Explainable Two-Layer Mode Machine Learning Method for Hyperspectral Image Classification
Explainable machine learning methods with a specific mathematical model provide insights into how the model works. We propose a new mode that contains a two-layer architecture for hyperspectral image (HSI) classification. In the front-end learning layer, superpixel segmentation and mathematical mode...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-05-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/11/5859 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849722546634620928 |
|---|---|
| author | Wenjia Chen Junwei Cheng Song Yang Li Sun |
| author_facet | Wenjia Chen Junwei Cheng Song Yang Li Sun |
| author_sort | Wenjia Chen |
| collection | DOAJ |
| description | Explainable machine learning methods with a specific mathematical model provide insights into how the model works. We propose a new mode that contains a two-layer architecture for hyperspectral image (HSI) classification. In the front-end learning layer, superpixel segmentation and mathematical models are combined to achieve the band selection, which obtains the data re-expression in a lower dimension. The mathematical model uses the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mrow><mi>l</mi></mrow><mrow><mn>2,1</mn></mrow></msub></mrow></semantics></math></inline-formula> norm and graph regularized term, which helps induce sparsity, improve robustness to outliers and noise, and enhance the explainability of the data re-expression. We employ the support vector machine or the K-nearest neighbor algorithms in the back-end layer to classify low-dimensional data. Finally, the two-layer mode classification method is applied to the three real HSI dataset classifications. Numerical results show that the overall classification accuracy of our method is improved. |
| format | Article |
| id | doaj-art-b0531e974cd74e79b0d21ff0b1b28864 |
| institution | DOAJ |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-b0531e974cd74e79b0d21ff0b1b288642025-08-20T03:11:18ZengMDPI AGApplied Sciences2076-34172025-05-011511585910.3390/app15115859Explainable Two-Layer Mode Machine Learning Method for Hyperspectral Image ClassificationWenjia Chen0Junwei Cheng1Song Yang2Li Sun3College of Information Science and Engineering, Shandong Agricultural University, Tai’an 271018, ChinaCollege of Information Science and Engineering, Shandong Agricultural University, Tai’an 271018, ChinaCollege of Information Science and Engineering, Shandong Agricultural University, Tai’an 271018, ChinaCollege of Information Science and Engineering, Shandong Agricultural University, Tai’an 271018, ChinaExplainable machine learning methods with a specific mathematical model provide insights into how the model works. We propose a new mode that contains a two-layer architecture for hyperspectral image (HSI) classification. In the front-end learning layer, superpixel segmentation and mathematical models are combined to achieve the band selection, which obtains the data re-expression in a lower dimension. The mathematical model uses the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mrow><mi>l</mi></mrow><mrow><mn>2,1</mn></mrow></msub></mrow></semantics></math></inline-formula> norm and graph regularized term, which helps induce sparsity, improve robustness to outliers and noise, and enhance the explainability of the data re-expression. We employ the support vector machine or the K-nearest neighbor algorithms in the back-end layer to classify low-dimensional data. Finally, the two-layer mode classification method is applied to the three real HSI dataset classifications. Numerical results show that the overall classification accuracy of our method is improved.https://www.mdpi.com/2076-3417/15/11/5859land coverland usesuperpixel segmentationsparse space clustering<i>l</i><sub>2,1</sub> normband selection |
| spellingShingle | Wenjia Chen Junwei Cheng Song Yang Li Sun Explainable Two-Layer Mode Machine Learning Method for Hyperspectral Image Classification Applied Sciences land cover land use superpixel segmentation sparse space clustering <i>l</i><sub>2,1</sub> norm band selection |
| title | Explainable Two-Layer Mode Machine Learning Method for Hyperspectral Image Classification |
| title_full | Explainable Two-Layer Mode Machine Learning Method for Hyperspectral Image Classification |
| title_fullStr | Explainable Two-Layer Mode Machine Learning Method for Hyperspectral Image Classification |
| title_full_unstemmed | Explainable Two-Layer Mode Machine Learning Method for Hyperspectral Image Classification |
| title_short | Explainable Two-Layer Mode Machine Learning Method for Hyperspectral Image Classification |
| title_sort | explainable two layer mode machine learning method for hyperspectral image classification |
| topic | land cover land use superpixel segmentation sparse space clustering <i>l</i><sub>2,1</sub> norm band selection |
| url | https://www.mdpi.com/2076-3417/15/11/5859 |
| work_keys_str_mv | AT wenjiachen explainabletwolayermodemachinelearningmethodforhyperspectralimageclassification AT junweicheng explainabletwolayermodemachinelearningmethodforhyperspectralimageclassification AT songyang explainabletwolayermodemachinelearningmethodforhyperspectralimageclassification AT lisun explainabletwolayermodemachinelearningmethodforhyperspectralimageclassification |