Multiscale Graph Transformer Network With Dynamic Superpixel Pyramid for Hyperspectral Image Classification
Hyperspectral image (HSI) classification plays a crucial role in remote sensing applications, leveraging the rich spectral and spatial information inherent in HSI. However, the diverse feature scales of ground objects pose challenges for traditional methods, which often employ fixed-scale feature ex...
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/11003964/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850131786610245632 |
|---|---|
| author | Tingting Wang Yao Sun Yunfeng Hu |
| author_facet | Tingting Wang Yao Sun Yunfeng Hu |
| author_sort | Tingting Wang |
| collection | DOAJ |
| description | Hyperspectral image (HSI) classification plays a crucial role in remote sensing applications, leveraging the rich spectral and spatial information inherent in HSI. However, the diverse feature scales of ground objects pose challenges for traditional methods, which often employ fixed-scale feature extraction and struggle to adapt dynamically to varying spatial structures. Moreover, conventional approaches typically emphasize either local spectral or spatial features while neglecting global contextual relationships. To address these limitations, we propose a multi-scale graph transformer network (MSGTN), which captures spatial features at different scales through multiscale graph convolutional networks (GCNs) with adaptive graph structures. First, we employ differentiable superpixel segmentation to construct flexible graph topologies in the feature space, enabling adaptive representation learning at various spatial scales. Second, a multiscale GCN is introduced to jointly model spectral and spatial features, enhancing the network's capacity to represent local structural information. Finally, a spectral–spatial transformer module is integrated to model complex global dependencies among superpixels, capturing long-range contextual relationships. Extensive experiments conducted on four widely used hyperspectral datasets, University of Pavia (UP), Indian Pines (IP), Salinas Valley (SA), and WHU-Hi-HongHu (WHHH), demonstrate the effectiveness of the proposed MSGTN. The model achieves classification accuracies of 99.01% (UP), 96.94% (IP), 99.43% (SA), and 97.95% (WHHH) using only 1%, 5%, 1%, and 1% of labeled training samples, respectively. Compared with state-of-the-art methods, MSGTN exhibits superior classification performance and enhanced robustness across varying data conditions. |
| format | Article |
| id | doaj-art-80ab66fba5f54526ab04f1a50d6725f3 |
| institution | OA Journals |
| 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-80ab66fba5f54526ab04f1a50d6725f32025-08-20T02:32:22ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118134381345610.1109/JSTARS.2025.357026711003964Multiscale Graph Transformer Network With Dynamic Superpixel Pyramid for Hyperspectral Image ClassificationTingting Wang0https://orcid.org/0000-0002-7605-4860Yao Sun1https://orcid.org/0000-0002-4924-7624Yunfeng Hu2https://orcid.org/0000-0003-4068-0664School of Communication Engineering, Jilin University, Changchun, ChinaNational Key Laboratory of Automotive Chassis Integration and Bionics, Jilin University, Changchun, ChinaSchool of Communication Engineering, Jilin University, Changchun, ChinaHyperspectral image (HSI) classification plays a crucial role in remote sensing applications, leveraging the rich spectral and spatial information inherent in HSI. However, the diverse feature scales of ground objects pose challenges for traditional methods, which often employ fixed-scale feature extraction and struggle to adapt dynamically to varying spatial structures. Moreover, conventional approaches typically emphasize either local spectral or spatial features while neglecting global contextual relationships. To address these limitations, we propose a multi-scale graph transformer network (MSGTN), which captures spatial features at different scales through multiscale graph convolutional networks (GCNs) with adaptive graph structures. First, we employ differentiable superpixel segmentation to construct flexible graph topologies in the feature space, enabling adaptive representation learning at various spatial scales. Second, a multiscale GCN is introduced to jointly model spectral and spatial features, enhancing the network's capacity to represent local structural information. Finally, a spectral–spatial transformer module is integrated to model complex global dependencies among superpixels, capturing long-range contextual relationships. Extensive experiments conducted on four widely used hyperspectral datasets, University of Pavia (UP), Indian Pines (IP), Salinas Valley (SA), and WHU-Hi-HongHu (WHHH), demonstrate the effectiveness of the proposed MSGTN. The model achieves classification accuracies of 99.01% (UP), 96.94% (IP), 99.43% (SA), and 97.95% (WHHH) using only 1%, 5%, 1%, and 1% of labeled training samples, respectively. Compared with state-of-the-art methods, MSGTN exhibits superior classification performance and enhanced robustness across varying data conditions.https://ieeexplore.ieee.org/document/11003964/Dynamic superpixel pyramidhyperspectral image (HSI) classificationmultiscale graphspectral–spatial transformer |
| spellingShingle | Tingting Wang Yao Sun Yunfeng Hu Multiscale Graph Transformer Network With Dynamic Superpixel Pyramid for Hyperspectral Image Classification IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Dynamic superpixel pyramid hyperspectral image (HSI) classification multiscale graph spectral–spatial transformer |
| title | Multiscale Graph Transformer Network With Dynamic Superpixel Pyramid for Hyperspectral Image Classification |
| title_full | Multiscale Graph Transformer Network With Dynamic Superpixel Pyramid for Hyperspectral Image Classification |
| title_fullStr | Multiscale Graph Transformer Network With Dynamic Superpixel Pyramid for Hyperspectral Image Classification |
| title_full_unstemmed | Multiscale Graph Transformer Network With Dynamic Superpixel Pyramid for Hyperspectral Image Classification |
| title_short | Multiscale Graph Transformer Network With Dynamic Superpixel Pyramid for Hyperspectral Image Classification |
| title_sort | multiscale graph transformer network with dynamic superpixel pyramid for hyperspectral image classification |
| topic | Dynamic superpixel pyramid hyperspectral image (HSI) classification multiscale graph spectral–spatial transformer |
| url | https://ieeexplore.ieee.org/document/11003964/ |
| work_keys_str_mv | AT tingtingwang multiscalegraphtransformernetworkwithdynamicsuperpixelpyramidforhyperspectralimageclassification AT yaosun multiscalegraphtransformernetworkwithdynamicsuperpixelpyramidforhyperspectralimageclassification AT yunfenghu multiscalegraphtransformernetworkwithdynamicsuperpixelpyramidforhyperspectralimageclassification |