A Joint Network of 3D-2D CNN Feature Hierarchy and Pyramidal Residual Model for Hyperspectral Image Classification
Since convolutional neural networks (CNN) can extract deeper features from hyperspectral images, they show good classification performance in the hyperspectral image (HSI) classification task. However, the performance of many CNN models is constrained by the complexity of hyperspectral pictures. The...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10847806/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832576742712147968 |
|---|---|
| author | Hongwei Wei Yufan Wang Yu Sun Jianfeng Zheng Xiaodong Yu |
| author_facet | Hongwei Wei Yufan Wang Yu Sun Jianfeng Zheng Xiaodong Yu |
| author_sort | Hongwei Wei |
| collection | DOAJ |
| description | Since convolutional neural networks (CNN) can extract deeper features from hyperspectral images, they show good classification performance in the hyperspectral image (HSI) classification task. However, the performance of many CNN models is constrained by the complexity of hyperspectral pictures. The typical 2D-CNN method is unable to capture the discriminant spectral-spatial features due to the massive dimensionality of HSI data as well as potential noise and redundancy, while the 3D-CNN method significantly increases the computational burden of the model. Moreover, the gradient problem of the deeper CNN architecture impedes the convergence performance of the neural network, leading to poor classification accuracy. To mitigate these problems, this paper proposes a joint network of Hybrid 3D-2D CNN and pyramid residual (J-NHPR) to realize the classification of HSI. The combined deep pyramid residual network can significantly enhance the performance of the suggested network with the HSI data by progressively improving the diversity of high-level spectral-spatial properties across layers. Our tests, which included eight distinct approaches for classification and three well-known HSI data sets, revealed that our recently built model, J-NHPR, can provide comparative advantage (in terms of classification accuracy) over other HSI classification methods. The source code of the J-NHPR model is available at <uri>https://github.com/Dreamvai/J-NHPR</uri>. |
| format | Article |
| id | doaj-art-278f43a84bee46ffabe0ed2904f8461b |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-278f43a84bee46ffabe0ed2904f8461b2025-01-31T00:00:44ZengIEEEIEEE Access2169-35362025-01-0113172731728410.1109/ACCESS.2025.353201610847806A Joint Network of 3D-2D CNN Feature Hierarchy and Pyramidal Residual Model for Hyperspectral Image ClassificationHongwei Wei0Yufan Wang1https://orcid.org/0009-0008-4744-8984Yu Sun2Jianfeng Zheng3Xiaodong Yu4https://orcid.org/0000-0003-3233-0840College of Computer Science and Information Engineering, Harbin Normal University, Harbin, ChinaCollege of Computer Science and Information Engineering, Harbin Normal University, Harbin, ChinaDepartment of Municipal and Environmental Engineering, Heilongjiang Institute of Construction Technology, Harbin, ChinaCollege of Computer Science and Information Engineering, Harbin Normal University, Harbin, ChinaCollege of Computer Science and Information Engineering, Harbin Normal University, Harbin, ChinaSince convolutional neural networks (CNN) can extract deeper features from hyperspectral images, they show good classification performance in the hyperspectral image (HSI) classification task. However, the performance of many CNN models is constrained by the complexity of hyperspectral pictures. The typical 2D-CNN method is unable to capture the discriminant spectral-spatial features due to the massive dimensionality of HSI data as well as potential noise and redundancy, while the 3D-CNN method significantly increases the computational burden of the model. Moreover, the gradient problem of the deeper CNN architecture impedes the convergence performance of the neural network, leading to poor classification accuracy. To mitigate these problems, this paper proposes a joint network of Hybrid 3D-2D CNN and pyramid residual (J-NHPR) to realize the classification of HSI. The combined deep pyramid residual network can significantly enhance the performance of the suggested network with the HSI data by progressively improving the diversity of high-level spectral-spatial properties across layers. Our tests, which included eight distinct approaches for classification and three well-known HSI data sets, revealed that our recently built model, J-NHPR, can provide comparative advantage (in terms of classification accuracy) over other HSI classification methods. The source code of the J-NHPR model is available at <uri>https://github.com/Dreamvai/J-NHPR</uri>.https://ieeexplore.ieee.org/document/10847806/CNNs2D-CNN3D-CNNHSIResNet |
| spellingShingle | Hongwei Wei Yufan Wang Yu Sun Jianfeng Zheng Xiaodong Yu A Joint Network of 3D-2D CNN Feature Hierarchy and Pyramidal Residual Model for Hyperspectral Image Classification IEEE Access CNNs 2D-CNN 3D-CNN HSI ResNet |
| title | A Joint Network of 3D-2D CNN Feature Hierarchy and Pyramidal Residual Model for Hyperspectral Image Classification |
| title_full | A Joint Network of 3D-2D CNN Feature Hierarchy and Pyramidal Residual Model for Hyperspectral Image Classification |
| title_fullStr | A Joint Network of 3D-2D CNN Feature Hierarchy and Pyramidal Residual Model for Hyperspectral Image Classification |
| title_full_unstemmed | A Joint Network of 3D-2D CNN Feature Hierarchy and Pyramidal Residual Model for Hyperspectral Image Classification |
| title_short | A Joint Network of 3D-2D CNN Feature Hierarchy and Pyramidal Residual Model for Hyperspectral Image Classification |
| title_sort | joint network of 3d 2d cnn feature hierarchy and pyramidal residual model for hyperspectral image classification |
| topic | CNNs 2D-CNN 3D-CNN HSI ResNet |
| url | https://ieeexplore.ieee.org/document/10847806/ |
| work_keys_str_mv | AT hongweiwei ajointnetworkof3d2dcnnfeaturehierarchyandpyramidalresidualmodelforhyperspectralimageclassification AT yufanwang ajointnetworkof3d2dcnnfeaturehierarchyandpyramidalresidualmodelforhyperspectralimageclassification AT yusun ajointnetworkof3d2dcnnfeaturehierarchyandpyramidalresidualmodelforhyperspectralimageclassification AT jianfengzheng ajointnetworkof3d2dcnnfeaturehierarchyandpyramidalresidualmodelforhyperspectralimageclassification AT xiaodongyu ajointnetworkof3d2dcnnfeaturehierarchyandpyramidalresidualmodelforhyperspectralimageclassification AT hongweiwei jointnetworkof3d2dcnnfeaturehierarchyandpyramidalresidualmodelforhyperspectralimageclassification AT yufanwang jointnetworkof3d2dcnnfeaturehierarchyandpyramidalresidualmodelforhyperspectralimageclassification AT yusun jointnetworkof3d2dcnnfeaturehierarchyandpyramidalresidualmodelforhyperspectralimageclassification AT jianfengzheng jointnetworkof3d2dcnnfeaturehierarchyandpyramidalresidualmodelforhyperspectralimageclassification AT xiaodongyu jointnetworkof3d2dcnnfeaturehierarchyandpyramidalresidualmodelforhyperspectralimageclassification |