Superpixel Nonlocal Weighting Joint Sparse Representation for Hyperspectral Image Classification
Joint sparse representation classification (JSRC) is a representative spectral–spatial classifier for hyperspectral images (HSIs). However, the JSRC is inappropriate for highly heterogeneous areas due to the spatial information being extracted from a fixed-sized neighborhood block, which is often un...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-04-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/9/2125 |
| _version_ | 1797503055818653696 |
|---|---|
| author | Aizhu Zhang Zhaojie Pan Hang Fu Genyun Sun Jun Rong Jinchang Ren Xiuping Jia Yanjuan Yao |
| author_facet | Aizhu Zhang Zhaojie Pan Hang Fu Genyun Sun Jun Rong Jinchang Ren Xiuping Jia Yanjuan Yao |
| author_sort | Aizhu Zhang |
| collection | DOAJ |
| description | Joint sparse representation classification (JSRC) is a representative spectral–spatial classifier for hyperspectral images (HSIs). However, the JSRC is inappropriate for highly heterogeneous areas due to the spatial information being extracted from a fixed-sized neighborhood block, which is often unable to conform to the naturally irregular structure of land cover. To address this problem, a superpixel-based JSRC with nonlocal weighting, i.e., superpixel-based nonlocal weighted JSRC (SNLW-JSRC), is proposed in this paper. In SNLW-JSRC, the superpixel representation of an HSI is first constructed based on an entropy rate segmentation method. This strategy forms homogeneous neighborhoods with naturally irregular structures and alleviates the inclusion of pixels from different classes in the process of spatial information extraction. Afterwards, the superpixel-based nonlocal weighting (SNLW) scheme is built to weigh the superpixel based on its structural and spectral information. In this way, the weight of one specific neighboring pixel is determined by the local structural similarity between the neighboring pixel and the central test pixel. Then, the obtained local weights are used to generate the weighted mean data for each superpixel. Finally, JSRC is used to produce the superpixel-level classification. This speeds up the sparse representation and makes the spatial content more centralized and compact. To verify the proposed SNLW-JSRC method, we conducted experiments on four benchmark hyperspectral datasets, namely Indian Pines, Pavia University, Salinas, and DFC2013. The experimental results suggest that the SNLW-JSRC can achieve better classification results than the other four SRC-based algorithms and the classical support vector machine algorithm. Moreover, the SNLW-JSRC can also outperform the other SRC-based algorithms, even with a small number of training samples. |
| first_indexed | 2024-03-10T03:45:02Z |
| format | Article |
| id | doaj.art-8b5ab1c91e1445b5986492ff5e0c62ed |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T03:45:02Z |
| publishDate | 2022-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-8b5ab1c91e1445b5986492ff5e0c62ed2023-11-23T09:10:57ZengMDPI AGRemote Sensing2072-42922022-04-01149212510.3390/rs14092125Superpixel Nonlocal Weighting Joint Sparse Representation for Hyperspectral Image ClassificationAizhu Zhang0Zhaojie Pan1Hang Fu2Genyun Sun3Jun Rong4Jinchang Ren5Xiuping Jia6Yanjuan Yao7College of Oceanography and Space Informatics, China University of Petroleum (East China), Qingdao 266580, ChinaCollege of Oceanography and Space Informatics, China University of Petroleum (East China), Qingdao 266580, ChinaCollege of Oceanography and Space Informatics, China University of Petroleum (East China), Qingdao 266580, ChinaCollege of Oceanography and Space Informatics, China University of Petroleum (East China), Qingdao 266580, ChinaPiesat Information Technology Co., Ltd., Beijing 100195, ChinaNational Subsea Centre, Robert Gordon University, Aberdeen AB10 7AQ, UKSchool of Engineering and Information Technology, University of New South Wales at Canberra, Canberra, ACT 2600, AustraliaSatellite Environment Center, Ministry of Environmental Protection, Beijing 100094, ChinaJoint sparse representation classification (JSRC) is a representative spectral–spatial classifier for hyperspectral images (HSIs). However, the JSRC is inappropriate for highly heterogeneous areas due to the spatial information being extracted from a fixed-sized neighborhood block, which is often unable to conform to the naturally irregular structure of land cover. To address this problem, a superpixel-based JSRC with nonlocal weighting, i.e., superpixel-based nonlocal weighted JSRC (SNLW-JSRC), is proposed in this paper. In SNLW-JSRC, the superpixel representation of an HSI is first constructed based on an entropy rate segmentation method. This strategy forms homogeneous neighborhoods with naturally irregular structures and alleviates the inclusion of pixels from different classes in the process of spatial information extraction. Afterwards, the superpixel-based nonlocal weighting (SNLW) scheme is built to weigh the superpixel based on its structural and spectral information. In this way, the weight of one specific neighboring pixel is determined by the local structural similarity between the neighboring pixel and the central test pixel. Then, the obtained local weights are used to generate the weighted mean data for each superpixel. Finally, JSRC is used to produce the superpixel-level classification. This speeds up the sparse representation and makes the spatial content more centralized and compact. To verify the proposed SNLW-JSRC method, we conducted experiments on four benchmark hyperspectral datasets, namely Indian Pines, Pavia University, Salinas, and DFC2013. The experimental results suggest that the SNLW-JSRC can achieve better classification results than the other four SRC-based algorithms and the classical support vector machine algorithm. Moreover, the SNLW-JSRC can also outperform the other SRC-based algorithms, even with a small number of training samples.https://www.mdpi.com/2072-4292/14/9/2125spatial–spectral fusionjoint sparse representation classification (JSRC)hyperspectral imagingsuperpixelnonlocal weighting |
| spellingShingle | Aizhu Zhang Zhaojie Pan Hang Fu Genyun Sun Jun Rong Jinchang Ren Xiuping Jia Yanjuan Yao Superpixel Nonlocal Weighting Joint Sparse Representation for Hyperspectral Image Classification Remote Sensing spatial–spectral fusion joint sparse representation classification (JSRC) hyperspectral imaging superpixel nonlocal weighting |
| title | Superpixel Nonlocal Weighting Joint Sparse Representation for Hyperspectral Image Classification |
| title_full | Superpixel Nonlocal Weighting Joint Sparse Representation for Hyperspectral Image Classification |
| title_fullStr | Superpixel Nonlocal Weighting Joint Sparse Representation for Hyperspectral Image Classification |
| title_full_unstemmed | Superpixel Nonlocal Weighting Joint Sparse Representation for Hyperspectral Image Classification |
| title_short | Superpixel Nonlocal Weighting Joint Sparse Representation for Hyperspectral Image Classification |
| title_sort | superpixel nonlocal weighting joint sparse representation for hyperspectral image classification |
| topic | spatial–spectral fusion joint sparse representation classification (JSRC) hyperspectral imaging superpixel nonlocal weighting |
| url | https://www.mdpi.com/2072-4292/14/9/2125 |
| work_keys_str_mv | AT aizhuzhang superpixelnonlocalweightingjointsparserepresentationforhyperspectralimageclassification AT zhaojiepan superpixelnonlocalweightingjointsparserepresentationforhyperspectralimageclassification AT hangfu superpixelnonlocalweightingjointsparserepresentationforhyperspectralimageclassification AT genyunsun superpixelnonlocalweightingjointsparserepresentationforhyperspectralimageclassification AT junrong superpixelnonlocalweightingjointsparserepresentationforhyperspectralimageclassification AT jinchangren superpixelnonlocalweightingjointsparserepresentationforhyperspectralimageclassification AT xiupingjia superpixelnonlocalweightingjointsparserepresentationforhyperspectralimageclassification AT yanjuanyao superpixelnonlocalweightingjointsparserepresentationforhyperspectralimageclassification |