An effective feature selection approach based on hybrid Grey Wolf Optimizer and Genetic Algorithm for hyperspectral image
Abstract Feature selection (FS) is a critical step in hyperspectral image (HSI) classification, essential for reducing data dimensionality while preserving classification accuracy. However, FS for HSIs remains an NP-hard challenge, as existing swarm intelligence and evolutionary algorithms (SIEAs) o...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-024-84934-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832594808795824128 |
|---|---|
| author | Yiqun Shang Minrui Zheng Jiayang Li Xinqi Zheng |
| author_facet | Yiqun Shang Minrui Zheng Jiayang Li Xinqi Zheng |
| author_sort | Yiqun Shang |
| collection | DOAJ |
| description | Abstract Feature selection (FS) is a critical step in hyperspectral image (HSI) classification, essential for reducing data dimensionality while preserving classification accuracy. However, FS for HSIs remains an NP-hard challenge, as existing swarm intelligence and evolutionary algorithms (SIEAs) often suffer from limited exploration capabilities or susceptibility to local optima, particularly in high-dimensional scenarios. To address these challenges, we propose GWOGA, a novel hybrid algorithm that combines Grey Wolf Optimizer (GWO) and Genetic Algorithm (GA), aiming to achieve an effective balance between exploration and exploitation. The innovation of GWOGA lies in three core strategies: (1) chaotic map and Opposition-Based Learning (OBL) for uniformly distributed population initialization, enhancing diversity and mitigating premature convergence; (2) elite learning strategy to prioritize high-ranking solutions, strengthening the search hierarchy and efficiency; and (3) a hybrid optimization mechanism where GWO ensures rapid early-stage convergence, while GA refines global search in later stages to escape local optima. Experiments on three benchmark HSIs (i.e., Indian Pines, KSC, and Botswana) demonstrate that GWOGA outperforms state-of-the-art algorithms, achieving higher classification accuracy with fewer selected bands. The results highlight GWOGA’s robustness, generalizability, and potential for real-world applications in HSI FS. |
| format | Article |
| id | doaj-art-0dbd3ae490c542aaba7a53371a31d26d |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-0dbd3ae490c542aaba7a53371a31d26d2025-01-19T12:18:29ZengNature PortfolioScientific Reports2045-23222025-01-0115112310.1038/s41598-024-84934-8An effective feature selection approach based on hybrid Grey Wolf Optimizer and Genetic Algorithm for hyperspectral imageYiqun Shang0Minrui Zheng1Jiayang Li2Xinqi Zheng3School of Information Engineering, China University of GeosciencesSchool of Public Administration and Policy, Renmin University of ChinaChengdu Institute of Survey and InvestigationSchool of Information Engineering, China University of GeosciencesAbstract Feature selection (FS) is a critical step in hyperspectral image (HSI) classification, essential for reducing data dimensionality while preserving classification accuracy. However, FS for HSIs remains an NP-hard challenge, as existing swarm intelligence and evolutionary algorithms (SIEAs) often suffer from limited exploration capabilities or susceptibility to local optima, particularly in high-dimensional scenarios. To address these challenges, we propose GWOGA, a novel hybrid algorithm that combines Grey Wolf Optimizer (GWO) and Genetic Algorithm (GA), aiming to achieve an effective balance between exploration and exploitation. The innovation of GWOGA lies in three core strategies: (1) chaotic map and Opposition-Based Learning (OBL) for uniformly distributed population initialization, enhancing diversity and mitigating premature convergence; (2) elite learning strategy to prioritize high-ranking solutions, strengthening the search hierarchy and efficiency; and (3) a hybrid optimization mechanism where GWO ensures rapid early-stage convergence, while GA refines global search in later stages to escape local optima. Experiments on three benchmark HSIs (i.e., Indian Pines, KSC, and Botswana) demonstrate that GWOGA outperforms state-of-the-art algorithms, achieving higher classification accuracy with fewer selected bands. The results highlight GWOGA’s robustness, generalizability, and potential for real-world applications in HSI FS.https://doi.org/10.1038/s41598-024-84934-8 |
| spellingShingle | Yiqun Shang Minrui Zheng Jiayang Li Xinqi Zheng An effective feature selection approach based on hybrid Grey Wolf Optimizer and Genetic Algorithm for hyperspectral image Scientific Reports |
| title | An effective feature selection approach based on hybrid Grey Wolf Optimizer and Genetic Algorithm for hyperspectral image |
| title_full | An effective feature selection approach based on hybrid Grey Wolf Optimizer and Genetic Algorithm for hyperspectral image |
| title_fullStr | An effective feature selection approach based on hybrid Grey Wolf Optimizer and Genetic Algorithm for hyperspectral image |
| title_full_unstemmed | An effective feature selection approach based on hybrid Grey Wolf Optimizer and Genetic Algorithm for hyperspectral image |
| title_short | An effective feature selection approach based on hybrid Grey Wolf Optimizer and Genetic Algorithm for hyperspectral image |
| title_sort | effective feature selection approach based on hybrid grey wolf optimizer and genetic algorithm for hyperspectral image |
| url | https://doi.org/10.1038/s41598-024-84934-8 |
| work_keys_str_mv | AT yiqunshang aneffectivefeatureselectionapproachbasedonhybridgreywolfoptimizerandgeneticalgorithmforhyperspectralimage AT minruizheng aneffectivefeatureselectionapproachbasedonhybridgreywolfoptimizerandgeneticalgorithmforhyperspectralimage AT jiayangli aneffectivefeatureselectionapproachbasedonhybridgreywolfoptimizerandgeneticalgorithmforhyperspectralimage AT xinqizheng aneffectivefeatureselectionapproachbasedonhybridgreywolfoptimizerandgeneticalgorithmforhyperspectralimage AT yiqunshang effectivefeatureselectionapproachbasedonhybridgreywolfoptimizerandgeneticalgorithmforhyperspectralimage AT minruizheng effectivefeatureselectionapproachbasedonhybridgreywolfoptimizerandgeneticalgorithmforhyperspectralimage AT jiayangli effectivefeatureselectionapproachbasedonhybridgreywolfoptimizerandgeneticalgorithmforhyperspectralimage AT xinqizheng effectivefeatureselectionapproachbasedonhybridgreywolfoptimizerandgeneticalgorithmforhyperspectralimage |