SCAD: A Siamese Cross-Attention Discrimination Network for Bitemporal Building Change Detection

被引:6
|
作者
Xu, Chuan [1 ]
Ye, Zhaoyi [1 ]
Mei, Liye [1 ,2 ]
Shen, Sen [3 ]
Zhang, Qi [1 ]
Sui, Haigang [4 ]
Yang, Wei [5 ]
Sun, Shaohua [6 ]
机构
[1] Hubei Univ Technol, Sch Comp Sci, Wuhan 430068, Peoples R China
[2] Wuhan Univ, Inst Technol Sci, Wuhan 430072, Peoples R China
[3] Naval Engn Univ, Sch Weap Engn, Wuhan 430032, Peoples R China
[4] Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & Re, Wuhan 430072, Peoples R China
[5] Wuchang Shouyi Univ, Sch Informat Sci & Engn, Wuhan 430064, Peoples R China
[6] Air Force Res Acad, Beijing 10085, Peoples R China
基金
中国国家自然科学基金;
关键词
building change detection; deep learning; Siamese cross-attention; feature fusion; differential context; COVER CHANGE DETECTION; IMAGE;
D O I
10.3390/rs14246213
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Building change detection (BCD) is crucial for urban construction and planning. The powerful discriminative ability of deep convolutions in deep learning-based BCD methods has considerably increased the accuracy and efficiency. However, dense and continuously distributed buildings contain a wide range of multi-scale features, which render current deep learning methods incapable of discriminating and incorporating multiple features effectively. In this work, we propose a Siamese cross-attention discrimination network (SCADNet) to identify complex information in bitemporal images and improve the change detection accuracy. Specifically, we first use the Siamese cross-attention (SCA) module to learn unchanged and changed feature information, combining multi-head cross-attention to improve the global validity of high-level semantic information. Second, we adapt a multi-scale feature fusion (MFF) module to integrate embedded tokens with context-rich channel transformer outputs. Then, upsampling is performed to fuse the extracted multi-scale information content to recover the original image information to the maximum extent. For information content with a large difference in contextual semantics, we perform filtering using a differential context discrimination (DCD) module, which can help the network to avoid pseudo-change occurrences. The experimental results show that the present SCADNet is able to achieve a significant change detection performance in terms of three public BCD datasets (LEVIR-CD, SYSU-CD, and WHU-CD). For these three datasets, we obtain F1 scores of 90.32%, 81.79%, and 88.62%, as well as OA values of 97.98%, 91.23%, and 98.88%, respectively.
引用
收藏
页数:21
相关论文
共 50 条
  • [21] A Siamese Multiscale Attention Decoding Network for Building Change Detection on High-Resolution Remote Sensing Images
    Chen, Yao
    Zhang, Jindou
    Shao, Zhenfeng
    Huang, Xiao
    Ding, Qing
    Li, Xianyi
    Huang, Youju
    REMOTE SENSING, 2023, 15 (21)
  • [22] Deception Detection System with Joint Cross-Attention
    Jiang, Peili
    Wang, Yunfan
    Li, Jiajun
    Wang, Ziyang
    PROCEEDINGS OF THE 2024 6TH INTERNATIONAL SYMPOSIUM ON SIGNAL PROCESSING SYSTEMS, SSPS 2024, 2024, : 40 - 47
  • [23] Tops, Bottoms, and Shoes: Building Capsule Wardrobes via Cross-Attention Tensor Network
    Chen, Huiyuan
    Lin, Yusan
    Wang, Fei
    Yang, Hao
    15TH ACM CONFERENCE ON RECOMMENDER SYSTEMS (RECSYS 2021), 2021, : 453 - 462
  • [24] Sarcasm Detection Base on Adaptive Incongruity Extraction Network and Incongruity Cross-Attention
    He, Yuanlin
    Chen, Mingju
    He, Yingying
    Qu, Zhining
    He, Fanglin
    Yu, Feihong
    Liao, Jun
    Wang, Zhenchuan
    APPLIED SCIENCES-BASEL, 2023, 13 (04):
  • [25] CS-HSNet: A Cross-Siamese Change Detection Network Based on Hierarchical-Split Attention
    Ke, Qingtian
    Zhang, Peng
    IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2021, 14 (14) : 9987 - 10002
  • [26] Cross-Attention Regression Flow for Defect Detection
    Liu, Binhui
    Guo, Tianchu
    Luo, Bin
    Cui, Zhen
    Yang, Jian
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2024, 33 : 5183 - 5193
  • [27] HFA-Net: High frequency attention siamese network for building change detection in VHR remote sensing images
    Zheng, Hanhong
    Gong, Maoguo
    Liu, Tongfei
    Jiang, Fenlong
    Zhan, Tao
    Lu, Di
    Zhang, Mingyang
    PATTERN RECOGNITION, 2022, 129
  • [28] Multiple Attention Siamese Network for High-Resolution Image Change Detection
    Huang, Jiru
    Shen, Qian
    Wang, Min
    Yang, Mengyuan
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60
  • [29] Remote sensing image change detection based on swin transformer and cross-attention mechanism
    Yan, Weidong
    Cao, Li
    Yan, Pei
    Zhu, Chaosheng
    Wang, Mengtian
    EARTH SCIENCE INFORMATICS, 2025, 18 (01)
  • [30] Adaptive Multi-Feature Fusion Visual Target Tracking Based on Siamese Neural Network with Cross-Attention Mechanism
    Zhou, Qian
    Xia, Haoran
    Yan, Hongzheng
    Yang, Ming
    Chen, Shidong
    2022 22ND IEEE/ACM INTERNATIONAL SYMPOSIUM ON CLUSTER, CLOUD AND INTERNET COMPUTING (CCGRID 2022), 2022, : 307 - 316