Orthogonal Capsule Network with Meta-Reinforcement Learning for Small Sample Hyperspectral Image Classification
Most current hyperspectral image classification (HSIC) models require a large number of training samples, and when the sample size is small, the classification performance decreases. To address this issue, we propose an innovative model that combines an orthogonal capsule network with meta-reinforce...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/2/215 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587587383984128 |
|---|---|
| author | Prince Yaw Owusu Amoako Guo Cao Boshan Shi Di Yang Benedict Boakye Acka |
| author_facet | Prince Yaw Owusu Amoako Guo Cao Boshan Shi Di Yang Benedict Boakye Acka |
| author_sort | Prince Yaw Owusu Amoako |
| collection | DOAJ |
| description | Most current hyperspectral image classification (HSIC) models require a large number of training samples, and when the sample size is small, the classification performance decreases. To address this issue, we propose an innovative model that combines an orthogonal capsule network with meta-reinforcement learning (OCN-MRL) for small sample HSIC. The OCN-MRL framework employs Meta-RL for feature selection and CapsNet for classification with a small data sample. The Meta-RL module through clustering, augmentation, and multiview techniques enables the model to adapt to new HSIC tasks with limited samples. Learning a meta-policy with a Q-learner generalizes across different tasks to effectively select discriminative features from the hyperspectral data. Integrating orthogonality into CapsNet reduces the network complexity while maintaining the ability to preserve spatial hierarchies and relationships in the data with a 3D convolution layer, suitably capturing complex patterns. Experimental results on four rich Chinese hyperspectral datasets demonstrate the OCN-MRL model’s competitiveness in both higher classification accuracy and less computational cost compared to existing CapsNet-based methods. |
| format | Article |
| id | doaj-art-9ffb3dd4406e4fa5b0ebc10fee5ff030 |
| institution | Kabale University |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-9ffb3dd4406e4fa5b0ebc10fee5ff0302025-01-24T13:47:46ZengMDPI AGRemote Sensing2072-42922025-01-0117221510.3390/rs17020215Orthogonal Capsule Network with Meta-Reinforcement Learning for Small Sample Hyperspectral Image ClassificationPrince Yaw Owusu Amoako0Guo Cao1Boshan Shi2Di Yang3Benedict Boakye Acka4School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, ChinaSchool of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, ChinaSchool of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, ChinaWyoming Geographic Information Science Center, University of Wyoming, Laramie, WY 82071, USADepartment of Computer Science, Faculty of Science, Valley View University, Accra 82071, GhanaMost current hyperspectral image classification (HSIC) models require a large number of training samples, and when the sample size is small, the classification performance decreases. To address this issue, we propose an innovative model that combines an orthogonal capsule network with meta-reinforcement learning (OCN-MRL) for small sample HSIC. The OCN-MRL framework employs Meta-RL for feature selection and CapsNet for classification with a small data sample. The Meta-RL module through clustering, augmentation, and multiview techniques enables the model to adapt to new HSIC tasks with limited samples. Learning a meta-policy with a Q-learner generalizes across different tasks to effectively select discriminative features from the hyperspectral data. Integrating orthogonality into CapsNet reduces the network complexity while maintaining the ability to preserve spatial hierarchies and relationships in the data with a 3D convolution layer, suitably capturing complex patterns. Experimental results on four rich Chinese hyperspectral datasets demonstrate the OCN-MRL model’s competitiveness in both higher classification accuracy and less computational cost compared to existing CapsNet-based methods.https://www.mdpi.com/2072-4292/17/2/215hyperspectral imagecapsule networkorthogonal layersmall sample classification |
| spellingShingle | Prince Yaw Owusu Amoako Guo Cao Boshan Shi Di Yang Benedict Boakye Acka Orthogonal Capsule Network with Meta-Reinforcement Learning for Small Sample Hyperspectral Image Classification Remote Sensing hyperspectral image capsule network orthogonal layer small sample classification |
| title | Orthogonal Capsule Network with Meta-Reinforcement Learning for Small Sample Hyperspectral Image Classification |
| title_full | Orthogonal Capsule Network with Meta-Reinforcement Learning for Small Sample Hyperspectral Image Classification |
| title_fullStr | Orthogonal Capsule Network with Meta-Reinforcement Learning for Small Sample Hyperspectral Image Classification |
| title_full_unstemmed | Orthogonal Capsule Network with Meta-Reinforcement Learning for Small Sample Hyperspectral Image Classification |
| title_short | Orthogonal Capsule Network with Meta-Reinforcement Learning for Small Sample Hyperspectral Image Classification |
| title_sort | orthogonal capsule network with meta reinforcement learning for small sample hyperspectral image classification |
| topic | hyperspectral image capsule network orthogonal layer small sample classification |
| url | https://www.mdpi.com/2072-4292/17/2/215 |
| work_keys_str_mv | AT princeyawowusuamoako orthogonalcapsulenetworkwithmetareinforcementlearningforsmallsamplehyperspectralimageclassification AT guocao orthogonalcapsulenetworkwithmetareinforcementlearningforsmallsamplehyperspectralimageclassification AT boshanshi orthogonalcapsulenetworkwithmetareinforcementlearningforsmallsamplehyperspectralimageclassification AT diyang orthogonalcapsulenetworkwithmetareinforcementlearningforsmallsamplehyperspectralimageclassification AT benedictboakyeacka orthogonalcapsulenetworkwithmetareinforcementlearningforsmallsamplehyperspectralimageclassification |