Multiple Kernel-Based SVM Classification of Hyperspectral Images by Combining Spectral, Spatial, and Semantic Information
In this study, we present a hyperspectral image classification method by combining spectral, spatial, and semantic information. The main steps of the proposed method are summarized as follows: First, principal component analysis transform is conducted on an original image to produce its extended mor...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/1/120 |
| _version_ | 1818085745373478912 |
|---|---|
| author | Yi Wang Wenke Yu Zhice Fang |
| author_facet | Yi Wang Wenke Yu Zhice Fang |
| author_sort | Yi Wang |
| collection | DOAJ |
| description | In this study, we present a hyperspectral image classification method by combining spectral, spatial, and semantic information. The main steps of the proposed method are summarized as follows: First, principal component analysis transform is conducted on an original image to produce its extended morphological profile, Gabor features, and superpixel-based segmentation map. To model spatial information, the extended morphological profile and Gabor features are used to represent structure and texture features, respectively. Moreover, the mean filtering is performed within each superpixel to maintain the homogeneity of the spatial features. Then, the k-means clustering and the entropy rate superpixel segmentation are combined to produce semantic feature vectors by using a bag of visual-words model for each superpixel. Next, three kernel functions are constructed to describe the spectral, spatial, and semantic information, respectively. Finally, the composite kernel technique is used to fuse all the features into a multiple kernel function that is fed into a support vector machine classifier to produce a final classification map. Experiments demonstrate that the proposed method is superior to the most popular kernel-based classification methods in terms of both visual inspection and quantitative analysis, even if only very limited training samples are available. |
| first_indexed | 2024-12-10T20:14:40Z |
| format | Article |
| id | doaj.art-30934ed3a18e4808aee8bf7b4875a053 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-12-10T20:14:40Z |
| publishDate | 2020-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-30934ed3a18e4808aee8bf7b4875a0532022-12-22T01:35:13ZengMDPI AGRemote Sensing2072-42922020-01-0112112010.3390/rs12010120rs12010120Multiple Kernel-Based SVM Classification of Hyperspectral Images by Combining Spectral, Spatial, and Semantic InformationYi Wang0Wenke Yu1Zhice Fang2Institute of Geophysics and Geomatics, China University of Geosciences, Wuhan 430074, ChinaInstitute of Geophysics and Geomatics, China University of Geosciences, Wuhan 430074, ChinaInstitute of Geophysics and Geomatics, China University of Geosciences, Wuhan 430074, ChinaIn this study, we present a hyperspectral image classification method by combining spectral, spatial, and semantic information. The main steps of the proposed method are summarized as follows: First, principal component analysis transform is conducted on an original image to produce its extended morphological profile, Gabor features, and superpixel-based segmentation map. To model spatial information, the extended morphological profile and Gabor features are used to represent structure and texture features, respectively. Moreover, the mean filtering is performed within each superpixel to maintain the homogeneity of the spatial features. Then, the k-means clustering and the entropy rate superpixel segmentation are combined to produce semantic feature vectors by using a bag of visual-words model for each superpixel. Next, three kernel functions are constructed to describe the spectral, spatial, and semantic information, respectively. Finally, the composite kernel technique is used to fuse all the features into a multiple kernel function that is fed into a support vector machine classifier to produce a final classification map. Experiments demonstrate that the proposed method is superior to the most popular kernel-based classification methods in terms of both visual inspection and quantitative analysis, even if only very limited training samples are available.https://www.mdpi.com/2072-4292/12/1/120hyperspectral imagesclassificationspectral-spatialmultiple kernelssemantic information |
| spellingShingle | Yi Wang Wenke Yu Zhice Fang Multiple Kernel-Based SVM Classification of Hyperspectral Images by Combining Spectral, Spatial, and Semantic Information Remote Sensing hyperspectral images classification spectral-spatial multiple kernels semantic information |
| title | Multiple Kernel-Based SVM Classification of Hyperspectral Images by Combining Spectral, Spatial, and Semantic Information |
| title_full | Multiple Kernel-Based SVM Classification of Hyperspectral Images by Combining Spectral, Spatial, and Semantic Information |
| title_fullStr | Multiple Kernel-Based SVM Classification of Hyperspectral Images by Combining Spectral, Spatial, and Semantic Information |
| title_full_unstemmed | Multiple Kernel-Based SVM Classification of Hyperspectral Images by Combining Spectral, Spatial, and Semantic Information |
| title_short | Multiple Kernel-Based SVM Classification of Hyperspectral Images by Combining Spectral, Spatial, and Semantic Information |
| title_sort | multiple kernel based svm classification of hyperspectral images by combining spectral spatial and semantic information |
| topic | hyperspectral images classification spectral-spatial multiple kernels semantic information |
| url | https://www.mdpi.com/2072-4292/12/1/120 |
| work_keys_str_mv | AT yiwang multiplekernelbasedsvmclassificationofhyperspectralimagesbycombiningspectralspatialandsemanticinformation AT wenkeyu multiplekernelbasedsvmclassificationofhyperspectralimagesbycombiningspectralspatialandsemanticinformation AT zhicefang multiplekernelbasedsvmclassificationofhyperspectralimagesbycombiningspectralspatialandsemanticinformation |