A Multi-scale Cross Domain Tea Dataset Augmentation Method

A Multi-scale Cross Domain Tea Dataset Augmentation Method | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 29 April 2025 V1 Latest version Share on A Multi-scale Cross Domain Tea Dataset Augmentation Method Authors : Taojie Yu , Jianneng Chen 0000-0002-3816-2692 [email protected] , Zhiyong Gui , Jiangming Jia , Yatao Li , Chennan Yu , and Chuanyu Wu Authors Info & Affiliations https://doi.org/10.22541/au.174593576.69063095/v1 229 views 105 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract To tackle challenges in tea shoot detection caused by lighting variations and cultivar differences, this study introduces Tea CycleGAN, combining multi-scale style transfer and spatial-consistent data augmentation. The framework transforms images between Longjing 43 (daylight) and Zhongcha 108 (night lighting) cultivars using a redesigned generator with SKConv modules for dynamic multi-scale feature fusion and attention-based weight optimization. A deep discriminator enhances texture/hue discrimination through added convolutional layers and batch normalization. The method employs a global-local transfer framework: 600×600-pixel plant backgrounds and 64×64-pixel shoot regions undergo separate style transfers, with a restoration-paste strategy preserving spatial consistency. Results show Tea CycleGAN achieves FID scores of 41.93 (600×600) and 26.49 (64×64), outperforming original CycleGAN by 32.98 and 30.67 reductions. Generated images exhibit superior hue diversity and edge clarity against CycleGAN, DCGAN, and WGAN. In YOLOv7 detection tasks, the method elevates performance from unaugmented data (mAP=73.94%, P=75.25%, R=45.25%) to 83.54% (+9.6), 86.03% (+10.78), and 55.18% (+9.93), surpassing Mosaic (80.13%, 83.52%, 50.92%) and Mixup (78.93%, 81.49%, 48.74%). The approach effectively mitigates lighting/scale impacts in agricultural AI, demonstrating SKConv’s multi-scale fusion and restoration-paste strategies significantly enhance synthetic data quality for tea-picking systems. Key innovations include cultivar-specific domain adaptation and hierarchical style transfer preserving biological spatial relationships, offering a robust solution for phenotypic variability in precision agriculture. Supplementary Material File (manuscript.docx) Download 10.24 MB Information & Authors Information Version history V1 Version 1 29 April 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords agricultural robotics computer vision sample acquisition unsupervised learning vision processing Authors Affiliations Taojie Yu Zhejiang Sci-Tech University School of Mechanical Engineering View all articles by this author Jianneng Chen 0000-0002-3816-2692 [email protected] Zhejiang Sci-Tech University School of Mechanical Engineering View all articles by this author Zhiyong Gui Zhejiang Sci-Tech University School of Mechanical Engineering View all articles by this author Jiangming Jia Zhejiang Sci-Tech University School of Mechanical Engineering View all articles by this author Yatao Li Zhejiang Sci-Tech University School of Mechanical Engineering View all articles by this author Chennan Yu Zhejiang Sci-Tech University School of Mechanical Engineering View all articles by this author Chuanyu Wu Zhejiang Ocean University View all articles by this author Metrics & Citations Metrics Article Usage 229 views 105 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Taojie Yu, Jianneng Chen, Zhiyong Gui, et al. A Multi-scale Cross Domain Tea Dataset Augmentation Method. Authorea . 29 April 2025. DOI: https://doi.org/10.22541/au.174593576.69063095/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.174593576.69063095/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fec7bf1bde806e7',t:'MTc3OTI5MjA5MQ=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();