Efficient Spectral-Spatial Fusion With Multiscale and Adaptive Attention for Hyperspectral Image Classification
In hyperspectral image (HSI) classification, convolutional neural networks (CNNs) are widely used due to their ability to leverage the rich spectral information across multiple bands. However, HSI classification still faces various challenges, including insufficient spectral-spatial representation,...
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/10745620/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841533434442285056 |
|---|---|
| author | Xiaoqing Wan Feng Chen Weizhe Gao Yupeng He Hui Liu Zhize Li |
| author_facet | Xiaoqing Wan Feng Chen Weizhe Gao Yupeng He Hui Liu Zhize Li |
| author_sort | Xiaoqing Wan |
| collection | DOAJ |
| description | In hyperspectral image (HSI) classification, convolutional neural networks (CNNs) are widely used due to their ability to leverage the rich spectral information across multiple bands. However, HSI classification still faces various challenges, including insufficient spectral-spatial representation, excessive redundant information, and difficulties in effectively integrating features of different scales, etc., which may lead to reduced classification accuracy. In order to reduce the computational cost and improve the classification accuracy of land cover categories, an efficient spectral-spatial fusion method (ESSF) is proposed, which is based on the following modules: a multiscale feature fusion module (MSFFM), an efficient adaptive spectral-spatial feature extraction module (EASSFEM), and a context-aware fusion network (CFN). First, the MSFFM utilizes CNNs to extract and fuse features from various scales to comprehensively capture detailed spectral information in HSIs. Second, the EASSFEM dynamically adjusts the feature extraction process to optimize the fusion and representation of spectral-spatial features. In addition, it incorporates an adaptive attention mechanism to enhance the focus on relevant spectral-spatial features. Finally, the CFN enhances the model's ability to understand contextual relationships within the images, thereby improving classification accuracy. Extensive experiments conducted on four public datasets (Houston2013, Botswana, WHU-Hi-HanChuan and WHU-Hi-HongHu) demonstrate that the proposed ESSF method significantly outperforms nine other state-of-the-art methods in terms of classification accuracy. |
| format | Article |
| id | doaj-art-b644dd905935470cacc3ab699c3e9197 |
| institution | Kabale University |
| 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-b644dd905935470cacc3ab699c3e91972025-01-16T00:00:31ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01181196121110.1109/JSTARS.2024.349235110745620Efficient Spectral-Spatial Fusion With Multiscale and Adaptive Attention for Hyperspectral Image ClassificationXiaoqing Wan0https://orcid.org/0000-0001-9948-9173Feng Chen1https://orcid.org/0009-0002-5966-7988Weizhe Gao2Yupeng He3https://orcid.org/0009-0009-4830-6467Hui Liu4https://orcid.org/0009-0005-6729-5821Zhize Li5https://orcid.org/0009-0003-5110-2058College of Computer Science and Technology, Hengyang Normal University, Hengyang, ChinaCollege of Computer Science and Technology, Hengyang Normal University, Hengyang, ChinaCollege of Computer Science and Technology, Hengyang Normal University, Hengyang, ChinaCollege of Computer Science and Technology, Hengyang Normal University, Hengyang, ChinaCollege of Computer Science and Technology, Hengyang Normal University, Hengyang, ChinaCollege of Computer Science and Technology, Hengyang Normal University, Hengyang, ChinaIn hyperspectral image (HSI) classification, convolutional neural networks (CNNs) are widely used due to their ability to leverage the rich spectral information across multiple bands. However, HSI classification still faces various challenges, including insufficient spectral-spatial representation, excessive redundant information, and difficulties in effectively integrating features of different scales, etc., which may lead to reduced classification accuracy. In order to reduce the computational cost and improve the classification accuracy of land cover categories, an efficient spectral-spatial fusion method (ESSF) is proposed, which is based on the following modules: a multiscale feature fusion module (MSFFM), an efficient adaptive spectral-spatial feature extraction module (EASSFEM), and a context-aware fusion network (CFN). First, the MSFFM utilizes CNNs to extract and fuse features from various scales to comprehensively capture detailed spectral information in HSIs. Second, the EASSFEM dynamically adjusts the feature extraction process to optimize the fusion and representation of spectral-spatial features. In addition, it incorporates an adaptive attention mechanism to enhance the focus on relevant spectral-spatial features. Finally, the CFN enhances the model's ability to understand contextual relationships within the images, thereby improving classification accuracy. Extensive experiments conducted on four public datasets (Houston2013, Botswana, WHU-Hi-HanChuan and WHU-Hi-HongHu) demonstrate that the proposed ESSF method significantly outperforms nine other state-of-the-art methods in terms of classification accuracy.https://ieeexplore.ieee.org/document/10745620/Attention mechanismconvolutional neural network (CNN)hyperspectral image (HSI)multiscale feature |
| spellingShingle | Xiaoqing Wan Feng Chen Weizhe Gao Yupeng He Hui Liu Zhize Li Efficient Spectral-Spatial Fusion With Multiscale and Adaptive Attention for Hyperspectral Image Classification IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Attention mechanism convolutional neural network (CNN) hyperspectral image (HSI) multiscale feature |
| title | Efficient Spectral-Spatial Fusion With Multiscale and Adaptive Attention for Hyperspectral Image Classification |
| title_full | Efficient Spectral-Spatial Fusion With Multiscale and Adaptive Attention for Hyperspectral Image Classification |
| title_fullStr | Efficient Spectral-Spatial Fusion With Multiscale and Adaptive Attention for Hyperspectral Image Classification |
| title_full_unstemmed | Efficient Spectral-Spatial Fusion With Multiscale and Adaptive Attention for Hyperspectral Image Classification |
| title_short | Efficient Spectral-Spatial Fusion With Multiscale and Adaptive Attention for Hyperspectral Image Classification |
| title_sort | efficient spectral spatial fusion with multiscale and adaptive attention for hyperspectral image classification |
| topic | Attention mechanism convolutional neural network (CNN) hyperspectral image (HSI) multiscale feature |
| url | https://ieeexplore.ieee.org/document/10745620/ |
| work_keys_str_mv | AT xiaoqingwan efficientspectralspatialfusionwithmultiscaleandadaptiveattentionforhyperspectralimageclassification AT fengchen efficientspectralspatialfusionwithmultiscaleandadaptiveattentionforhyperspectralimageclassification AT weizhegao efficientspectralspatialfusionwithmultiscaleandadaptiveattentionforhyperspectralimageclassification AT yupenghe efficientspectralspatialfusionwithmultiscaleandadaptiveattentionforhyperspectralimageclassification AT huiliu efficientspectralspatialfusionwithmultiscaleandadaptiveattentionforhyperspectralimageclassification AT zhizeli efficientspectralspatialfusionwithmultiscaleandadaptiveattentionforhyperspectralimageclassification |