Hyperspectral Band Selection with Unique Pixel Extraction and Adaptive Neighbor Clustering
Band selection is an effective way to reduce redundant information, while preserving the physical properties of hyperspectral images (HSI). However, most band selection methods merely consider the relevance and separability between pairs of bands and ignore those for different ground objects. To sol...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/2/315 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587595464310784 |
|---|---|
| author | Bing Han Mingqing Liu Zhenyu Ma Ke Zhang Yanke Xu Jingyu Wang Qi Wang |
| author_facet | Bing Han Mingqing Liu Zhenyu Ma Ke Zhang Yanke Xu Jingyu Wang Qi Wang |
| author_sort | Bing Han |
| collection | DOAJ |
| description | Band selection is an effective way to reduce redundant information, while preserving the physical properties of hyperspectral images (HSI). However, most band selection methods merely consider the relevance and separability between pairs of bands and ignore those for different ground objects. To solve these issues, we propose a Unique Pixel extraction and Adaptive Neighbor Clustering (UPANC) band selection method in this theoretical study. First, in consideration of the characteristics of HSI data and tasks, unique pixels are obtained with a low-rank representation, where the importance of bands is analyzed from both spectral and spatial perspectives. Second, an adaptive neighbor clustering method is designed based on the unique pixels, which groups bands into several clusters through optimizing the graph structure under label smoothness. With support vector machines (SVM) as the classifier, the UPANC method achieved good performance, where the overall accuracy scores were 89.05%, 82.62%, and 92.07% on the Houston, IndianPines, and Pavia University datasets, respectively. The experimental results illustrated the advantages of the UPANC method, which could select optimal bands to enhance the performance in land cover observation. |
| format | Article |
| id | doaj-art-c07e7223ca1845ab834cd5758a333292 |
| institution | Kabale University |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-c07e7223ca1845ab834cd5758a3332922025-01-24T13:48:05ZengMDPI AGRemote Sensing2072-42922025-01-0117231510.3390/rs17020315Hyperspectral Band Selection with Unique Pixel Extraction and Adaptive Neighbor ClusteringBing Han0Mingqing Liu1Zhenyu Ma2Ke Zhang3Yanke Xu4Jingyu Wang5Qi Wang6School of Astronautics, Northwestern Polytechnical University, Xi’an 710072, ChinaSchool of Astronautics, Northwestern Polytechnical University, Xi’an 710072, ChinaSchool of Astronautics, Northwestern Polytechnical University, Xi’an 710072, ChinaSchool of Astronautics, Northwestern Polytechnical University, Xi’an 710072, ChinaNational Key Laboratory of Air-Based Information Perception and Fusion, Luoyang 471000, ChinaSchool of Astronautics, Northwestern Polytechnical University, Xi’an 710072, ChinaSchool of Artificial Intelligence, OPtics and ElectroNics (iOPEN), Northwestern Polytechnical University, Xi’an 710072, ChinaBand selection is an effective way to reduce redundant information, while preserving the physical properties of hyperspectral images (HSI). However, most band selection methods merely consider the relevance and separability between pairs of bands and ignore those for different ground objects. To solve these issues, we propose a Unique Pixel extraction and Adaptive Neighbor Clustering (UPANC) band selection method in this theoretical study. First, in consideration of the characteristics of HSI data and tasks, unique pixels are obtained with a low-rank representation, where the importance of bands is analyzed from both spectral and spatial perspectives. Second, an adaptive neighbor clustering method is designed based on the unique pixels, which groups bands into several clusters through optimizing the graph structure under label smoothness. With support vector machines (SVM) as the classifier, the UPANC method achieved good performance, where the overall accuracy scores were 89.05%, 82.62%, and 92.07% on the Houston, IndianPines, and Pavia University datasets, respectively. The experimental results illustrated the advantages of the UPANC method, which could select optimal bands to enhance the performance in land cover observation.https://www.mdpi.com/2072-4292/17/2/315band selectionunique pixelslow-rank representationspectral clustering |
| spellingShingle | Bing Han Mingqing Liu Zhenyu Ma Ke Zhang Yanke Xu Jingyu Wang Qi Wang Hyperspectral Band Selection with Unique Pixel Extraction and Adaptive Neighbor Clustering Remote Sensing band selection unique pixels low-rank representation spectral clustering |
| title | Hyperspectral Band Selection with Unique Pixel Extraction and Adaptive Neighbor Clustering |
| title_full | Hyperspectral Band Selection with Unique Pixel Extraction and Adaptive Neighbor Clustering |
| title_fullStr | Hyperspectral Band Selection with Unique Pixel Extraction and Adaptive Neighbor Clustering |
| title_full_unstemmed | Hyperspectral Band Selection with Unique Pixel Extraction and Adaptive Neighbor Clustering |
| title_short | Hyperspectral Band Selection with Unique Pixel Extraction and Adaptive Neighbor Clustering |
| title_sort | hyperspectral band selection with unique pixel extraction and adaptive neighbor clustering |
| topic | band selection unique pixels low-rank representation spectral clustering |
| url | https://www.mdpi.com/2072-4292/17/2/315 |
| work_keys_str_mv | AT binghan hyperspectralbandselectionwithuniquepixelextractionandadaptiveneighborclustering AT mingqingliu hyperspectralbandselectionwithuniquepixelextractionandadaptiveneighborclustering AT zhenyuma hyperspectralbandselectionwithuniquepixelextractionandadaptiveneighborclustering AT kezhang hyperspectralbandselectionwithuniquepixelextractionandadaptiveneighborclustering AT yankexu hyperspectralbandselectionwithuniquepixelextractionandadaptiveneighborclustering AT jingyuwang hyperspectralbandselectionwithuniquepixelextractionandadaptiveneighborclustering AT qiwang hyperspectralbandselectionwithuniquepixelextractionandadaptiveneighborclustering |