Graph Convolutional Enhanced Discriminative Broad Learning System for Hyperspectral Image Classification
Recently, broad learning system (BLS) have demonstrated excellent performance in hyperspectral images (HSI) classification. However, due to the complex geometric structure and spatial layout of HSI, the linear sparse features in broad learning system are difficult to fully represent hyperspectral da...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9866761/ |
| _version_ | 1811182704207593472 |
|---|---|
| author | Tuya |
| author_facet | Tuya |
| author_sort | Tuya |
| collection | DOAJ |
| description | Recently, broad learning system (BLS) have demonstrated excellent performance in hyperspectral images (HSI) classification. However, due to the complex geometric structure and spatial layout of HSI, the linear sparse features in broad learning system are difficult to fully represent hyperspectral data. In addition, the features learned by broad learning system lack more effective discriminative ability, which leads to the limited expressive ability of features. To address the issues, we propose a graph convolutional enhanced discriminative broad learning system (GCDBLS) for HSI Classification. GCDBLS aggregates the node information in the adjacency graph through graph convolution, and then learns the context relationship, so as to obtain rich nonlinear spatial spectral features in hyperspectral images; In order to extract more discriminative hyperspectral image features, GCDBLS introduces the concept of local intra-class scatter and local inter-class scatter. By minimizing the local intra-class feature distance and maximizing the local inter-class feature distance, GCDBLS can improve the discrimination ability of BLS extracted features. On three HSI datasets, the experiments compared with the latest classification methods show that the proposed method achieves good results, and improves the classification performance of hyperspectral images. |
| first_indexed | 2024-04-11T09:35:08Z |
| format | Article |
| id | doaj.art-08a6b48cf8034ff89cda94015da7fa9d |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-11T09:35:08Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-08a6b48cf8034ff89cda94015da7fa9d2022-12-22T04:31:41ZengIEEEIEEE Access2169-35362022-01-0110902999031110.1109/ACCESS.2022.32015379866761Graph Convolutional Enhanced Discriminative Broad Learning System for Hyperspectral Image Classification Tuya0College of Computer Science and Technology, Inner Mongolia Minzu University, Tongliao, ChinaRecently, broad learning system (BLS) have demonstrated excellent performance in hyperspectral images (HSI) classification. However, due to the complex geometric structure and spatial layout of HSI, the linear sparse features in broad learning system are difficult to fully represent hyperspectral data. In addition, the features learned by broad learning system lack more effective discriminative ability, which leads to the limited expressive ability of features. To address the issues, we propose a graph convolutional enhanced discriminative broad learning system (GCDBLS) for HSI Classification. GCDBLS aggregates the node information in the adjacency graph through graph convolution, and then learns the context relationship, so as to obtain rich nonlinear spatial spectral features in hyperspectral images; In order to extract more discriminative hyperspectral image features, GCDBLS introduces the concept of local intra-class scatter and local inter-class scatter. By minimizing the local intra-class feature distance and maximizing the local inter-class feature distance, GCDBLS can improve the discrimination ability of BLS extracted features. On three HSI datasets, the experiments compared with the latest classification methods show that the proposed method achieves good results, and improves the classification performance of hyperspectral images.https://ieeexplore.ieee.org/document/9866761/Broad learning systemhyperspectral imagegraph convolutionalintra-class feature distanceinter-class feature distance |
| spellingShingle | Tuya Graph Convolutional Enhanced Discriminative Broad Learning System for Hyperspectral Image Classification IEEE Access Broad learning system hyperspectral image graph convolutional intra-class feature distance inter-class feature distance |
| title | Graph Convolutional Enhanced Discriminative Broad Learning System for Hyperspectral Image Classification |
| title_full | Graph Convolutional Enhanced Discriminative Broad Learning System for Hyperspectral Image Classification |
| title_fullStr | Graph Convolutional Enhanced Discriminative Broad Learning System for Hyperspectral Image Classification |
| title_full_unstemmed | Graph Convolutional Enhanced Discriminative Broad Learning System for Hyperspectral Image Classification |
| title_short | Graph Convolutional Enhanced Discriminative Broad Learning System for Hyperspectral Image Classification |
| title_sort | graph convolutional enhanced discriminative broad learning system for hyperspectral image classification |
| topic | Broad learning system hyperspectral image graph convolutional intra-class feature distance inter-class feature distance |
| url | https://ieeexplore.ieee.org/document/9866761/ |
| work_keys_str_mv | AT tuya graphconvolutionalenhanceddiscriminativebroadlearningsystemforhyperspectralimageclassification |