DA-IMRN: Dual-Attention-Guided Interactive Multi-Scale Residual Network for Hyperspectral Image Classification
Deep learning-based fusion of spectral-spatial information is increasingly dominant for hyperspectral image (HSI) classification. However, due to insufficient samples, current feature fusion methods often neglect joint interactions. In this paper, to further improve the classification accuracy, we p...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/3/530 |
| _version_ | 1797485062584795136 |
|---|---|
| author | Liang Zou Zhifan Zhang Haijia Du Meng Lei Yong Xue Z. Jane Wang |
| author_facet | Liang Zou Zhifan Zhang Haijia Du Meng Lei Yong Xue Z. Jane Wang |
| author_sort | Liang Zou |
| collection | DOAJ |
| description | Deep learning-based fusion of spectral-spatial information is increasingly dominant for hyperspectral image (HSI) classification. However, due to insufficient samples, current feature fusion methods often neglect joint interactions. In this paper, to further improve the classification accuracy, we propose a dual-attention-guided interactive multi-scale residual network (DA-IMRN) to explore the joint spectral-spatial information and assign pixel-wise labels for HSIs without information leakage. In DA-IMRN, two branches focusing on spatial and spectral information separately are employed for feature extraction. A bidirectional-attention mechanism is employed to guide the interactive feature learning between two branches and promote refined feature maps. In addition, we extract deep multi-scale features corresponding to multiple receptive fields from limited samples via a multi-scale spectral/spatial residual block, to improve classification performance. Experimental results on three benchmark datasets (i.e., Salinas Valley, Pavia University, and Indian Pines) support that attention-guided multi-scale feature learning can effectively explore the joint spectral-spatial information. The proposed method outperforms state-of-the-art methods with the overall accuracy of 91.26%, 93.33%, and 82.38%, and the average accuracy of 94.22%, 89.61%, and 80.35%, respectively. |
| first_indexed | 2024-03-09T23:13:28Z |
| format | Article |
| id | doaj.art-2e0e7c9d3464483b84b603d8a0c2d80b |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T23:13:28Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-2e0e7c9d3464483b84b603d8a0c2d80b2023-11-23T17:39:13ZengMDPI AGRemote Sensing2072-42922022-01-0114353010.3390/rs14030530DA-IMRN: Dual-Attention-Guided Interactive Multi-Scale Residual Network for Hyperspectral Image ClassificationLiang Zou0Zhifan Zhang1Haijia Du2Meng Lei3Yong Xue4Z. Jane Wang5Engineering Research Center of Intelligent Control for Underground Space, Ministry of Education, China University of Mining and Technology, Xuzhou 221116, ChinaEngineering Research Center of Intelligent Control for Underground Space, Ministry of Education, China University of Mining and Technology, Xuzhou 221116, ChinaEngineering Research Center of Intelligent Control for Underground Space, Ministry of Education, China University of Mining and Technology, Xuzhou 221116, ChinaEngineering Research Center of Intelligent Control for Underground Space, Ministry of Education, China University of Mining and Technology, Xuzhou 221116, ChinaSchool of Environment Science and Spatial Informatics, China University of Mining and Technology, Xuzhou 221116, ChinaDepartment of Electrical and Computer Engineering, University of British Columbia, Vancouver, BC V6T 1Z4, CanadaDeep learning-based fusion of spectral-spatial information is increasingly dominant for hyperspectral image (HSI) classification. However, due to insufficient samples, current feature fusion methods often neglect joint interactions. In this paper, to further improve the classification accuracy, we propose a dual-attention-guided interactive multi-scale residual network (DA-IMRN) to explore the joint spectral-spatial information and assign pixel-wise labels for HSIs without information leakage. In DA-IMRN, two branches focusing on spatial and spectral information separately are employed for feature extraction. A bidirectional-attention mechanism is employed to guide the interactive feature learning between two branches and promote refined feature maps. In addition, we extract deep multi-scale features corresponding to multiple receptive fields from limited samples via a multi-scale spectral/spatial residual block, to improve classification performance. Experimental results on three benchmark datasets (i.e., Salinas Valley, Pavia University, and Indian Pines) support that attention-guided multi-scale feature learning can effectively explore the joint spectral-spatial information. The proposed method outperforms state-of-the-art methods with the overall accuracy of 91.26%, 93.33%, and 82.38%, and the average accuracy of 94.22%, 89.61%, and 80.35%, respectively.https://www.mdpi.com/2072-4292/14/3/530hyperspectral image classificationinteractiondual-attention mechanismmulti-scale spectral/spatial residual block |
| spellingShingle | Liang Zou Zhifan Zhang Haijia Du Meng Lei Yong Xue Z. Jane Wang DA-IMRN: Dual-Attention-Guided Interactive Multi-Scale Residual Network for Hyperspectral Image Classification Remote Sensing hyperspectral image classification interaction dual-attention mechanism multi-scale spectral/spatial residual block |
| title | DA-IMRN: Dual-Attention-Guided Interactive Multi-Scale Residual Network for Hyperspectral Image Classification |
| title_full | DA-IMRN: Dual-Attention-Guided Interactive Multi-Scale Residual Network for Hyperspectral Image Classification |
| title_fullStr | DA-IMRN: Dual-Attention-Guided Interactive Multi-Scale Residual Network for Hyperspectral Image Classification |
| title_full_unstemmed | DA-IMRN: Dual-Attention-Guided Interactive Multi-Scale Residual Network for Hyperspectral Image Classification |
| title_short | DA-IMRN: Dual-Attention-Guided Interactive Multi-Scale Residual Network for Hyperspectral Image Classification |
| title_sort | da imrn dual attention guided interactive multi scale residual network for hyperspectral image classification |
| topic | hyperspectral image classification interaction dual-attention mechanism multi-scale spectral/spatial residual block |
| url | https://www.mdpi.com/2072-4292/14/3/530 |
| work_keys_str_mv | AT liangzou daimrndualattentionguidedinteractivemultiscaleresidualnetworkforhyperspectralimageclassification AT zhifanzhang daimrndualattentionguidedinteractivemultiscaleresidualnetworkforhyperspectralimageclassification AT haijiadu daimrndualattentionguidedinteractivemultiscaleresidualnetworkforhyperspectralimageclassification AT menglei daimrndualattentionguidedinteractivemultiscaleresidualnetworkforhyperspectralimageclassification AT yongxue daimrndualattentionguidedinteractivemultiscaleresidualnetworkforhyperspectralimageclassification AT zjanewang daimrndualattentionguidedinteractivemultiscaleresidualnetworkforhyperspectralimageclassification |