses.hal.science/tel-04370250. van der Woerd, J., Dorbath, C., Ousadou, F., Dorbath, L., Delouis, B., Jacques, E., Tapponnier, P., Hahou, Y., Menzhi, M., Frogneux, M. and Haessler, H. (2014). The Al Hoceima Mw 6.4 earthquake of 24 February 2004 and its aftershocks sequence. Journal of Geodynamics , 77, pp.89–109. doi:https://doi.org/10.1016/j.jog.2013.12.004. Rossi, M., Minicozzi, G., Pascarella, G. and Capasso, A. (2020). ESG, Competitive Advantage and Financial performances: a Preliminary Research. Handle.net, 1(1), pp.969–986. Saah, D., Tenneson, K., Matin, M., Uddin, K., Cutter, P., Poortinga, A., Nguyen, Q.H., Patterson, M., Johnson, G., Markert, K., Flores, A., Anderson, E., Weigel, A., Ellenberg, W.L., Bhargava, R., Aekakkararungroj, A., Bhandari, B., Khanal, N., Housman, I.W. and Potapov, P. (2019). Land Cover Mapping in Data Scarce Environments: Challenges and Opportunities. Frontiers in Environmental Science, 7. doi:https://doi.org/10.3389/fenvs.2019.00150. 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-7769060","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":527860980,"identity":"31bbc8a2-c55a-4948-88a7-5789417495b9","order_by":0,"name":"Wardia Boutalkhoukhte","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABJUlEQVRIie3Qv0vDQBTA8YuBdHlt1juupP/ChUAgaNN/JXKQKVldFHoinIvVtUX/iW6OFwJxEeeIS0PBqYNT6SBirCJirD82h/sOx+PBhweHkE73L4P128UYIeNQvO7ww48kQvBG2HpHxn8mFL4BTLWzeXsVAjk/Ksjk8tFBrTy768uwVw+zrfLgC9LhHkQcaLeIyfSaeQhivp1K7gqImZkUDeIr8ClEJjg48Ukl2a7A9SaVyhAoQWZibSLDdzIU2F7SQKqBsBc1edpEcqAvZCpZhDBY1JCqPldfSWWDDPIOdy/iKyDjmAcT6bkSYi84ueFc4nuWp6cNQo5H2Wyxs+/gkme3I+n07FZelau9sH9m82qeLJu/bH5eWB8H1QQ6nU6n+0XP7EdeG9+PJRMAAAAASUVORK5CYII=","orcid":"https://orcid.org/0009-0007-0789-7518","institution":"Cadi Ayyad University Faculty of Sciences Semlalia: Universite Cadi Ayyad Faculte des Sciences Semlalia","correspondingAuthor":true,"prefix":"","firstName":"Wardia","middleName":"","lastName":"Boutalkhoukhte","suffix":""},{"id":527860981,"identity":"ad0f56c7-df9e-40de-8fd1-dc8fc090d393","order_by":1,"name":"Ahmed Algouti","email":"","orcid":"","institution":"Cadi Ayyad University Faculty of Sciences Semlalia: Universite Cadi Ayyad Faculte des Sciences Semlalia","correspondingAuthor":false,"prefix":"","firstName":"Ahmed","middleName":"","lastName":"Algouti","suffix":""}],"badges":[],"createdAt":"2025-10-02 21:11:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7769060/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7769060/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94400037,"identity":"eb423076-bd6a-4590-b551-006e242df256","added_by":"auto","created_at":"2025-10-27 13:57:59","extension":"xml","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7110,"visible":true,"origin":"","legend":"","description":"","filename":"beeeBEEED2500935.xml","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/41521b9c973ab02853c2e7e7.xml"},{"id":94399872,"identity":"cd0278e3-b87f-4e0e-a408-5edcf21cdcfa","added_by":"auto","created_at":"2025-10-27 13:57:50","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":957,"visible":true,"origin":"","legend":"","description":"","filename":"BEEED250093520166.go.xml","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/8af38a03dbc65cebcbafec2d.xml"},{"id":94399829,"identity":"88f53fe5-a29d-4ea6-8f89-c66bfbd76ee6","added_by":"auto","created_at":"2025-10-27 13:57:49","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":968,"visible":true,"origin":"","legend":"","description":"","filename":"BEEED2500935Import.xml","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/bf1105468fe6a206899a2005.xml"},{"id":94397929,"identity":"ee24ea7a-4f76-4753-96f6-46c688fa06bf","added_by":"auto","created_at":"2025-10-27 13:56:55","extension":"xml","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":97817,"visible":true,"origin":"","legend":"","description":"","filename":"BEEED25009350enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/b2e100dd54747eb06cf3f636.xml"},{"id":94399849,"identity":"9fe5286d-dfb8-4ce7-9581-26ce87b2ed1c","added_by":"auto","created_at":"2025-10-27 13:57:49","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":145909,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/661fa78e98f47f3f71914bde.jpeg"},{"id":94397843,"identity":"f8cb255e-df25-4378-a2fc-cc0e4ed97830","added_by":"auto","created_at":"2025-10-27 13:56:51","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":192923,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage10.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/c98211098950dd42e54f30b9.jpeg"},{"id":94398223,"identity":"3a49f736-618f-4098-95df-62100b510392","added_by":"auto","created_at":"2025-10-27 13:57:01","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":240230,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage11.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/219fa5014131c40c32b08296.jpeg"},{"id":94399654,"identity":"92ef8286-0af7-4699-a906-b7799f686b70","added_by":"auto","created_at":"2025-10-27 13:57:41","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":189213,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage12.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/49d571c42138d02d7db20074.jpeg"},{"id":94397836,"identity":"571c353b-7e79-4831-a249-df22bdf02caf","added_by":"auto","created_at":"2025-10-27 13:56:50","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":279735,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage13.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/e90d25047b16aa98248cfb43.jpeg"},{"id":94396991,"identity":"381b6d21-6819-4fa6-acbc-502dabf99404","added_by":"auto","created_at":"2025-10-27 13:56:23","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56512,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/b76fa6527164c52fd475900a.jpeg"},{"id":94397290,"identity":"fa459605-14d9-4fe4-813f-f029d4a5a496","added_by":"auto","created_at":"2025-10-27 13:56:36","extension":"jpeg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":722938,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/7c1dd1d62d608d9cb70b480c.jpeg"},{"id":94398538,"identity":"bd893139-c72e-4db8-84ba-f83169190d18","added_by":"auto","created_at":"2025-10-27 13:57:07","extension":"jpeg","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":104827,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/172de63543a4817cc4f6cff9.jpeg"},{"id":94397007,"identity":"d22071f8-1ddb-4835-8578-cb3ca6ce8352","added_by":"auto","created_at":"2025-10-27 13:56:23","extension":"jpeg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":211537,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/41be840596a8c9c61431e611.jpeg"},{"id":94399416,"identity":"944c8e63-db57-4cc2-aaf9-d979af7bf182","added_by":"auto","created_at":"2025-10-27 13:57:32","extension":"jpeg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":248015,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/2de90c096f00a2f83ec755b1.jpeg"},{"id":94399045,"identity":"fb662e40-69cb-4858-aa1b-c83ffe0efbb0","added_by":"auto","created_at":"2025-10-27 13:57:21","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1005603,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/32dc4ca4789ec4847566b3df.png"},{"id":94399925,"identity":"8f1cc5f5-9f91-46f6-acea-54a8520e52eb","added_by":"auto","created_at":"2025-10-27 13:57:52","extension":"jpeg","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":204076,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/b52be08fc253a7c56022ba00.jpeg"},{"id":94400069,"identity":"eab08424-eeb1-4f31-aa91-be568b77dbd8","added_by":"auto","created_at":"2025-10-27 13:58:00","extension":"jpeg","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":222598,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/62346e3e2922f2e43b5d697a.jpeg"},{"id":94398598,"identity":"81bff22e-3c09-4e54-b11e-e5412807c880","added_by":"auto","created_at":"2025-10-27 13:57:08","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":97772,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/1ff5000fe419933f09e00876.png"},{"id":94399040,"identity":"b39ab0bb-af57-4e7d-96e6-5d581235429b","added_by":"auto","created_at":"2025-10-27 13:57:20","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":155852,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/b76ec416b3285de1fd0f470f.png"},{"id":94397539,"identity":"6e277841-3ba8-409a-adf1-b156463d4699","added_by":"auto","created_at":"2025-10-27 13:56:42","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":199686,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/9fb135b3eb4d66a19d753d70.png"},{"id":94399926,"identity":"32364350-fe82-425e-8886-a7f831dc74a6","added_by":"auto","created_at":"2025-10-27 13:57:52","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":153325,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage12.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/7de4c3e5481350fd4b733c69.png"},{"id":94399930,"identity":"c47f9af2-3c9f-4554-b02c-e37b4ccccf9d","added_by":"auto","created_at":"2025-10-27 13:57:52","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":244764,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage13.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/0c66f1fb692ec289aaa6c5bd.png"},{"id":94398879,"identity":"670f10ab-0276-4f85-99d8-0efdb742aea5","added_by":"auto","created_at":"2025-10-27 13:57:14","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27767,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/1399fc66eecb1d730319ba89.png"},{"id":94398287,"identity":"72a20dcf-8e0d-4352-8580-248a769394a2","added_by":"auto","created_at":"2025-10-27 13:57:02","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":755633,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/99e8ea534a0e4b184c2fb345.png"},{"id":94399446,"identity":"06eaa16e-2d85-48cc-af73-7d77ae4b819c","added_by":"auto","created_at":"2025-10-27 13:57:33","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33787,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/d2161fce990717d40bd8d4ad.png"},{"id":94489286,"identity":"ed17d4e6-cc80-4d71-ab17-dacbaef556de","added_by":"auto","created_at":"2025-10-27 17:04:06","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":181321,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/9ed00a6933a337d8884d3f43.png"},{"id":94398490,"identity":"2de3f5f9-ea5c-47fb-bf14-d6f5cac87e80","added_by":"auto","created_at":"2025-10-27 13:57:07","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":205487,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/b70b60c2739e7be5663ebb32.png"},{"id":94397012,"identity":"e64ddcff-5b0f-4ff7-8395-b9df407a837b","added_by":"auto","created_at":"2025-10-27 13:56:24","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":231088,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/b18c2897a1da68e7058fab7c.png"},{"id":94399095,"identity":"4ddb3458-94fd-4afb-ace9-1e565c4c3c0a","added_by":"auto","created_at":"2025-10-27 13:57:21","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":173975,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/cd389e70e5f2302af60ba8a4.png"},{"id":94399596,"identity":"a3ed06fe-6fad-45ef-86bf-c32f549025c7","added_by":"auto","created_at":"2025-10-27 13:57:39","extension":"png","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":196185,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/49580c967dcae2c4c7b612e1.png"},{"id":94398999,"identity":"1028423f-0340-4e10-b5e6-b50b09194372","added_by":"auto","created_at":"2025-10-27 13:57:19","extension":"xml","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":96747,"visible":true,"origin":"","legend":"","description":"","filename":"BEEED25009350structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/4350d9d8f1f5d7814d360388.xml"},{"id":94398587,"identity":"c9bbebe4-2b07-475f-96d1-7bb77cd41aef","added_by":"auto","created_at":"2025-10-27 13:57:08","extension":"html","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":109396,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/9fe07b4700fdc971cc2ef5fc.html"},{"id":94398652,"identity":"6e54912a-552f-47ad-8db3-2667ad596203","added_by":"auto","created_at":"2025-10-27 13:57:10","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":145909,"visible":true,"origin":"","legend":"\u003cp\u003eGlobal distribution of tectonic plate boundaries\u003c/p\u003e","description":"","filename":"image1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/4c5dfbf8347e9556287b2144.jpeg"},{"id":94400034,"identity":"0ef977d4-46be-40b9-bd1f-8c618cecf432","added_by":"auto","created_at":"2025-10-27 13:57:58","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":65287,"visible":true,"origin":"","legend":"\u003cp\u003eStructural tectonic map of the Al Hoceima region showing fault lines, the Nekkour\u003c/p\u003e","description":"","filename":"image2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/68156978fc96622552b6fa01.jpeg"},{"id":94399009,"identity":"c076260f-1f1c-4786-b724-e6e97cefca95","added_by":"auto","created_at":"2025-10-27 13:57:19","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":722938,"visible":true,"origin":"","legend":"\u003cp\u003ePost-earthquake satellite image of Gaziantep\u003c/p\u003e","description":"","filename":"image3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/7aafa83793be912328713c2f.jpeg"},{"id":94397291,"identity":"43b088b1-242a-44b6-bf37-c49f27fccee3","added_by":"auto","created_at":"2025-10-27 13:56:36","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":104827,"visible":true,"origin":"","legend":"\u003cp\u003eMethodological framework for post-seismic damage assessment using VHR satellite\u003c/p\u003e","description":"","filename":"image4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/9b33ab57e897f1aa7c7b005a.jpeg"},{"id":94397295,"identity":"89b77fce-4b28-417d-954b-1618b39638b6","added_by":"auto","created_at":"2025-10-27 13:56:36","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":255137,"visible":true,"origin":"","legend":"\u003cp\u003eMinimum Distance supervised classification for Gaziantep region\u003c/p\u003e","description":"","filename":"image5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/98e4a67132240f7cfebb38ce.jpeg"},{"id":94399785,"identity":"a4b504b3-b96b-4373-979e-d471aead01b1","added_by":"auto","created_at":"2025-10-27 13:57:44","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":297336,"visible":true,"origin":"","legend":"\u003cp\u003eMaximum Likelihood supervised classification method for Gaziantep region\u003c/p\u003e","description":"","filename":"image6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/3dd273c276aeb03e5e37710b.jpeg"},{"id":94400138,"identity":"5966b211-383a-4219-88c8-e13cb070a56b","added_by":"auto","created_at":"2025-10-27 13:58:01","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1005603,"visible":true,"origin":"","legend":"\u003cp\u003eLand cover classification map integrating damaged structures, natural elements, and\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/07bb71cf159495b14eedbf6a.png"},{"id":94396978,"identity":"2f7ff289-69ef-4725-b2bd-c3d1d379257b","added_by":"auto","created_at":"2025-10-27 13:56:23","extension":"jpeg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":204076,"visible":true,"origin":"","legend":"\u003cp\u003ePost-classification damage mapping using supervised machine learning algorithm\u003c/p\u003e","description":"","filename":"image8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/22ca14e4327ecdd61420f65e.jpeg"},{"id":94399940,"identity":"913e7daf-73cf-4398-b787-41c4419b7f1e","added_by":"auto","created_at":"2025-10-27 13:57:53","extension":"jpeg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":222598,"visible":true,"origin":"","legend":"\u003cp\u003eClassification map output of Al Hoceima derived using the KNN algorithm\u003c/p\u003e","description":"","filename":"image9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/340f0cf43439e78818269d98.jpeg"},{"id":94397846,"identity":"da6edd05-c276-4266-800c-7dada7497765","added_by":"auto","created_at":"2025-10-27 13:56:51","extension":"jpeg","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":192923,"visible":true,"origin":"","legend":"\u003cp\u003ePost-earthquake classification of Al Hoceima using the Decision Tree (DT) model\u003c/p\u003e","description":"","filename":"image10.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/83c4668444c30c99cb0e44e8.jpeg"},{"id":94398658,"identity":"7a81e127-ef32-462f-9925-b9d15357afbb","added_by":"auto","created_at":"2025-10-27 13:57:10","extension":"jpeg","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":240230,"visible":true,"origin":"","legend":"\u003cp\u003eClassification output of Al Hoceima using the Random Forest (RF) algorithm\u003c/p\u003e","description":"","filename":"image11.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/336eb4e983741c6938700eb3.jpeg"},{"id":94399656,"identity":"143a91d8-c800-4ef8-a210-08a4c892f47f","added_by":"auto","created_at":"2025-10-27 13:57:41","extension":"jpeg","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":189213,"visible":true,"origin":"","legend":"\u003cp\u003eClassification result for Al Hoceima using the Mask-RCNN deep learning model\u003c/p\u003e","description":"","filename":"image12.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/8bbf5a4fd4b797cfb9f76db7.jpeg"},{"id":94398626,"identity":"41e810f9-6c75-477f-afa4-73a5f8d8a5b5","added_by":"auto","created_at":"2025-10-27 13:57:09","extension":"jpeg","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":279735,"visible":true,"origin":"","legend":"\u003cp\u003eDestruction map generated by the Mask-RCNN model\u003c/p\u003e","description":"","filename":"image13.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/e01ceafcdc6e86e853dcc873.jpeg"},{"id":101751301,"identity":"0282f455-27d4-4dac-a3e2-8558bb9cea6c","added_by":"auto","created_at":"2026-02-03 10:19:06","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5856890,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7769060/v1/4a4aef59-2737-45fb-835a-e6ec35b2a860.pdf"}],"financialInterests":"","formattedTitle":"Application of Artificial Intelligence Algorithms for Earthquake Damage Mapping Using Very High-Resolution Satellite Imagery: Case Studies from Al Hoceima, Morocco, and Gaziantep, Turkey","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eEarthquakes are among the most destructive natural hazards, capable of causing extensive human and economic losses and long-term societal disruptions (Chaudhary \u0026amp; Piracha \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The growing intersection between seismic activity and unregulated development has led to heightened structural vulnerability in North Africa and the Middle East.\u003c/p\u003e\u003cp\u003eTraditional methods used to assess post-earthquake damage have historically relied on ground-based inspections. These approaches have faced significant limitations during large-scale events, although known for their accuracy at local scales. Restricted access to disaster zones, prolonged deployment times, and high operational costs have constrained their utility, especially in densely populated or remote areas (Subash Ghimire \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eRemote sensing has emerged as a transformative tool for disaster assessment in response to these challenges. Very high-resolution satellite imagery has enabled broad, repeatable, and rapid evaluations of earthquake-induced destruction. The capacity to observe spatial damage patterns from orbit has supported improved situational awareness in environments where timely decision-making is critical (Karls and Braun \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe integration of artificial intelligence (AI) and deep learning (DL) into satellite-based assessment has further enhanced analytical capabilities. These technologies have enabled the automation of damage detection, offering scalable alternatives to manual interpretation. Existing studies often remain constrained by localised case studies that lack cross-regional validation and broader applicability.\u003c/p\u003e\u003cp\u003eThe present study builds upon this foundation by examining the applicability of AI-assisted damage assessment across distinct seismic environments using satellite-derived data.\u003c/p\u003e"},{"header":"2. Study Area and Data","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Geographical and Tectonic Context\u003c/h2\u003e\u003cp\u003eAl Hoceima in northern Morocco and Gaziantep in southeastern Turkey have been selected based on their tectonic relevance and documented seismic activity. Both locations lie along active fault zones where plate interactions have intensified earthquake risks (van der Woerd et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). The placement within global tectonic frameworks is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eAl Hoceima is located in the central Rif region, where the convergence of the African and Eurasian tectonic plates has created a seismically complex environment. The 2004 earthquake (Mw 6.3) led to significant structural damage and loss of life (ARAB Oussama 2021). The area's tectonic layout, which highlights structural deformation zones that justify its selection, is depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eGaziantep is located along the East Anatolian Fault, which remains one of Turkey's most active tectonic structures. The 2023 earthquake caused widespread destruction, which emphasised the city\u0026rsquo;s vulnerability to seismic shocks (OzturkArslan and Korkmaz 2023). Key structural features relevant to the region's seismic behaviour are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Satellite and Ancillary Datasets\u003c/h2\u003e\u003cp\u003eHigh-resolution satellite imagery was employed to capture pre- and post-earthquake conditions in both sites. QuickBird imagery (50 cm) was used for Al Hoceima, while Airbus imagery (2 m) supported damage mapping in Gaziantep. Supplementary GIS layers such as road networks and building footprints were used to strengthen spatial analysis. Data Preprocessing involved geometric correction, radiometric normalisation, and derivation of NDVI and NDBI indices. Image co-registration was conducted within QGIS and ArcGIS Pro environments, ensuring spatial alignment. Ancillary layers, e.g., cadastral maps, road networks, and building footprints, were also integrated for spatial accuracy (Benchekroun and Chakir 2004; Gaziantep Municipality Report 2023).\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Methodology","content":"\u003cp\u003eA structured remote sensing framework was developed to evaluate post-earthquake structural damage in Al Hoceima (Morocco) and Gaziantep (Turkey). The approach combined very high-resolution (VHR) satellite imagery with AI-based classification models, allowing consistent detection across contrasting urban and seismic settings. Key methodological steps are illustrated in Figure 4.\u003c/p\u003e\n\u003cp\u003eQuickBird multispectral imagery from April 2004 was utilised for Al Hoceima, while Airbus RGB imagery from 2023 was used for Gaziantep. Both datasets provided sub-meter to 2-meter resolution, which supported urban-scale damage delineation. Ancillary geospatial layers, including building footprints, cadastral maps, and road networks, were integrated to enhance spatial coherence.\u003c/p\u003e\n\u003cp\u003eGround truth data were established using field reports, municipal documentation, and manual digitisation of imagery. Training samples captured five thematic classes: intact structures, moderate and complete collapses, vegetation, and open spaces.\u003c/p\u003e\n\u003cp\u003eMultiple AI classifiers were tested. Random Forest served as a traditional machine learning benchmark. Deep learning models, including U-Net and Mask R-CNN, were selected for their capabilities in semantic segmentation and object-level classification.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.1 Random Forest (RF)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Random Forest algorithm, consisting of an ensemble of decision trees, was implemented using Python\u0026rsquo;s Scikit-learn library. Its robustness in handling high-dimensional multispectral data and its reduced susceptibility to overfitting made it a suitable baseline. Key hyperparameters such as the number of trees, maximum depth, and minimum samples per leaf were tuned through grid search validation (Zourarah et al. 2005).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.2 U-Net Deep Learning Model\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eU-Net, a convolutional neural network designed for semantic segmentation, was deployed for pixel-level classification of damage zones. Input layers included RGB bands and spectral indices. The architecture comprised an encoder-decoder structure with skip connections that improved localisation. The model was implemented using Keras and TensorFlow frameworks and trained using Google Colab GPU instances for 200 epochs.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.3 Mask R-CNN\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMask R-CNN was applied to detect and delineate individual damaged buildings. Its two-stage framework, Region Proposal Network (RPN) and classification enabled precise object extraction. Model performance was evaluated using confusion matrices and the following metrics: overall accuracy, precision, recall, F1-score, and Kappa coefficient. Ground truth polygons were used for testing. Accuracy results are summarised in Tables 1-6.\u003c/p\u003e\n\u003cp\u003eTable 1: Confusion matrix for the SVM model\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" align=\"\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"9\" valign=\"top\" style=\"width: 618px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResult of KNN Classification\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"8\" valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eReference\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUndamaged building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTree\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBare soil\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRoad\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal reference\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e18\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e13\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e40\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUndamaged building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e40\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e41\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTree\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e14\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e17\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e12\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBare soil\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e62\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e70\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRoad\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e12\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e21\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal classified\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e23\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e47\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e18\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 53px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e85\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e18\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e201\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePr\u0026eacute;cision\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e75,879397\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"7\" rowspan=\"2\" valign=\"top\" style=\"width: 437px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 72px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eKappa\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 109px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0,68261563\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 2:Confusion matrix for the KNN model\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" align=\"\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"9\" valign=\"top\" style=\"width: 606px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResult of the SVM classification\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"8\" valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eReference\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eDestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eUndestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eTree\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 49px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBare soil\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eRoad\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eTotal reference\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e15\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e11\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 49px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e41\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUndestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e25\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 49px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e34\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTree\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e13\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 49px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e17\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 49px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e14\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBare soil\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 49px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e58\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e72\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRoad\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 49px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e8\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e13\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e23\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal classified\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e25\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e39\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e18\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 49px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e18\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e76\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e25\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e200\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePrecision\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e65,60846561\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"7\" rowspan=\"2\" valign=\"top\" style=\"width: 426px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eKappa\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 108px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0,554051111\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 3: Confusion matrix for the Decision Tree (DT) model\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"10\" valign=\"top\" style=\"width: 593px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResult of the DT classification\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"8\" valign=\"top\" style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eR\u0026eacute;f\u0026eacute;rence\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUndamaged building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTree\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBare soil\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRoad\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal reference\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 0px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e26\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e39\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 0px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUndamaged building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e40\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e42\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 0px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTree\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e14\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e17\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 0px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e12\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 0px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBare soil\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e59\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e70\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 0px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRoad\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e19\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e21\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 0px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal classified\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e26\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e47\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e16\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e13\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e69\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e30\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 65px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e200\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 0px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 69px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePr\u0026eacute;cision\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e83,2335025\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"8\" rowspan=\"2\" valign=\"top\" style=\"width: 419px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 69px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eKappa\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0,79143899\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 4:Confusion matrix for the Random Forest (RF) model\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"9\" valign=\"top\" style=\"width: 594px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResult of the RF classification\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"8\" valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eReference\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eDestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eUndestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eTree\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBare soil\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eRoad\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal reference\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e15\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e16\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e39\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUndestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e40\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e40\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTree\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e11\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e16\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e13\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBare soil\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e66\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e70\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRoad\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e11\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e22\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal classified\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e15\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e49\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e16\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e9\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e98\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e13\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e200\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePr\u0026eacute;cision\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e74,3718593\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"7\" rowspan=\"2\" valign=\"top\" style=\"width: 420px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eKappa\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0,655989424\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 5: Confusion matrix for the Naive Bayes model\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"615\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"9\" valign=\"top\" style=\"width: 615px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResult of the Bayes classification\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"8\" valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eReference\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUndamaged building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTree\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 51px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBare soil\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRoad\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal reference\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e27\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 51px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e39\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUndamaged building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e37\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 51px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e41\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTree\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e11\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 51px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e16\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 51px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e13\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBare soil\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 51px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e48\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e69\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRoad\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 51px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e13\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e22\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal classified\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 80px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e43\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e40\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e15\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 51px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e18\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 44px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e57\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 45px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e27\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 63px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e200\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePrecision\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e72,5888325\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"7\" rowspan=\"2\" valign=\"top\" style=\"width: 434px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eKappa\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0,65341761\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable 6:Confusion matrix for the SharkRF model\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"9\" valign=\"top\" style=\"width: 612px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResult of the SHARKRF classification\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"8\" valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eReference\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eDestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eUndestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eTree\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBare soil\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eRoad\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eTotal reference\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e14\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e15\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e39\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUndestroyed building\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e33\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e41\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTree\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e11\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e16\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eField\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e12\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBare soil\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e64\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e70\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRoad\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e11\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e22\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal classified\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 77px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e14\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 105px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e48\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 41px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e19\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 48px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e104\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 43px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e12\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e200\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePr\u0026eacute;cision\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e64,97461929\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"7\" rowspan=\"2\" valign=\"top\" style=\"width: 435px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 71px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eKappa\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 106px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0,527265772\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003ePost-processing was executed in QGIS to remove classification noise and ensure topological integrity. Damage density hotspots were derived via kernel density estimation in Gaziantep. Model accuracy was assessed through metrics such as precision, recall, F1-score, and Kappa coefficient using reserved ground truth polygons.\u003c/p\u003e\n\u003cp\u003eThis dual-case design enabled a cross-contextual evaluation of model performance under varied urban geometries and seismic impact levels.\u003c/p\u003e"},{"header":"4. Results and Discussion","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e4.1 Evaluation Framework and Interpretation Strategy\u003c/h2\u003e\u003cp\u003ePost-earthquake damage detection models were tested in Al Hoceima and Gaziantep using very high-resolution imagery and artificial intelligence (AI). The evaluation employed a binary classification scheme, \u0026ldquo;Destruction\u0026rdquo; vs. \u0026ldquo;No Destruction\u0026rdquo;, to support operational crisis response. Performance metrics included overall accuracy and Cohen\u0026rsquo;s kappa coefficient. An 80/20 training-validation split ensured model generalizability.\u003c/p\u003e\u003cp\u003eConfusion matrices revealed trends in misclassification, particularly false positives in nondamaged areas and false negatives in undetected damaged areas. False positives for non-damaged areas misclassified as damaged and false negatives for undetected actual damage were particularly critical in boundary zones between collapsed structures and open land (Labiak \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). These diagnostic patterns guided model refinement.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e4.2 Traditional Classifiers: MD and MLH\u003c/h2\u003e\u003cp\u003eMinimum Distance (MD) and Maximum Likelihood (MLH) were initially tested for baseline comparison. Both rely on pixel-level spectral similarity but ignore spatial context and object geometry, rendering them inadequate in post-earthquake debris environments (Rossi et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eVisual assessments (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) showed substantial confusion between rubble and bare surfaces. MD achieved 47.36% accuracy with a kappa of 0.48, while MLH scored 46.25% and 0.42, respectively. These values suggest minimal reliability for field deployment.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eSpectral confusion, radiometric variability, and lack of spatial awareness accounted for their poor performance. Their operational limitations make them unsuitable for urban seismic mapping under crisis conditions.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e4.3 Machine Learning Model Comparison\u003c/h2\u003e\u003cp\u003eMachine learning (ML) models showed improved but varied performance. Results are detailed in Tables\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e to \u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e.\u003c/p\u003e\u003cp\u003eSVM (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) achieved 65.6% accuracy and a kappa of 0.55\u003c/p\u003e\u003cp\u003eKNN (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) improved to 75.8% and 0.68\u003c/p\u003e\u003cp\u003eDecision Tree (DT) (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) delivered the highest ML scores: 83.2% accuracy, 0.79 kappa.\u003c/p\u003e\u003cp\u003eRandom Forest (RF) (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) followed at 74.3% and 0.65\u003c/p\u003e\u003cp\u003eNaive Bayes and SharkRF (Tables\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e and \u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) showed lower reliability\u003c/p\u003e\u003cp\u003eWhile DT showed strength in handling complex boundaries, performance dropped in heterogeneous urban zones. ML models lacked the spatial granularity required to differentiate structural forms in high-density neighbourhoods. This limited their applicability for emergency damage segmentation. (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e4.4 Deep Learning Results: Mask-RCNN\u003c/h2\u003e\u003cp\u003eMask-RCNN outperformed all other models. As a region-based convolutional neural network, it captured both spectral and spatial patterns, detecting object boundaries with high precision. In Gaziantep, accuracy reached 90.48% with a kappa of 0.93. Al Hoceima returned 90.2% accuracy and 0.91 kappa (Figs.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003e and \u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e13\u003c/span\u003e), indicating excellent agreement with ground truth (Alafandy et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eSuccess stemmed from the model\u0026rsquo;s object-based segmentation, spatial coherence, and architectural capacity to learn structural variation. These capabilities proved vital in recognising collapsed buildings amidst debris and overlapping features.\u003c/p\u003e\u003cp\u003eLimitations included the requirement for annotated training data and vulnerability to atmospheric inconsistencies. Despite this, Mask-RCNN offered rapid post-training deployment, ideal for operational crisis mapping (AlAfandy et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e4.5 Comparative Insights and Operational Implications\u003c/h2\u003e\u003cp\u003eA cross-model comparison confirmed the superiority of Mask-RCNN. Traditional models underperformed significantly. Compared with advanced machine learning algorithms such as DT and RF, or the deep learning model Mask-RCNN, traditional methods like MD and MLH are outperformed. While modern techniques reached accuracy levels above 80% and kappa values near 0.8, traditional approaches barely exceeded the 45% mark, with kappa coefficients under 0.5 (Alloghani et al. 2020). Only Mask-RCNN produced maps with sufficient fidelity for emergency planning (Tables\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e13\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThese differences are not only reflected in metrics but also the visual quality of the classification outputs. As seen by comparing Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003e, which present the results for Al Hoceima and Turkey, deep learning delivers cleaner boundaries, reduced noise, and a closer match to the true spatial distribution of damage. Although MD and MLH require lower computational resources and are easier to implement, which may justify their use in resource-constrained contexts, their limitations in damage detection must be carefully considered before deployment in crisis scenarios.\u003c/p\u003e\u003cp\u003eDeep learning approaches reduced processing time and subjective interpretation errors. Their real-time application capacity enhances field decision-making. Reliance on high-quality imagery and training data may constrain use in data-scarce environments (Saah et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eHybrid frameworks combining open-source AI, cloud computing, and auxiliary datasets are recommended. Image fusion and AI-based super-resolution could mitigate reliance on costly VHR data (Shafique et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e"},{"header":"5 Conclusion","content":"\u003cp\u003eAccurate identification of seismic damage has been recognised as essential for effective disaster response and recovery. Timely and spatially precise assessments are required so that emergency interventions can be prioritised and resources can be allocated where most needed. In the absence of reliable geospatial intelligence, critical delays in crisis response are often observed.\u003c/p\u003e\u003cp\u003eThe integration of very high-resolution (VHR) satellite imagery with artificial intelligence methods has been validated as a powerful approach to support post-earthquake damage mapping. By enabling detailed observation across large geographic extents, satellite-based remote sensing provides a scalable alternative where ground-based surveys are impractical or unsafe. This study confirmed that advanced object-based deep learning models, particularly Mask-RCNN, consistently outperformed traditional pixel-based classifiers in detecting damaged structures within complex urban environments.\u003c/p\u003e\u003cp\u003eThe cases of Al Hoceima and Gaziantep were selected due to their tectonic relevance and diverse urban characteristics. Through their analysis, the framework demonstrated strong model transferability and high classification accuracy, supported by kappa coefficients exceeding 0.9. These results emphasised the importance of integrating spectral and spatial information, especially where debris patterns and structural collapse are spatially heterogeneous.\u003c/p\u003e\u003cp\u003eThe study concludes that AI-driven methodologies provide significant advantages in terms of operational speed, reproducibility, and the reduction of human error. As satellite access improves and AI tools become increasingly open-source, future applications will benefit from enhanced model generalizability. Strengthening collaboration between scientific, governmental, and humanitarian sectors will remain vital in transforming these technological capabilities into real-world solutions for seismic resilience and disaster risk reduction.\u003c/p\u003e"},{"header":"6 Future Directions","content":"\u003cp\u003eSeveral future research directions are proposed. The development of open and collaborative datasets is strongly recommended to improve the generalizability and robustness of deep learning models (Alloghani et al. 2020). Hybrid approaches should be explored, integrating deep learning with complementary data such as socioeconomic or environmental variables, to enhance accuracy and operational relevance in real-world scenarios (Shinde \u0026amp; Shah 2018). It is also advised to investigate methodologies that reduce reliance on VHR satellite images through the fusion of multiple satellite data sources or by implementing advanced AI techniques capable of enhancing the effective resolution of available imagery (Shravantandale 2023)\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003e\u003cstrong\u003eAI\u0026nbsp;\u003c/strong\u003eArtificial Intelligence\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDL\u0026nbsp;\u003c/strong\u003eDeep Learning\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNDVI\u0026nbsp;\u003c/strong\u003eNormalized Difference Vegetation Index\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNDBI\u0026nbsp;\u003c/strong\u003eNaturalistic Developmental Behavioral Interventions\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eVHR\u0026nbsp;\u003c/strong\u003eVery Highresolution\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRF\u0026nbsp;\u003c/strong\u003eRandom Forest\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRPN\u0026nbsp;\u003c/strong\u003eRegion Proposal Network\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMD\u0026nbsp;\u003c/strong\u003eMinimum Distance\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMLH\u0026nbsp;\u003c/strong\u003eMaximum Likelihood\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eML\u0026nbsp;\u003c/strong\u003eMachine learning\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDT\u0026nbsp;\u003c/strong\u003eDecision Tree\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRCNN\u0026nbsp;\u003c/strong\u003eRegion-based Convolutional Neural Network\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSVM\u0026nbsp;\u003c/strong\u003eSupport Vector Machine\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003eCompeting Interests\u003c/h2\u003e\u003cp\u003eNo conflicts of interest (student/supervisor relationship)\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eNot applicable (no external funding).\u003c/p\u003e\u003ch2\u003eAuthor Contributions\u003c/h2\u003e\u003cp\u003eAll authors contributed to the study conception and design. Conceptualization, methodology, data acquisition and processing, writing \u0026ndash; original draft, visualization by Wardia Boutalkhoukhte. Supervision, validation, methodology, writing \u0026ndash; review \u0026amp; editing, critical revision by Ahmed Algouti. All authors read and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgments\u003c/h2\u003e\u003cp\u003eWe thank colleagues at the Faculty of Sciences Semlalia, Cadi Ayyad University, for constructive discussions and logistical support, and the agencies that provided satellite imagery and ancillary geospatial datasets used in this study.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eAvailable upon request from the corresponding author.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAlAfandy, K.A., Hicham, H., Lazaar, M. and Achhab, M.A. (2020). Investment of Classic Deep CNNs and SVM for Classifying Remote Sensing Images. \u003cem\u003eAdvances in Science, Technology and Engineering Systems Journal\u003c/em\u003e, 5(5), pp.652\u0026ndash;659. doi:https://doi.org/10.25046/aj050580.\u003c/li\u003e\n\u003cli\u003eARAB Oussama (2021). CONTRIBUTION UTILISANT L\u0026rsquo;ATTENUATION DES ONDES DE CODA ET LA LIQUEFACTION DES SOL DANS LA REGION DU RIF. \u003cem\u003eImist.ma\u003c/em\u003e. doi:http://toubkal.imist.ma/handle/123456789/33470.\u003c/li\u003e\n\u003cli\u003eChaudhary, M.T. and Piracha, A. (2021). Natural Disasters\u0026mdash;Origins, Impacts, Management. \u003cem\u003eEncyclopedia\u003c/em\u003e, [online] 1(4), pp.1101\u0026ndash;1131. Available at: https://www.mdpi.com/2673-8392/1/4/84 \u003c/li\u003e\n\u003cli\u003eKarls, E. and Braun, A. (2019). \u003cem\u003eRadar satellite imagery for humanitarian response Bridging the gap between technology and application Dissertation der Mathematisch-Naturwissenschaftlichen Fakult\u0026auml;t\u003c/em\u003e. Available at: https://publikationen.uni-tuebingen.de/xmlui/bitstream/handle/10900/91317/Braun2019%20Radar%20satellite%20imagery%20for%20humanitarian%20response%20UB.pdf?sequence=1 \u003c/li\u003e\n\u003cli\u003eLabiak, R. (2025). \u003cem\u003eA Method for detection and quantification of building damage using post-disaster LiDAR data\u003c/em\u003e. RIT Digital Institutional Repository. Available at: https://repository.rit.edu/theses/3023/ \u003c/li\u003e\n\u003cli\u003eOzturk, M., Arslan, M.H. and Korkmaz, H.H. (2023). Effect on RC buildings of 6 February 2023 Turkey earthquake doublets and new doctrines for seismic design. \u003cem\u003eEngineering Failure Analysis\u003c/em\u003e, 153, p.107521. doi:https://doi.org/10.1016/j.engfailanal.2023.107521.\u003c/li\u003e\n\u003cli\u003eShafique, A., Cao, G., Khan, Z., Asad, M. and Aslam, M. (2022). Deep Learning-Based Change Detection in Remote Sensing Images: A Review. \u003cem\u003eRemote Sensing\u003c/em\u003e, 14(4), p.871. doi:https://doi.org/10.3390/rs14040871.\u003c/li\u003e\n\u003cli\u003eSubash Ghimire (2023). Data-driven-based models for seismic damage assessment of buildings : using machine learning at a regional scale and in-situ structural monitoring at building scale. \u003cem\u003eHal.science\u003c/em\u003e. doi:https://theses.hal.science/tel-04370250.\u003c/li\u003e\n\u003cli\u003evan der Woerd, J., Dorbath, C., Ousadou, F., Dorbath, L., Delouis, B., Jacques, E., Tapponnier, P., Hahou, Y., Menzhi, M., Frogneux, M. and Haessler, H. (2014). The Al Hoceima Mw 6.4 earthquake of 24 February 2004 and its aftershocks sequence. \u003cem\u003eJournal of Geodynamics\u003c/em\u003e, 77, pp.89\u0026ndash;109. doi:https://doi.org/10.1016/j.jog.2013.12.004.\u003c/li\u003e\n\u003cli\u003eRossi, M., Minicozzi, G., Pascarella, G. and Capasso, A. (2020). ESG, Competitive Advantage and Financial performances: a Preliminary Research. Handle.net, 1(1), pp.969\u0026ndash;986.\u003c/li\u003e\n\u003cli\u003eSaah, D., Tenneson, K., Matin, M., Uddin, K., Cutter, P., Poortinga, A., Nguyen, Q.H., Patterson, M., Johnson, G., Markert, K., Flores, A., Anderson, E., Weigel, A., Ellenberg, W.L., Bhargava, R., Aekakkararungroj, A., Bhandari, B., Khanal, N., Housman, I.W. and Potapov, P. (2019). Land Cover Mapping in Data Scarce Environments: Challenges and Opportunities. Frontiers in Environmental Science, 7. doi:https://doi.org/10.3389/fenvs.2019.00150.\u003c/li\u003e\n\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":"Earthquake, Remote Sensing, Deep Learning, Classification, Satellite Imagery, Artificial Intelligence Algorithms","lastPublishedDoi":"10.21203/rs.3.rs-7769060/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7769060/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSeismic events continue to challenge the resilience of urban systems, particularly in regions marked by tectonic complexity and high population density. In such contexts, timely and accurate damage assessment remains critical for effective emergency response. This study explores the potential of very high-resolution (VHR) satellite imagery integrated with artificial intelligence (AI) to enhance the spatial understanding of post-earthquake structural impacts. By focusing on two distinct seismic events, the 2004 Al Hoceima earthquake in Morocco and the 2023 Gaziantep earthquake in Turkey. The research underscores the utility of AI-supported remote sensing in overcoming the limitations of conventional ground-based methods. Emphasis is placed on the capacity of VHR imagery to capture subtle spatial variations that are often overlooked during manual assessments. The findings highlight the potential of integrating AI with satellite data as a scalable, transferable, and practical solution for disaster risk management. Broader implications are drawn for future applications in real-time decision making, urban resilience planning, and the development of data-driven humanitarian strategies.\u003c/p\u003e","manuscriptTitle":"Application of Artificial Intelligence Algorithms for Earthquake Damage Mapping Using Very High-Resolution Satellite Imagery: Case Studies from Al Hoceima, Morocco, and Gaziantep, Turkey","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-26 01:15:58","doi":"10.21203/rs.3.rs-7769060/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":"4f6fcc75-24cb-449c-9420-eba8b05e2182","owner":[],"postedDate":"October 26th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-01-27T22:28:18+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-26 01:15:58","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7769060","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7769060","identity":"rs-7769060","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}