DTT-CGINet: A Dual Temporal Transformer Network with Multi-Scale Contour-Guided Graph Interaction for Change Detection
Deep learning has dramatically enhanced remote sensing change detection. However, existing neural network models often face challenges like false positives and missed detections due to factors like lighting changes, scale differences, and noise interruptions. Additionally, change detection results o...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/16/5/844 |
| _version_ | 1797263975075807232 |
|---|---|
| author | Ming Chen Wanshou Jiang Yuan Zhou |
| author_facet | Ming Chen Wanshou Jiang Yuan Zhou |
| author_sort | Ming Chen |
| collection | DOAJ |
| description | Deep learning has dramatically enhanced remote sensing change detection. However, existing neural network models often face challenges like false positives and missed detections due to factors like lighting changes, scale differences, and noise interruptions. Additionally, change detection results often fail to capture target contours accurately. To address these issues, we propose a novel transformer-based hybrid network. In this study, we analyze the structural relationship in bi-temporal images and introduce a cross-attention-based transformer to model this relationship. First, we use a tokenizer to express the high-level features of the bi-temporal image into several semantic tokens. Then, we use a dual temporal transformer (DTT) encoder to capture dense spatiotemporal contextual relationships among the tokens. The features extracted at the coarse scale are refined into finer details through the DTT decoder. Concurrently, we input the backbone’s low-level features into a contour-guided graph interaction module (CGIM) that utilizes joint attention to capture semantic relationships between object regions and the contour. Then, we use the feature pyramid decoder to integrate the multi-scale outputs of the CGIM. The convolutional block attention modules (CBAMs) employ channel and spatial attention to reweight feature maps. Finally, the classifier discriminates change pixels and generates the final change map of the difference feature map. Several experiments have demonstrated that our model shows significant advantages over other methods in terms of efficiency, accuracy, and visual effects. |
| first_indexed | 2024-04-25T00:21:32Z |
| format | Article |
| id | doaj.art-cf1ba009c0f740ed8f26cf9b3c651ff7 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-04-25T00:21:32Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-cf1ba009c0f740ed8f26cf9b3c651ff72024-03-12T16:54:12ZengMDPI AGRemote Sensing2072-42922024-02-0116584410.3390/rs16050844DTT-CGINet: A Dual Temporal Transformer Network with Multi-Scale Contour-Guided Graph Interaction for Change DetectionMing Chen0Wanshou Jiang1Yuan Zhou2State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, ChinaDeep learning has dramatically enhanced remote sensing change detection. However, existing neural network models often face challenges like false positives and missed detections due to factors like lighting changes, scale differences, and noise interruptions. Additionally, change detection results often fail to capture target contours accurately. To address these issues, we propose a novel transformer-based hybrid network. In this study, we analyze the structural relationship in bi-temporal images and introduce a cross-attention-based transformer to model this relationship. First, we use a tokenizer to express the high-level features of the bi-temporal image into several semantic tokens. Then, we use a dual temporal transformer (DTT) encoder to capture dense spatiotemporal contextual relationships among the tokens. The features extracted at the coarse scale are refined into finer details through the DTT decoder. Concurrently, we input the backbone’s low-level features into a contour-guided graph interaction module (CGIM) that utilizes joint attention to capture semantic relationships between object regions and the contour. Then, we use the feature pyramid decoder to integrate the multi-scale outputs of the CGIM. The convolutional block attention modules (CBAMs) employ channel and spatial attention to reweight feature maps. Finally, the classifier discriminates change pixels and generates the final change map of the difference feature map. Several experiments have demonstrated that our model shows significant advantages over other methods in terms of efficiency, accuracy, and visual effects.https://www.mdpi.com/2072-4292/16/5/844change detectiontransformerattentiongraph convolutional network (GCN)remote sensing |
| spellingShingle | Ming Chen Wanshou Jiang Yuan Zhou DTT-CGINet: A Dual Temporal Transformer Network with Multi-Scale Contour-Guided Graph Interaction for Change Detection Remote Sensing change detection transformer attention graph convolutional network (GCN) remote sensing |
| title | DTT-CGINet: A Dual Temporal Transformer Network with Multi-Scale Contour-Guided Graph Interaction for Change Detection |
| title_full | DTT-CGINet: A Dual Temporal Transformer Network with Multi-Scale Contour-Guided Graph Interaction for Change Detection |
| title_fullStr | DTT-CGINet: A Dual Temporal Transformer Network with Multi-Scale Contour-Guided Graph Interaction for Change Detection |
| title_full_unstemmed | DTT-CGINet: A Dual Temporal Transformer Network with Multi-Scale Contour-Guided Graph Interaction for Change Detection |
| title_short | DTT-CGINet: A Dual Temporal Transformer Network with Multi-Scale Contour-Guided Graph Interaction for Change Detection |
| title_sort | dtt cginet a dual temporal transformer network with multi scale contour guided graph interaction for change detection |
| topic | change detection transformer attention graph convolutional network (GCN) remote sensing |
| url | https://www.mdpi.com/2072-4292/16/5/844 |
| work_keys_str_mv | AT mingchen dttcginetadualtemporaltransformernetworkwithmultiscalecontourguidedgraphinteractionforchangedetection AT wanshoujiang dttcginetadualtemporaltransformernetworkwithmultiscalecontourguidedgraphinteractionforchangedetection AT yuanzhou dttcginetadualtemporaltransformernetworkwithmultiscalecontourguidedgraphinteractionforchangedetection |