A Region-Based Feature Fusion Network for VHR Image Change Detection
Deep learning (DL)-based architectures have shown a strong capacity to identify changes. However, existing change detection (CD) networks still suffer from limited applicability when it comes to multi-scale targets and spatially misaligned objects. For the sake of tackling the above problems, a regi...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/21/5577 |
| _version_ | 1797466578046943232 |
|---|---|
| author | Pan Chen Cong Li Bing Zhang Zhengchao Chen Xuan Yang Kaixuan Lu Lina Zhuang |
| author_facet | Pan Chen Cong Li Bing Zhang Zhengchao Chen Xuan Yang Kaixuan Lu Lina Zhuang |
| author_sort | Pan Chen |
| collection | DOAJ |
| description | Deep learning (DL)-based architectures have shown a strong capacity to identify changes. However, existing change detection (CD) networks still suffer from limited applicability when it comes to multi-scale targets and spatially misaligned objects. For the sake of tackling the above problems, a region-based feature fusion network (RFNet) for CD of very high spatial resolution (VHR) remote sensing images is proposed. RFNet uses a fully convolutional Siamese network backbone where a multi-stage feature interaction module (MFIM) is embedded in the dual encoder and a series of region-based feature fusion modules (RFFMs) is used to generate change information. The MFIM fuses features in different stages to enhance the interaction of multi-scale information and help the network better distinguish complex ground objects. The RFFM is built based on region similarity (RSIM), which measures the similarity of bitemporal features with neighborhoods. The RFFM can reduce the impact of spatially offset bitemporal targets and accurately identify changes in bitemporal images. We also design a deep supervise strategy by directly introducing RSIM into loss calculation and shortening the error propagation distance. We validate RFNet with two popular CD datasets: the SECOND dataset and the WHU dataset. The qualitative and quantitative comparison results demonstrate the high capacity and strong robustness of RFNet. We also conduct robustness experiments and the results demonstrate that RFNet can deal with spatially shifted bitemporal images. |
| first_indexed | 2024-03-09T18:40:46Z |
| format | Article |
| id | doaj.art-16a6137095e34137841d86e0177f7ec9 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T18:40:46Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-16a6137095e34137841d86e0177f7ec92023-11-24T06:41:17ZengMDPI AGRemote Sensing2072-42922022-11-011421557710.3390/rs14215577A Region-Based Feature Fusion Network for VHR Image Change DetectionPan Chen0Cong Li1Bing Zhang2Zhengchao Chen3Xuan Yang4Kaixuan Lu5Lina Zhuang6Key Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAirborne Remote Sensing Center, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAirborne Remote Sensing Center, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAirborne Remote Sensing Center, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Computational Optical Imaging Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaDeep learning (DL)-based architectures have shown a strong capacity to identify changes. However, existing change detection (CD) networks still suffer from limited applicability when it comes to multi-scale targets and spatially misaligned objects. For the sake of tackling the above problems, a region-based feature fusion network (RFNet) for CD of very high spatial resolution (VHR) remote sensing images is proposed. RFNet uses a fully convolutional Siamese network backbone where a multi-stage feature interaction module (MFIM) is embedded in the dual encoder and a series of region-based feature fusion modules (RFFMs) is used to generate change information. The MFIM fuses features in different stages to enhance the interaction of multi-scale information and help the network better distinguish complex ground objects. The RFFM is built based on region similarity (RSIM), which measures the similarity of bitemporal features with neighborhoods. The RFFM can reduce the impact of spatially offset bitemporal targets and accurately identify changes in bitemporal images. We also design a deep supervise strategy by directly introducing RSIM into loss calculation and shortening the error propagation distance. We validate RFNet with two popular CD datasets: the SECOND dataset and the WHU dataset. The qualitative and quantitative comparison results demonstrate the high capacity and strong robustness of RFNet. We also conduct robustness experiments and the results demonstrate that RFNet can deal with spatially shifted bitemporal images.https://www.mdpi.com/2072-4292/14/21/5577deep learningchange detectionspatially offset images |
| spellingShingle | Pan Chen Cong Li Bing Zhang Zhengchao Chen Xuan Yang Kaixuan Lu Lina Zhuang A Region-Based Feature Fusion Network for VHR Image Change Detection Remote Sensing deep learning change detection spatially offset images |
| title | A Region-Based Feature Fusion Network for VHR Image Change Detection |
| title_full | A Region-Based Feature Fusion Network for VHR Image Change Detection |
| title_fullStr | A Region-Based Feature Fusion Network for VHR Image Change Detection |
| title_full_unstemmed | A Region-Based Feature Fusion Network for VHR Image Change Detection |
| title_short | A Region-Based Feature Fusion Network for VHR Image Change Detection |
| title_sort | region based feature fusion network for vhr image change detection |
| topic | deep learning change detection spatially offset images |
| url | https://www.mdpi.com/2072-4292/14/21/5577 |
| work_keys_str_mv | AT panchen aregionbasedfeaturefusionnetworkforvhrimagechangedetection AT congli aregionbasedfeaturefusionnetworkforvhrimagechangedetection AT bingzhang aregionbasedfeaturefusionnetworkforvhrimagechangedetection AT zhengchaochen aregionbasedfeaturefusionnetworkforvhrimagechangedetection AT xuanyang aregionbasedfeaturefusionnetworkforvhrimagechangedetection AT kaixuanlu aregionbasedfeaturefusionnetworkforvhrimagechangedetection AT linazhuang aregionbasedfeaturefusionnetworkforvhrimagechangedetection AT panchen regionbasedfeaturefusionnetworkforvhrimagechangedetection AT congli regionbasedfeaturefusionnetworkforvhrimagechangedetection AT bingzhang regionbasedfeaturefusionnetworkforvhrimagechangedetection AT zhengchaochen regionbasedfeaturefusionnetworkforvhrimagechangedetection AT xuanyang regionbasedfeaturefusionnetworkforvhrimagechangedetection AT kaixuanlu regionbasedfeaturefusionnetworkforvhrimagechangedetection AT linazhuang regionbasedfeaturefusionnetworkforvhrimagechangedetection |