[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84968-en":3,"doc-seo-84968-105":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},84968,7971461740886,"Theodore","https://ap-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2",8,"Research & Report","ASFR-Net Adversarial Alignment and Spatio-Frequency Refinement Network for Heterogeneous Remote Sensing Image Change Detection","Heterogeneous remote sensing change detection must separate true land-cover alterations from strong modality discrepancies introduced by different sensing mechanisms, which often cause pseudo-changes and limit accuracy. ASFR-Net proposes an end-to-end adversarial spatio-frequency refinement framework. MIRLearner extracts modality-invariant representations to bridge the domain gap, while SFEM suppresses sensor-specific noise via frequency-domain enhancement. Refined multi-level difference features drive a decoder with cascaded hierarchical guided fusion to produce accurate change maps. A new VisNIR-HCD high-resolution benchmark for building changes is released. Experiments show state-of-the-art performance on VisNIR-HCD and public datasets, with open-source code availability.","ASFR-Net: Adversarial Alignment and Spatio-Frequency Refinement Network for Heterogeneous Remote Sensing Image Change Detection  \nXin-Jie Wu, Zhi-Hui You, Si-Bao Chen, Qing-Ling Shu, Xiao Wang, Jin Tang, and Bin Luo  \narXiv :2607 .07 16 1v 1 [ cs .CV] 8 Jul 2026  \nAbstract—The core challenge of heterogeneous change detection in remote sensing imagery lies in effectively decoupling genuine land-cover changes from significant modal disparities caused by distinct imaging mechanisms. These intrinsic inconsistencies are prone to introducing pseudo-changes, thereby constraining detection accuracy. To address this, we propose a novel, endto-end adversarial spatio-frequency refinement network (ASFRNet). Initially, a modality-invariant representation learner (MIRLearner) guides the backbone to extract modality-invariant features, effectively bridging the primary domain gap. Subsequently, to address persistent residual modal differences, we design an innovative spatio-frequency synergistic enhancement module (SFEM), which identifies and suppresses sensor-specific noise and artifacts that are difficult to discern in the spatial domain by leveraging frequency-domain processing. Multi-level difference features are then computed from these refined representationsand fed into a decoder equipped with cascaded hierarchical guided fusion module (HGFM) blocks to generate precise change maps. To alleviate the data scarcity in heterogeneous tasks, we construct and release a new high-resolution benchmark specifically focused on building changes: the visible-nearinfrared heterogeneous change detection (VisNIR-HCD) dataset. It presents unique scientific challenges arising from deceptive visual similarity and non-linear spectral inversions, providing a robust platform for evaluating model generalization. Extensive experiments on VisNIR-HCD and public datasets demonstrate that ASFR-Net achieves state-of-the-art (SOTA) performance, significantly outperforming existing methods. The source code and the VisNIR-HCD dataset are publicly available at [https:](https:)//[github.com/LuoYang2024/ASFR-Net](github.com/LuoYang2024/ASFR-Net).  \nIndex Terms—Change detection (CD), domain adaptation (DA), frequency domain analysis, multimodal, heterogeneous image, remote sensing (RS).  \nI. INTRODUCTION  \nCHANGE DETECTION (CD) stands as a fundamen  \ntal cornerstone in remote sensing image interpretation, dedicated to discerning significant land-cover alterations by analyzing bi-temporal images of the same geographical area acquired at different timestamps [1] . By delivering timely and precise insights into Earth surface dynamics, CD plays an  \nThis work was supported in part by the NSFC Key Project of Joint Fund for Enterprise Innovation and Development under Grant U24A20342 and in part by the National Natural Science Foundation of China under Grant 62576006 and Grant 61976004 . (Xin-Jie Wu and Zhi-Hui You contributed equally to this work.) (Corresponding author: Si-Bao Chen.)  \nXin-Jie Wu, Si-Bao Chen, Qing-Ling Shu, Xiao Wang, Jin Tang, and Bin Luo are with the MOE Key Laboratory of ICSP, IMIS Laboratory of Anhui Province, Anhui Provincial Key Laboratory of Multimodal Cognitive Computation, Zenmorn-AHU AI Joint Laboratory, School of Computer Science and Technology, Anhui University, Hefei 230601, China (e-mail: luoyang [unique@outlook.com](unique@outlook.com); [sbchen@ahu.edu.cn](sbchen@ahu.edu.cn); [2563489133@qq.com](2563489133@qq.com); xi  \n[aowang@ahu.edu.cn](aowang@ahu.edu.cn); [tangjin@ahu.edu.cn](tangjin@ahu.edu.cn); [luobin@ahu.edu.cn](luobin@ahu.edu.cn)).  \nZhi-Hui You is with the School of Public Safety and Emergency Management, Anhui University of Science and Technology, Hefei 231131, China (e-mail: [youzh@aust.edu.cn](youzh@aust.edu.cn)).  \nindispensable role across a spectrum of critical applications, ranging from post-disaster damage assessment [2] and urban expansion monitoring [3] to ecosystem sustainability analysis. In recent years, the explosive pro","cbCail4UnObdtgfu","https://ap.wps.com/l/cbCail4UnObdtgfu","pdf",5560185,1,16,"English","en",105,"# Abstract\n# Index Terms\n# I. Introduction","[{\"question\":\"What makes heterogeneous remote sensing change detection difficult?\",\"answer\":\"It requires decoupling genuine land-cover changes from modality disparities caused by different imaging mechanisms, which can introduce pseudo-changes and degrade detection accuracy.\"},{\"question\":\"How does ASFR-Net address the domain gap between modalities?\",\"answer\":\"It uses a modality-invariant representation learner (MIRLearner) to guide the backbone in extracting modality-invariant features, bridging the primary domain gap.\"},{\"question\":\"What role does the spatio-frequency enhancement module play?\",\"answer\":\"The SFEM identifies and suppresses sensor-specific noise and artifacts that are hard to discern in the spatial domain by leveraging frequency-domain processing.\"}]",1784199816,40,{"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},"asfr-net-adversarial-alignment-and-spatio-frequency-refinement-network-for-heterogeneous-remote-sensing-image-change-detection","",{"@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/asfr-net-adversarial-alignment-and-spatio-frequency-refinement-network-for-heterogeneous-remote-sensing-image-change-detection/84968/",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","2026-07-22","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},"What makes heterogeneous remote sensing change detection difficult?","Question",{"text":75,"@type":76},"It requires decoupling genuine land-cover changes from modality disparities caused by different imaging mechanisms, which can introduce pseudo-changes and degrade detection accuracy.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does ASFR-Net address the domain gap between modalities?",{"text":80,"@type":76},"It uses a modality-invariant representation learner (MIRLearner) to guide the backbone in extracting modality-invariant features, bridging the primary domain gap.",{"name":82,"@type":73,"acceptedAnswer":83},"What role does the spatio-frequency enhancement module play?",{"text":84,"@type":76},"The SFEM identifies and suppresses sensor-specific noise and artifacts that are hard to discern in the spatial domain by leveraging frequency-domain processing.","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},[93,97,101,105,110,115,119,122,127,130,134],{"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":28,"slug":118},7,"Healthcare","healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]