Multiscale Superpixelwise Locality Preserving Projection for Hyperspectral Image Classification
Manifold learning is a powerful dimensionality reduction tool for a hyperspectral image (HSI) classification to relieve the curse of dimensionality and to reveal the intrinsic low-dimensional manifold. However, a specific characteristic of HSIs, i.e., irregular spatial dependency, is not taken into...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-05-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/9/10/2161 |
| _version_ | 1831842872439078912 |
|---|---|
| author | Lin He Xianjun Chen Jun Li Xiaofeng Xie |
| author_facet | Lin He Xianjun Chen Jun Li Xiaofeng Xie |
| author_sort | Lin He |
| collection | DOAJ |
| description | Manifold learning is a powerful dimensionality reduction tool for a hyperspectral image (HSI) classification to relieve the curse of dimensionality and to reveal the intrinsic low-dimensional manifold. However, a specific characteristic of HSIs, i.e., irregular spatial dependency, is not taken into consideration in the method design, which can yield many spatially homogenous subregions in an HSI scence. Conventional manifold learning methods, such as a locality preserving projection (LPP), pursue a unified projection on the entire HSI, while neglecting the local homogeneities on the HSI manifold caused by those spatially homogenous subregions. In this work, we propose a novel multiscale superpixelwise LPP (MSuperLPP) for HSI classification to overcome the challenge. First, we partition an HSI into homogeneous subregions with a multiscale superpixel segmentation. Then, on each scale, subregion specific LPPs and the associated preliminary classifications are performed. Finally, we aggregate the classification results from all scales using a decision fusion strategy to achieve the final result. Experimental results on three real hyperspectral data sets validate the effectiveness of our method. |
| first_indexed | 2024-12-23T06:57:05Z |
| format | Article |
| id | doaj.art-f6dfa2152d0e46e9952a8b5dcfafcdfa |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-12-23T06:57:05Z |
| publishDate | 2019-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-f6dfa2152d0e46e9952a8b5dcfafcdfa2022-12-21T17:56:18ZengMDPI AGApplied Sciences2076-34172019-05-01910216110.3390/app9102161app9102161Multiscale Superpixelwise Locality Preserving Projection for Hyperspectral Image ClassificationLin He0Xianjun Chen1Jun Li2Xiaofeng Xie3School of Automation Science and Engineering, South China University of Technology, Guangzhou 510640, ChinaSchool of Automation Science and Engineering, South China University of Technology, Guangzhou 510640, ChinaGuangdong Provincial Key Laboratory of Urbanization and Geo-Simulation, Center of Integrated Geographic Information Analysis, School of Geography and Planning, Sun Yat-sen University, Guangzhou 510275, ChinaMechanical and Electrical Engineering College, Hainan University, Haikou 570228, ChinaManifold learning is a powerful dimensionality reduction tool for a hyperspectral image (HSI) classification to relieve the curse of dimensionality and to reveal the intrinsic low-dimensional manifold. However, a specific characteristic of HSIs, i.e., irregular spatial dependency, is not taken into consideration in the method design, which can yield many spatially homogenous subregions in an HSI scence. Conventional manifold learning methods, such as a locality preserving projection (LPP), pursue a unified projection on the entire HSI, while neglecting the local homogeneities on the HSI manifold caused by those spatially homogenous subregions. In this work, we propose a novel multiscale superpixelwise LPP (MSuperLPP) for HSI classification to overcome the challenge. First, we partition an HSI into homogeneous subregions with a multiscale superpixel segmentation. Then, on each scale, subregion specific LPPs and the associated preliminary classifications are performed. Finally, we aggregate the classification results from all scales using a decision fusion strategy to achieve the final result. Experimental results on three real hyperspectral data sets validate the effectiveness of our method.https://www.mdpi.com/2076-3417/9/10/2161hyperspectral image manifold learningdimensionality reductionlocal homogeneityirregular spatial dependencymultiscale superpixel segmentationcovariance featureclassification |
| spellingShingle | Lin He Xianjun Chen Jun Li Xiaofeng Xie Multiscale Superpixelwise Locality Preserving Projection for Hyperspectral Image Classification Applied Sciences hyperspectral image manifold learning dimensionality reduction local homogeneity irregular spatial dependency multiscale superpixel segmentation covariance feature classification |
| title | Multiscale Superpixelwise Locality Preserving Projection for Hyperspectral Image Classification |
| title_full | Multiscale Superpixelwise Locality Preserving Projection for Hyperspectral Image Classification |
| title_fullStr | Multiscale Superpixelwise Locality Preserving Projection for Hyperspectral Image Classification |
| title_full_unstemmed | Multiscale Superpixelwise Locality Preserving Projection for Hyperspectral Image Classification |
| title_short | Multiscale Superpixelwise Locality Preserving Projection for Hyperspectral Image Classification |
| title_sort | multiscale superpixelwise locality preserving projection for hyperspectral image classification |
| topic | hyperspectral image manifold learning dimensionality reduction local homogeneity irregular spatial dependency multiscale superpixel segmentation covariance feature classification |
| url | https://www.mdpi.com/2076-3417/9/10/2161 |
| work_keys_str_mv | AT linhe multiscalesuperpixelwiselocalitypreservingprojectionforhyperspectralimageclassification AT xianjunchen multiscalesuperpixelwiselocalitypreservingprojectionforhyperspectralimageclassification AT junli multiscalesuperpixelwiselocalitypreservingprojectionforhyperspectralimageclassification AT xiaofengxie multiscalesuperpixelwiselocalitypreservingprojectionforhyperspectralimageclassification |