SCAD: A Siamese Cross-Attention Discrimination Network for Bitemporal Building Change Detection
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 ran...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/24/6213 |
| _version_ | 1797455474455478272 |
|---|---|
| author | Chuan Xu Zhaoyi Ye Liye Mei Sen Shen Qi Zhang Haigang Sui Wei Yang Shaohua Sun |
| author_facet | Chuan Xu Zhaoyi Ye Liye Mei Sen Shen Qi Zhang Haigang Sui Wei Yang Shaohua Sun |
| author_sort | Chuan Xu |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-03-09T15:54:52Z |
| format | Article |
| id | doaj.art-d4d447c8411d4883870ee1a0a9b1125b |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T15:54:52Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-d4d447c8411d4883870ee1a0a9b1125b2023-11-24T17:45:59ZengMDPI AGRemote Sensing2072-42922022-12-011424621310.3390/rs14246213SCAD: A Siamese Cross-Attention Discrimination Network for Bitemporal Building Change DetectionChuan Xu0Zhaoyi Ye1Liye Mei2Sen Shen3Qi Zhang4Haigang Sui5Wei Yang6Shaohua Sun7School of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaSchool of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaSchool of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaSchool of Weapon Engineering, Naval Engineering University, Wuhan 430032, ChinaSchool of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430072, ChinaSchool of Information Science and Engineering, Wuchang Shouyi University, Wuhan 430064, ChinaAir Force Research Academy, Beijing 10085, ChinaBuilding 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.https://www.mdpi.com/2072-4292/14/24/6213building change detectiondeep learningSiamese cross-attentionfeature fusiondifferential context |
| spellingShingle | Chuan Xu Zhaoyi Ye Liye Mei Sen Shen Qi Zhang Haigang Sui Wei Yang Shaohua Sun SCAD: A Siamese Cross-Attention Discrimination Network for Bitemporal Building Change Detection Remote Sensing building change detection deep learning Siamese cross-attention feature fusion differential context |
| title | SCAD: A Siamese Cross-Attention Discrimination Network for Bitemporal Building Change Detection |
| title_full | SCAD: A Siamese Cross-Attention Discrimination Network for Bitemporal Building Change Detection |
| title_fullStr | SCAD: A Siamese Cross-Attention Discrimination Network for Bitemporal Building Change Detection |
| title_full_unstemmed | SCAD: A Siamese Cross-Attention Discrimination Network for Bitemporal Building Change Detection |
| title_short | SCAD: A Siamese Cross-Attention Discrimination Network for Bitemporal Building Change Detection |
| title_sort | scad a siamese cross attention discrimination network for bitemporal building change detection |
| topic | building change detection deep learning Siamese cross-attention feature fusion differential context |
| url | https://www.mdpi.com/2072-4292/14/24/6213 |
| work_keys_str_mv | AT chuanxu scadasiamesecrossattentiondiscriminationnetworkforbitemporalbuildingchangedetection AT zhaoyiye scadasiamesecrossattentiondiscriminationnetworkforbitemporalbuildingchangedetection AT liyemei scadasiamesecrossattentiondiscriminationnetworkforbitemporalbuildingchangedetection AT senshen scadasiamesecrossattentiondiscriminationnetworkforbitemporalbuildingchangedetection AT qizhang scadasiamesecrossattentiondiscriminationnetworkforbitemporalbuildingchangedetection AT haigangsui scadasiamesecrossattentiondiscriminationnetworkforbitemporalbuildingchangedetection AT weiyang scadasiamesecrossattentiondiscriminationnetworkforbitemporalbuildingchangedetection AT shaohuasun scadasiamesecrossattentiondiscriminationnetworkforbitemporalbuildingchangedetection |