Smart City Gnosys
Smart city article details
| Title | D2Sformer: Dual Attention-Dynamic Bidirectional Transformer For Semantic Segmentation Of Urban Remote Sensing Images |
|---|---|
| ID_Doc | 17083 |
| Authors | Yan Y.; Li J.; Zhang J.; Wang L.; Zhuo L. |
| Year | 2025 |
| Published | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 18 |
| DOI | http://dx.doi.org/10.1109/JSTARS.2025.3566159 |
| Abstract | Semantic segmentation of urban remote sensing images (URSIs) is crucial for the development of smart cities and the acceleration of urban information. In view of the characteristics of URSIs with intricate backgrounds, dense distributions of feature objects, and varying object scales, we propose a dual attention-dynamic bidirectional transformer (D2SFormer) for semantic segmentation of URSIs. First, a dual attention transformer encoder has a global branch of cross-shaped window self-attention and a local branch of convolutional channel attention block (CCAB), which helps to enhance pixel-level semantic discrimination. Then, CCAB combines the local sensitivity of Conv with the nonlocal modeling capability of efficient cannel attention for overlapping region resolution. Finally, the decoder dynamically fuses multiscale features to reconstruct boundaries and suppress artifacts by using a bidirectional feature pyramid network and dynamic upsampling. Experimental results demonstrate that our D2SFormer achieves mIoUs of 87.84%, 81.24%, and 53.19% on the Potsdam, Vaihingen, and LoveDA datasets, respectively, which is highly competitive with existing methods. © 2008-2012 IEEE. |
| Author Keywords | dual attention; dynamic bidirectional Transformer; semantic segmentation; Urban remote sensing images (URSIs) |
Similar Articles
| Id | Similarity | Authors | Title | Published |
|---|---|---|---|---|
26188  | 0.892 | Tang S.; He L. | Fattformer: Feature Attention Refinement Model For Semantic Segmentation In High-Resolution Remote Sensing Images | Proceedings of SPIE - The International Society for Optical Engineering, 13506 (2025) |
| 8040 | 0.875 | Gonthina N.; Narasimha Prasad L.V. | An Enhanced Convolutional Neural Network Architecture For Semantic Segmentation In High-Resolution Remote Sensing Images | Discover Computing, 28, 1 (2025) |
| 48265 | 0.871 | Arulananth T.S.; Kuppusamy P.G.; Ayyasamy R.K.; Alhashmi S.M.; Mahalakshmi M.; Vasanth K.; Chinnasamy P. | Semantic Segmentation Of Urban Environments: Leveraging U-Net Deep Learning Model For Cityscape Image Analysis | PLoS ONE, 19, 4 April (2024) |
| 48263 | 0.865 | Vetsa P.; Buddharaju A.; Dasari V.; Bogila Y.; Sundaresan A.A.; Paul M.B. | Semantic Segmentation Of Remote Sensing Images Of Urban Regions Using Deep Learning Methods | AIP Conference Proceedings, 3291, 1 (2025) |
| 48255 | 0.862 | Jia P.; Chen C.; Zhang D.; Sang Y.; Zhang L. | Semantic Segmentation Of Deep Learning Remote Sensing Images Based On Band Combination Principle: Application In Urban Planning And Land Use | Computer Communications, 217 (2024) |