[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83048-en":3,"doc-seo-83048-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},83048,13056703019404,"Miles","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Enhanced Seam Segmentation for Automated Welding Robot in Construction Through Transfer Learning","Reliable seam segmentation is essential for autonomous robotic welding in construction, where harsh illumination, specular reflections, and thin weld geometries degrade segmentation quality. This study proposes a reflection-robust seam segmentation framework that enhances a BiSeNetV2 backbone using transfer learning and a hybrid Cross-Entropy–Lovász loss. Without increasing architectural complexity, the method improves reflection robustness via learning-stability-oriented optimization. Results reach 81.76% Joint IoU and 90.73% mIoU, raising Joint IoU by +22.36 points over an OHEM baseline while keeping FLOPs, parameters, and inference speed unchanged.","arXiv :2607 .06 150v 1 [ cs .CV] 7 Jul 2026  \nEnhanced Seam Segmentation for Automated Welding Robot in Construction Through Transfer Learning: Addressing Limitations of Bilateral Segmentation Network  \nKeonvin Park 1 , Yong Ann Voeurn2 , Hyeokjun Kweon3,*,+ , and Doyun Lee2,*,+  \n1 Interdisciplinary Program in Artificial Intelligence, Seoul National University, Seoul, 08826, Republic of Korea  \n2 Department of Civil Engineering and Construction, Georgia Southern University, Statesboro, GA 30460, USA  \n3The Graduate School of Advanced Imaging Science, Multimedia & Film, Chung-Ang University, Seoul, 06974, Republic of Korea  \n* Co-corresponding authors: [hyeokjunkweon@cau.ac.kr](hyeokjunkweon@cau.ac.kr),[doyunlee@georgiasouthern.edu](doyunlee@georgiasouthern.edu)[ ](doyunlee@georgiasouthern.edu)+these authors contributed equally to this work  \nABSTRACT  \nReliable seam segmentation is essential for autonomous robotic welding in construction, where harsh illumination, specular reflections, and thin weld geometries often degrade segmentation performance. This study proposes a reflection-robust seam segmentation framework that enhances a BiSeNetV2 backbone through transfer learning and a hybrid Cross-Entropy–Lovász loss. Rather than increasing architectural complexity, the proposed framework improves reflection robustness through learningstability-oriented optimization. Experimental results show that the proposed method achieves 81 .76% Joint IoU and 90.73% mIoU, improving Joint IoU by +22.36 percentage points over the OHEM-based baseline while maintaining identical FLOPs, parameter count, and inference speed. The proposed approach also recovers 96.33% of severe zero-IoU failure cases under reflective conditions. Comparative experiments across BiSeNetV2, DeepLabV3+, UNet, and SegFormer further demonstrate that the proposed optimization strategy is particularly effective for lightweight real-time segmentation architectures. Qualitative analyses additionally show improved seam continuity and reflection robustness in challenging welding environments. These findings suggest that the proposed framework provides a practical and lightweight perception solution for robotic welding applications involving reflective metallic surfaces.  \nIntroduction  \nAutomated Robotic Welding in Construction  \nWelding remains a fundamental process in structural steel fabrication, prefabrication, and on-site assembly. However, the construction industry increasingly faces shortages of skilled welders, rising labor costs, and growing demands for safety and quality consistency 1–3. In response, robotic welding systems have attracted significant attention as a practical solution for improving productivity, reducing human exposure to hazardous environments, and enabling more consistent weld quality4. Recent advances in computer vision and deep learning have accelerated the development of vision-based robotic welding systems capable of operating in unstructured construction environments5, 6. Semantic segmentation models can directly identify weld seam regions from images, enabling downstream robotic processes such as seam tracking, waypoint generation, and robotic trajectory planning7–9. In particular, lightweight real-time architectures such as BiSeNetV2 provide an effective balance between segmentation accuracy and inference efficiency, making them attractive for edge-deployed robotic platforms 10. Despite these advances, robust weld seam segmentation in construction environments remains challenging. Metallic surfaces frequently produce strong specular reflections, saturated highlights, and illumination artifacts that obscure thin seam boundaries and degrade segmentation reliability 11–14. These issues often lead to fragmented seam predictions, discontinuous masks, and unstable centerline extraction, which may propagate into unreliable robotic trajectories during downstream welding operations8, 9, 15, 16.  \nLimitations of Existing Welding Seam Segmentation Appro","cbCaibMFrGlQMqH1","https://ap.wps.com/l/cbCaibMFrGlQMqH1","pdf",2365486,1,19,"English","en",105,"# Abstract\n# Introduction\n## Automated Robotic Welding in Construction\n## Limitations of Existing Welding Seam Segmentation Approaches\n# Objectives and Contributions","[{\"question\":\"为什么施工场景中的焊缝分割更难？\",\"answer\":\"施工环境的强照明、镜面反射以及焊缝几何较薄会遮蔽焊缝边界，降低分割可靠性，导致焊缝预测破碎和中心线提取不稳定。\"},{\"question\":\"本文提出了什么方法来增强镜面反射鲁棒性？\",\"answer\":\"通过在BiSeNetV2上引入迁移学习，并采用混合的Cross-Entropy–Lovász损失，结合面向学习稳定性的优化来提升对反射的鲁棒性。\"},{\"question\":\"该方法相对基线提升了哪些指标，且是否影响实时性？\",\"answer\":\"实验显示Joint IoU达到81.76%、mIoU达到90.73%，Joint IoU相对OHEM基线提升+22.36个百分点；同时保持相同FLOPs、参数量与推理速度。\"}]",1784184859,48,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"enhanced-seam-segmentation-for-automated-welding-robot-in-construction-through-transfer-learning","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/enhanced-seam-segmentation-for-automated-welding-robot-in-construction-through-transfer-learning/83048/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"为什么施工场景中的焊缝分割更难？","Question",{"text":75,"@type":76},"施工环境的强照明、镜面反射以及焊缝几何较薄会遮蔽焊缝边界，降低分割可靠性，导致焊缝预测破碎和中心线提取不稳定。","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"本文提出了什么方法来增强镜面反射鲁棒性？",{"text":80,"@type":76},"通过在BiSeNetV2上引入迁移学习，并采用混合的Cross-Entropy–Lovász损失，结合面向学习稳定性的优化来提升对反射的鲁棒性。",{"name":82,"@type":73,"acceptedAnswer":83},"该方法相对基线提升了哪些指标，且是否影响实时性？",{"text":84,"@type":76},"实验显示Joint IoU达到81.76%、mIoU达到90.73%，Joint IoU相对OHEM基线提升+22.36个百分点；同时保持相同FLOPs、参数量与推理速度。","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},"General","general"]