Collaborative Superpixelwised PCA for Hyperspectral Image Classification
Extracting spectral-spatial features from Hyperspectral imagery (HSI) has been proven to be efficient for classification tasks. A recently developed superpixelwised principal component analysis (PCA) (SuperPCA), which has shown its promising performance, is a prominent technique in spectral ...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10816304/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841546167508271104 |
|---|---|
| author | Chao Yao Junrui Gu Zehua Guo Miao Ma Qingrui Guo Gong Cheng |
| author_facet | Chao Yao Junrui Gu Zehua Guo Miao Ma Qingrui Guo Gong Cheng |
| author_sort | Chao Yao |
| collection | DOAJ |
| description | Extracting spectral-spatial features from Hyperspectral imagery (HSI) has been proven to be efficient for classification tasks. A recently developed superpixelwised principal component analysis (PCA) (SuperPCA), which has shown its promising performance, is a prominent technique in spectral–spatial feature extraction. However, we have discovered that SuperPCA may lead to an intraclass dispersion problem, which can result in a decrease in classification accuracy. In this article, a novel method called collaborative superpixelwised PCA (CSPCA) is proposed to address this issue. The main idea behind CSPCA is to collaboratively learn the projections for each superpixel. Specifically, CSPCA first employs a superpixel segmentation technique to generate superpixels. Next, the mean vectors of samples within each superpixel are utilized to model the manifold structure of the data. Then, a novel objective function is formulated, which aims to simultaneously preserve the obtained manifold structure between superpixels and the structure within each superpixel. To optimize the objective function, the Manopt toolbox is employed in the proposed method. The effectiveness of the proposed approach is validated through experimental evaluations conducted on five HSI datasets. |
| format | Article |
| id | doaj-art-d78f2d862cb34665b740284a7650f3e8 |
| institution | Kabale University |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-d78f2d862cb34665b740284a7650f3e82025-01-11T00:00:33ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01182589260110.1109/JSTARS.2024.352096010816304Collaborative Superpixelwised PCA for Hyperspectral Image ClassificationChao Yao0https://orcid.org/0000-0003-0988-6349Junrui Gu1Zehua Guo2Miao Ma3https://orcid.org/0000-0001-7735-6695Qingrui Guo4Gong Cheng5https://orcid.org/0000-0001-5030-0683School of Computer Science, Shaanxi Normal University, Xi'an, ChinaSchool of Computer Science, Shaanxi Normal University, Xi'an, ChinaSchool of Automation, Beijing Institute of Technology, Beijing, ChinaSchool of Computer Science, Shaanxi Normal University, Xi'an, ChinaState Grid Xinjiang Electric Power Research Institute, Urumqi, ChinaSchool of Automation, Northwestern Polytechnical University, Xi'an, ChinaExtracting spectral-spatial features from Hyperspectral imagery (HSI) has been proven to be efficient for classification tasks. A recently developed superpixelwised principal component analysis (PCA) (SuperPCA), which has shown its promising performance, is a prominent technique in spectral–spatial feature extraction. However, we have discovered that SuperPCA may lead to an intraclass dispersion problem, which can result in a decrease in classification accuracy. In this article, a novel method called collaborative superpixelwised PCA (CSPCA) is proposed to address this issue. The main idea behind CSPCA is to collaboratively learn the projections for each superpixel. Specifically, CSPCA first employs a superpixel segmentation technique to generate superpixels. Next, the mean vectors of samples within each superpixel are utilized to model the manifold structure of the data. Then, a novel objective function is formulated, which aims to simultaneously preserve the obtained manifold structure between superpixels and the structure within each superpixel. To optimize the objective function, the Manopt toolbox is employed in the proposed method. The effectiveness of the proposed approach is validated through experimental evaluations conducted on five HSI datasets.https://ieeexplore.ieee.org/document/10816304/Dimension reductionhyperspectral imagery (HSI) classificationspectral–spatial feature learningsuperpixelwised PCA |
| spellingShingle | Chao Yao Junrui Gu Zehua Guo Miao Ma Qingrui Guo Gong Cheng Collaborative Superpixelwised PCA for Hyperspectral Image Classification IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Dimension reduction hyperspectral imagery (HSI) classification spectral–spatial feature learning superpixelwised PCA |
| title | Collaborative Superpixelwised PCA for Hyperspectral Image Classification |
| title_full | Collaborative Superpixelwised PCA for Hyperspectral Image Classification |
| title_fullStr | Collaborative Superpixelwised PCA for Hyperspectral Image Classification |
| title_full_unstemmed | Collaborative Superpixelwised PCA for Hyperspectral Image Classification |
| title_short | Collaborative Superpixelwised PCA for Hyperspectral Image Classification |
| title_sort | collaborative superpixelwised pca for hyperspectral image classification |
| topic | Dimension reduction hyperspectral imagery (HSI) classification spectral–spatial feature learning superpixelwised PCA |
| url | https://ieeexplore.ieee.org/document/10816304/ |
| work_keys_str_mv | AT chaoyao collaborativesuperpixelwisedpcaforhyperspectralimageclassification AT junruigu collaborativesuperpixelwisedpcaforhyperspectralimageclassification AT zehuaguo collaborativesuperpixelwisedpcaforhyperspectralimageclassification AT miaoma collaborativesuperpixelwisedpcaforhyperspectralimageclassification AT qingruiguo collaborativesuperpixelwisedpcaforhyperspectralimageclassification AT gongcheng collaborativesuperpixelwisedpcaforhyperspectralimageclassification |