SGANet: A Siamese Geometry-Aware Network for Remote Sensing Change Detection
The significant progress in the fields of deep learning and computer vision has propelled the development of remote sensing change detection. However, we noticed that previous methods still rely on the single visual modality and cannot effectively utilize other prior information, such as elevation o...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10884698/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849764553440624640 |
|---|---|
| author | Jiangwei Chen Sijun Dong Xiaoliang Meng |
| author_facet | Jiangwei Chen Sijun Dong Xiaoliang Meng |
| author_sort | Jiangwei Chen |
| collection | DOAJ |
| description | The significant progress in the fields of deep learning and computer vision has propelled the development of remote sensing change detection. However, we noticed that previous methods still rely on the single visual modality and cannot effectively utilize other prior information, such as elevation or depth maps. Therefore, this article presents a novel Siamese geometry-aware network (SGANet) intended for RGB-D remote sensing change detection. By incorporating both RGB data and geometry priors such as relative depth estimations derived from a monocular depth estimation model such as DepthAnythingV2, SGANet surpasses the limitations of traditional methods that primarily depend on visual data. The proposed network employs a shared siamese encoder architecture with a lightweight decoder head for efficient change map prediction. Within the encoder blocks, we integrated a local feature extraction block that excels at capturing fine-grained features and a global cross-attention block that focuses on contextual features between different modalities. Furthermore, we engineered a dual-path fusion structure that facilitates a seamless integration of vision and geometry features. Extensive experiments on the LEVIR-CD, WHU-CD, SYSU-CD, and S2Looking-CD datasets demonstrated that SGANet achieved substantial enhancements in F1-Score and intersection over union compared to benchmark methods that are in vogue. By integrating geometry priors and effective multimodal fusion mechanisms, SGANet promoted the development of geometry-aware change detection, further enhancing optimal performance. |
| format | Article |
| id | doaj-art-ed259761a48547799ac3519768ff593d |
| institution | DOAJ |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-ed259761a48547799ac3519768ff593d2025-08-20T03:05:06ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01186232624810.1109/JSTARS.2025.353973310884698SGANet: A Siamese Geometry-Aware Network for Remote Sensing Change DetectionJiangwei Chen0https://orcid.org/0000-0002-3576-8007Sijun Dong1https://orcid.org/0000-0003-4218-6790Xiaoliang Meng2https://orcid.org/0000-0002-3271-9314School of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaThe significant progress in the fields of deep learning and computer vision has propelled the development of remote sensing change detection. However, we noticed that previous methods still rely on the single visual modality and cannot effectively utilize other prior information, such as elevation or depth maps. Therefore, this article presents a novel Siamese geometry-aware network (SGANet) intended for RGB-D remote sensing change detection. By incorporating both RGB data and geometry priors such as relative depth estimations derived from a monocular depth estimation model such as DepthAnythingV2, SGANet surpasses the limitations of traditional methods that primarily depend on visual data. The proposed network employs a shared siamese encoder architecture with a lightweight decoder head for efficient change map prediction. Within the encoder blocks, we integrated a local feature extraction block that excels at capturing fine-grained features and a global cross-attention block that focuses on contextual features between different modalities. Furthermore, we engineered a dual-path fusion structure that facilitates a seamless integration of vision and geometry features. Extensive experiments on the LEVIR-CD, WHU-CD, SYSU-CD, and S2Looking-CD datasets demonstrated that SGANet achieved substantial enhancements in F1-Score and intersection over union compared to benchmark methods that are in vogue. By integrating geometry priors and effective multimodal fusion mechanisms, SGANet promoted the development of geometry-aware change detection, further enhancing optimal performance.https://ieeexplore.ieee.org/document/10884698/Deep learning for Earth observationGeometry-aware perceptionmultimodal data fusionremote sensing change detection |
| spellingShingle | Jiangwei Chen Sijun Dong Xiaoliang Meng SGANet: A Siamese Geometry-Aware Network for Remote Sensing Change Detection IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Deep learning for Earth observation Geometry-aware perception multimodal data fusion remote sensing change detection |
| title | SGANet: A Siamese Geometry-Aware Network for Remote Sensing Change Detection |
| title_full | SGANet: A Siamese Geometry-Aware Network for Remote Sensing Change Detection |
| title_fullStr | SGANet: A Siamese Geometry-Aware Network for Remote Sensing Change Detection |
| title_full_unstemmed | SGANet: A Siamese Geometry-Aware Network for Remote Sensing Change Detection |
| title_short | SGANet: A Siamese Geometry-Aware Network for Remote Sensing Change Detection |
| title_sort | sganet a siamese geometry aware network for remote sensing change detection |
| topic | Deep learning for Earth observation Geometry-aware perception multimodal data fusion remote sensing change detection |
| url | https://ieeexplore.ieee.org/document/10884698/ |
| work_keys_str_mv | AT jiangweichen sganetasiamesegeometryawarenetworkforremotesensingchangedetection AT sijundong sganetasiamesegeometryawarenetworkforremotesensingchangedetection AT xiaoliangmeng sganetasiamesegeometryawarenetworkforremotesensingchangedetection |