Scholarlink

<p>Change detection is widely utilized across various domains, such as disaster monitoring, where it aids in identifying differences between images captured at different time intervals. However, current methods often lack constraints on intermediate features and fail to comprehensively model the temporal relationships among these features. Additionally, they rely on simplistic fusion mechanisms, leading to suboptimal network performance. In this paper, we propose: (1) a Feature Information Interaction Module (FIIM) based on spatial attention to enhance semantic information; (2) a Feature Pair Fusion Module (FPFM) with dual-branch structure to model bi-temporal relationships; and (3) a Multi-Scale Supervision Method (MSSM) using contrastive learning to better constrain intermediate features. Comparative experiments conducted on the CDD and LEVIR-CD datasets demonstrate the superiority of our proposed network over existing state-of-the-art methods. Code repository: <a href="https://github.com/joyeuxni/SNIIF-Net">https://github.com/joyeuxni/SNIIF-Net</a>.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Siamese change detection based on information interaction and fusion network

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月10日

查看全文
  • Yanni Zhang,
  • Lei Yang,
  • Caigen Zhou,
  • Jiachen Wen,
  • Licai Zhu

摘要

Change detection is widely utilized across various domains, such as disaster monitoring, where it aids in identifying differences between images captured at different time intervals. However, current methods often lack constraints on intermediate features and fail to comprehensively model the temporal relationships among these features. Additionally, they rely on simplistic fusion mechanisms, leading to suboptimal network performance. In this paper, we propose: (1) a Feature Information Interaction Module (FIIM) based on spatial attention to enhance semantic information; (2) a Feature Pair Fusion Module (FPFM) with dual-branch structure to model bi-temporal relationships; and (3) a Multi-Scale Supervision Method (MSSM) using contrastive learning to better constrain intermediate features. Comparative experiments conducted on the CDD and LEVIR-CD datasets demonstrate the superiority of our proposed network over existing state-of-the-art methods. Code repository: https://github.com/joyeuxni/SNIIF-Net.