EUVIMEDEUVIMED entdecken
Crossref

A Prototype-Aware Learning and Dual-View Regularization Network for Weakly Supervised Change Detection in VHR Remote Sensing Images

Yongqi Chen, Shou Feng, Chunhui Zhao, Yingjie Tang, Wei Li, Ran Tao

IEEE Transactions on Geoscience and Remote Sensing · 2025 · 63 · 1-15

DOI: 10.1109/tgrs.2025.3608817 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Geoscience and Remote Sensing, 2025. Verfasst von Yongqi Chen und 5 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 65 Literaturangaben. Gefördert von: National Natural Science Foundation of China, China Postdoctoral Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The change detection (CD) is a critical task for monitoring the spatiotemporal evolution of the Earth’s surface. Recently, due to the advantages of reduced annotation cost and improved labeling efficiency, weakly supervised CD (WSCD) has attracted increasing attention. However, existing WSCD methods encounter several critical challenges, including incomplete activation of class activation maps (CAMs), interference from noisy pseudo-labels during training, and instability in change recognition caused by illumination and environmental variations. To address these issues, we propose a prototype-aware learning and dual-view regularization network (PDRNet) for image-level WSCD. Specifically, to address the issue of incomplete activation caused by the tendency of CAM to focus excessively on locally discriminative regions, PDRNet devises a prototype-aware module (PAM), which captures stable category prototypes and refines CAM quality by reactivating hierarchical features. Furthermore, to mitigate the network’s sensitivity to noisy pseudo-labels, a dual-view regularization strategy (DRS) is designed to partition pseudo-labels into clean and noisy regions. Region-specific regularization is subsequently employed to improve the robustness of the model against noisy supervision. Finally, to enhance the capability of identifying changed regions, PDRNet constructs a wavelet-based change enhancement module (WCEM) to decompose bitemporal features into multiple frequency bands. This facilitates the comprehensive utilization of low-frequency structural semantics and high-frequency texture details. Extensive experiments and analyses conducted on three publicly available CD datasets yield the superiority of PDRNet.

Förderung

  • National Natural Science Foundation of China
  • China Postdoctoral Science Foundation

Zitieren

APA

Chen, Y., Feng, S., Zhao, C., Tang, Y., Li, W., & Tao, R. (2025). A Prototype-Aware Learning and Dual-View Regularization Network for Weakly Supervised Change Detection in VHR Remote Sensing Images. IEEE Transactions on Geoscience and Remote Sensing, 63, 1-15. https://doi.org/10.1109/tgrs.2025.3608817

Vancouver

Chen Y, Feng S, Zhao C, Tang Y, Li W, Tao R. A Prototype-Aware Learning and Dual-View Regularization Network for Weakly Supervised Change Detection in VHR Remote Sensing Images. IEEE Transactions on Geoscience and Remote Sensing. 2025;63:1-15. doi:10.1109/tgrs.2025.3608817

BibTeX
@article{Chen2025,
  author = {Chen, Yongqi and Feng, Shou and Zhao, Chunhui and Tang, Yingjie and Li, Wei and Tao, Ran},
  title = {A Prototype-Aware Learning and Dual-View Regularization Network for Weakly Supervised Change Detection in VHR Remote Sensing Images},
  journal = {IEEE Transactions on Geoscience and Remote Sensing},
  year = {2025},
  volume = {63},
  pages = {1--15},
  doi = {10.1109/tgrs.2025.3608817}
}
Kostenlos starten →