Unsupervised Change Detection in Heterogeneous Remote Sensing Images via Dynamic Mask Guidance
Remote Sensing · 29.03.2026 · 18 (7) · 1022
DOI: 10.3390/rs18071022 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 29.03.2026. Verfasst von Paixin Xie und 4 weiteren Autor:innen. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Guangdong Province Natural Science Foundation, project of Southern Marine Science and Engineering Guangdong Laboratory, Zhuhai Basic and Applied Basic Research Foundation.
Abstract (Originalsprache)
Unsupervised change detection (CD) in heterogeneous remote sensing images is intrinsically difficult due to severe sensor-specific discrepancies. In the absence of ground truth, these discrepancies result in ambiguous optimization objectives that make it difficult for models to distinguish true land-cover changes from modality-driven pseudo-changes. To address these challenges, we propose MaskUCD, a novel unsupervised framework that reformulates heterogeneous CD as a dynamic mask-driven constraint scheduling problem. Fundamentally distinct from conventional strategies that enforce selective feature alignment, MaskUCD employs a spatially adaptive optimization mechanism. Specifically, the iteratively refined mask serves as a geometric reference to guide optimization. It enforces strict feature alignment in mask-unchanged regions to suppress modality-induced discrepancies, while simultaneously promoting feature divergence in mask-changed regions to emphasize semantic inconsistencies. In this way, explicit optimization objectives are established, together with an intrinsic interpretability constraint that guides the CD process. This strategy treats the mask as a structural guide for representation learning rather than a ground-truth reference, thereby avoiding error accumulation caused by directly using inaccurate masks as supervisory signals. To facilitate this optimization, we design a specialized asymmetric autoencoder with a hybrid encoder architecture, utilizing multi-scale frequency analysis and global context modeling to enhance feature representation capabilities. Consequently, this design enables the generation of refined and semantically consistent masks, which provide increasingly precise structural guidance, yielding converged and discriminative difference maps. Extensive experiments demonstrate that MaskUCD achieves state-of-the-art performance and superior robustness compared to existing advanced methods.
Förderung
- Guangdong Province Natural Science Foundation
- project of Southern Marine Science and Engineering Guangdong Laboratory
- Zhuhai Basic and Applied Basic Research Foundation
- Major Special Projects of State Key Laboratory under the Jilin Province Science and Technology Development Plan
- Scientific Research Startup Foundation of Guangzhou Institute of Science Technology
- Guangdong Key Discipline Research Capacity Enhancement Project
Zitieren
APA
Xie, P., Chen, G., Zhou, Q., Li, X., & Yan, J. (2026). Unsupervised Change Detection in Heterogeneous Remote Sensing Images via Dynamic Mask Guidance. Remote Sensing, 18(7), 1022. https://doi.org/10.3390/rs18071022
Vancouver
Xie P, Chen G, Zhou Q, Li X, Yan J. Unsupervised Change Detection in Heterogeneous Remote Sensing Images via Dynamic Mask Guidance. Remote Sensing. 2026;18(7):1022. doi:10.3390/rs18071022
BibTeX
@article{Xie2026,
author = {Xie, Paixin and Chen, Gao and Zhou, Qingfeng and Li, Xiaoyan and Yan, Jingwen},
title = {Unsupervised Change Detection in Heterogeneous Remote Sensing Images via Dynamic Mask Guidance},
journal = {Remote Sensing},
year = {2026},
volume = {18},
number = {7},
pages = {1022},
doi = {10.3390/rs18071022}
}
EUVIMED