Multiscale Weighted Adjacent Superpixel-Based Composite Kernel for Hyperspectral Image Classification
This paper presents a composite kernel method (MWASCK) based on multiscale weighted adjacent superpixels (ASs) to classify hyperspectral image (HSI). The MWASCK adequately exploits spatial-spectral features of weighted adjacent superpixels to guarantee that more accurate spectral features can be ext...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-02-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/4/820 |
| _version_ | 1797395537726537728 |
|---|---|
| author | Yaokang Zhang Yunjie Chen |
| author_facet | Yaokang Zhang Yunjie Chen |
| author_sort | Yaokang Zhang |
| collection | DOAJ |
| description | This paper presents a composite kernel method (MWASCK) based on multiscale weighted adjacent superpixels (ASs) to classify hyperspectral image (HSI). The MWASCK adequately exploits spatial-spectral features of weighted adjacent superpixels to guarantee that more accurate spectral features can be extracted. Firstly, we use a superpixel segmentation algorithm to divide HSI into multiple superpixels. Secondly, the similarities between each target superpixel and its ASs are calculated to construct the spatial features. Finally, a weighted AS-based composite kernel (WASCK) method for HSI classification is proposed. In order to avoid seeking for the optimal superpixel scale and fuse the multiscale spatial features, the MWASCK method uses multiscale weighted superpixel neighbor information. Experiments from two real HSIs indicate that superior performance of the WASCK and MWASCK methods compared with some popular classification methods. |
| first_indexed | 2024-03-09T00:36:57Z |
| format | Article |
| id | doaj.art-e2babd02dfbf4a54996954c122046d5a |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T00:36:57Z |
| publishDate | 2021-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-e2babd02dfbf4a54996954c122046d5a2023-12-11T18:08:26ZengMDPI AGRemote Sensing2072-42922021-02-0113482010.3390/rs13040820Multiscale Weighted Adjacent Superpixel-Based Composite Kernel for Hyperspectral Image ClassificationYaokang Zhang0Yunjie Chen1School of Mathematics and Statistics, Nanjing University of Information Science & Technology, Nanjing 210044, ChinaSchool of Mathematics and Statistics, Nanjing University of Information Science & Technology, Nanjing 210044, ChinaThis paper presents a composite kernel method (MWASCK) based on multiscale weighted adjacent superpixels (ASs) to classify hyperspectral image (HSI). The MWASCK adequately exploits spatial-spectral features of weighted adjacent superpixels to guarantee that more accurate spectral features can be extracted. Firstly, we use a superpixel segmentation algorithm to divide HSI into multiple superpixels. Secondly, the similarities between each target superpixel and its ASs are calculated to construct the spatial features. Finally, a weighted AS-based composite kernel (WASCK) method for HSI classification is proposed. In order to avoid seeking for the optimal superpixel scale and fuse the multiscale spatial features, the MWASCK method uses multiscale weighted superpixel neighbor information. Experiments from two real HSIs indicate that superior performance of the WASCK and MWASCK methods compared with some popular classification methods.https://www.mdpi.com/2072-4292/13/4/820hyperspectral image (HSI)multiscale superpixelspectral-spatial classificationcomposite kernel |
| spellingShingle | Yaokang Zhang Yunjie Chen Multiscale Weighted Adjacent Superpixel-Based Composite Kernel for Hyperspectral Image Classification Remote Sensing hyperspectral image (HSI) multiscale superpixel spectral-spatial classification composite kernel |
| title | Multiscale Weighted Adjacent Superpixel-Based Composite Kernel for Hyperspectral Image Classification |
| title_full | Multiscale Weighted Adjacent Superpixel-Based Composite Kernel for Hyperspectral Image Classification |
| title_fullStr | Multiscale Weighted Adjacent Superpixel-Based Composite Kernel for Hyperspectral Image Classification |
| title_full_unstemmed | Multiscale Weighted Adjacent Superpixel-Based Composite Kernel for Hyperspectral Image Classification |
| title_short | Multiscale Weighted Adjacent Superpixel-Based Composite Kernel for Hyperspectral Image Classification |
| title_sort | multiscale weighted adjacent superpixel based composite kernel for hyperspectral image classification |
| topic | hyperspectral image (HSI) multiscale superpixel spectral-spatial classification composite kernel |
| url | https://www.mdpi.com/2072-4292/13/4/820 |
| work_keys_str_mv | AT yaokangzhang multiscaleweightedadjacentsuperpixelbasedcompositekernelforhyperspectralimageclassification AT yunjiechen multiscaleweightedadjacentsuperpixelbasedcompositekernelforhyperspectralimageclassification |