Dual-Branch Attention-Assisted CNN for Hyperspectral Image Classification
Convolutional neural network (CNN)-based hyperspectral image (HSI) classification models have developed rapidly in recent years due to their superiority. However, recent deep learning methods based on CNN tend to be deep networks with multiple parameters, which inevitably resulted in information red...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/23/6158 |
| _version_ | 1797462211584589824 |
|---|---|
| author | Wei Huang Zhuobing Zhao Le Sun Ming Ju |
| author_facet | Wei Huang Zhuobing Zhao Le Sun Ming Ju |
| author_sort | Wei Huang |
| collection | DOAJ |
| description | Convolutional neural network (CNN)-based hyperspectral image (HSI) classification models have developed rapidly in recent years due to their superiority. However, recent deep learning methods based on CNN tend to be deep networks with multiple parameters, which inevitably resulted in information redundancy and increased computational cost. We propose a dual-branch attention-assisted CNN (DBAA-CNN) for HSI classification to address these problems. The network consists of spatial-spectral and spectral attention branches. The spatial-spectral branch integrates multi-scale spatial information with cross-channel attention by extracting spatial–spectral information jointly utilizing a 3-D CNN and a pyramid squeeze-and-excitation attention (PSA) module. The spectral branch maps the original features to the spectral interaction space for feature representation and learning by adding an attention module. Finally, the spectral and spatial features are combined and input into the linear layer to generate the sample label. We conducted tests with three common hyperspectral datasets to test the efficacy of the framework. Our method outperformed state-of-the-art HSI classification algorithms based on classification accuracy and processing time. |
| first_indexed | 2024-03-09T17:33:17Z |
| format | Article |
| id | doaj.art-da70caa21f4f407a8d3d5207c8e17e49 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T17:33:17Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-da70caa21f4f407a8d3d5207c8e17e492023-11-24T12:06:46ZengMDPI AGRemote Sensing2072-42922022-12-011423615810.3390/rs14236158Dual-Branch Attention-Assisted CNN for Hyperspectral Image ClassificationWei Huang0Zhuobing Zhao1Le Sun2Ming Ju3College of Computer and Communication Engineering, Zhengzhou University of Light Industry, Zhengzhou 450002, ChinaCollege of Computer and Communication Engineering, Zhengzhou University of Light Industry, Zhengzhou 450002, ChinaSchool of Computer Science, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaCollege of Computer and Communication Engineering, Zhengzhou University of Light Industry, Zhengzhou 450002, ChinaConvolutional neural network (CNN)-based hyperspectral image (HSI) classification models have developed rapidly in recent years due to their superiority. However, recent deep learning methods based on CNN tend to be deep networks with multiple parameters, which inevitably resulted in information redundancy and increased computational cost. We propose a dual-branch attention-assisted CNN (DBAA-CNN) for HSI classification to address these problems. The network consists of spatial-spectral and spectral attention branches. The spatial-spectral branch integrates multi-scale spatial information with cross-channel attention by extracting spatial–spectral information jointly utilizing a 3-D CNN and a pyramid squeeze-and-excitation attention (PSA) module. The spectral branch maps the original features to the spectral interaction space for feature representation and learning by adding an attention module. Finally, the spectral and spatial features are combined and input into the linear layer to generate the sample label. We conducted tests with three common hyperspectral datasets to test the efficacy of the framework. Our method outperformed state-of-the-art HSI classification algorithms based on classification accuracy and processing time.https://www.mdpi.com/2072-4292/14/23/6158hyperspectral image (HSI) classificationpyramid squeeze-and-excitation attention (PSA)spatial–spectralcross-channel attention |
| spellingShingle | Wei Huang Zhuobing Zhao Le Sun Ming Ju Dual-Branch Attention-Assisted CNN for Hyperspectral Image Classification Remote Sensing hyperspectral image (HSI) classification pyramid squeeze-and-excitation attention (PSA) spatial–spectral cross-channel attention |
| title | Dual-Branch Attention-Assisted CNN for Hyperspectral Image Classification |
| title_full | Dual-Branch Attention-Assisted CNN for Hyperspectral Image Classification |
| title_fullStr | Dual-Branch Attention-Assisted CNN for Hyperspectral Image Classification |
| title_full_unstemmed | Dual-Branch Attention-Assisted CNN for Hyperspectral Image Classification |
| title_short | Dual-Branch Attention-Assisted CNN for Hyperspectral Image Classification |
| title_sort | dual branch attention assisted cnn for hyperspectral image classification |
| topic | hyperspectral image (HSI) classification pyramid squeeze-and-excitation attention (PSA) spatial–spectral cross-channel attention |
| url | https://www.mdpi.com/2072-4292/14/23/6158 |
| work_keys_str_mv | AT weihuang dualbranchattentionassistedcnnforhyperspectralimageclassification AT zhuobingzhao dualbranchattentionassistedcnnforhyperspectralimageclassification AT lesun dualbranchattentionassistedcnnforhyperspectralimageclassification AT mingju dualbranchattentionassistedcnnforhyperspectralimageclassification |