HDTFF-Net: Hierarchical Deep Texture Features Fusion Network for High-Resolution Remote Sensing Scene Classification
Fusing features from different feature descriptors or different convolutional layers can improve the understanding of scene and enhance the classification accuracy. In this article, we propose a hierarchical deep texture feature fusion network, abbreviated as HDTFF-Net, aiming to improve the classif...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10194293/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849340581892849664 |
|---|---|
| author | Wanying Song Yifan Cong Shiru Zhang Yan Wu Peng Zhang |
| author_facet | Wanying Song Yifan Cong Shiru Zhang Yan Wu Peng Zhang |
| author_sort | Wanying Song |
| collection | DOAJ |
| description | Fusing features from different feature descriptors or different convolutional layers can improve the understanding of scene and enhance the classification accuracy. In this article, we propose a hierarchical deep texture feature fusion network, abbreviated as HDTFF-Net, aiming to improve the classification accuracy of high-resolution remote sensing scene classification. The proposed HDTFF-Net can effectively combine the shallow texture information from manual features and the deep texture information by convolutional neural networks (CNNs). First, for deeply excavating the multiscale and multidirectional shallow texture features in images, an improved Wavelet feature extraction module and a Gabor feature extraction module are designed by fully fusing the structural features into the backbone neural network. Then, to make the output texture features more discriminative and interpretative, we incorporate the above texture feature extraction modules into traditional CNNs (Tra-CNNs), and design two improved deep networks, namely Wave-CNN and Gabor-CNN. Finally, according to the Dempster-Shafer evidence theory, the designed Wave-CNN and Gabor-CNN are fused with the Tra-CNN by a decision-level fusion strategy, which can effectively capture the deep texture features by different feature descriptors and improve the classification performance. Experiments on high-resolution remote sensing images demonstrate the effectiveness of the proposed HDTFF-Net, and verify that it can greatly improve the classification performance. |
| format | Article |
| id | doaj-art-1b7a7ff63db843efb5f22da13cf52d26 |
| institution | Kabale University |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-1b7a7ff63db843efb5f22da13cf52d262025-08-20T03:43:52ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352023-01-01167327734210.1109/JSTARS.2023.329849210194293HDTFF-Net: Hierarchical Deep Texture Features Fusion Network for High-Resolution Remote Sensing Scene ClassificationWanying Song0https://orcid.org/0000-0002-3777-067XYifan Cong1https://orcid.org/0009-0000-1047-6967Shiru Zhang2https://orcid.org/0000-0002-1677-2721Yan Wu3https://orcid.org/0000-0001-7502-2341Peng Zhang4https://orcid.org/0000-0002-8065-0948Xi'an Key Laboratory of Network Convergence Communication, School of Communication and Information Engineering, Xi'an University of Science and Technology, Xi'an, ChinaXi'an Key Laboratory of Network Convergence Communication, School of Communication and Information Engineering, Xi'an University of Science and Technology, Xi'an, ChinaXi'an Key Laboratory of Network Convergence Communication, School of Communication and Information Engineering, Xi'an University of Science and Technology, Xi'an, ChinaSchool of Electronics Engineering, Xidian University, Xi'an, ChinaSchool of Electronics Engineering, Xidian University, Xi'an, ChinaFusing features from different feature descriptors or different convolutional layers can improve the understanding of scene and enhance the classification accuracy. In this article, we propose a hierarchical deep texture feature fusion network, abbreviated as HDTFF-Net, aiming to improve the classification accuracy of high-resolution remote sensing scene classification. The proposed HDTFF-Net can effectively combine the shallow texture information from manual features and the deep texture information by convolutional neural networks (CNNs). First, for deeply excavating the multiscale and multidirectional shallow texture features in images, an improved Wavelet feature extraction module and a Gabor feature extraction module are designed by fully fusing the structural features into the backbone neural network. Then, to make the output texture features more discriminative and interpretative, we incorporate the above texture feature extraction modules into traditional CNNs (Tra-CNNs), and design two improved deep networks, namely Wave-CNN and Gabor-CNN. Finally, according to the Dempster-Shafer evidence theory, the designed Wave-CNN and Gabor-CNN are fused with the Tra-CNN by a decision-level fusion strategy, which can effectively capture the deep texture features by different feature descriptors and improve the classification performance. Experiments on high-resolution remote sensing images demonstrate the effectiveness of the proposed HDTFF-Net, and verify that it can greatly improve the classification performance.https://ieeexplore.ieee.org/document/10194293/Convolutional neural network (CNN)deep Gabor featuresdeep wavelet featuresDempster-Shafer (D-S) evidential theoryremote sensing scene classification (RSSC) |
| spellingShingle | Wanying Song Yifan Cong Shiru Zhang Yan Wu Peng Zhang HDTFF-Net: Hierarchical Deep Texture Features Fusion Network for High-Resolution Remote Sensing Scene Classification IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Convolutional neural network (CNN) deep Gabor features deep wavelet features Dempster-Shafer (D-S) evidential theory remote sensing scene classification (RSSC) |
| title | HDTFF-Net: Hierarchical Deep Texture Features Fusion Network for High-Resolution Remote Sensing Scene Classification |
| title_full | HDTFF-Net: Hierarchical Deep Texture Features Fusion Network for High-Resolution Remote Sensing Scene Classification |
| title_fullStr | HDTFF-Net: Hierarchical Deep Texture Features Fusion Network for High-Resolution Remote Sensing Scene Classification |
| title_full_unstemmed | HDTFF-Net: Hierarchical Deep Texture Features Fusion Network for High-Resolution Remote Sensing Scene Classification |
| title_short | HDTFF-Net: Hierarchical Deep Texture Features Fusion Network for High-Resolution Remote Sensing Scene Classification |
| title_sort | hdtff net hierarchical deep texture features fusion network for high resolution remote sensing scene classification |
| topic | Convolutional neural network (CNN) deep Gabor features deep wavelet features Dempster-Shafer (D-S) evidential theory remote sensing scene classification (RSSC) |
| url | https://ieeexplore.ieee.org/document/10194293/ |
| work_keys_str_mv | AT wanyingsong hdtffnethierarchicaldeeptexturefeaturesfusionnetworkforhighresolutionremotesensingsceneclassification AT yifancong hdtffnethierarchicaldeeptexturefeaturesfusionnetworkforhighresolutionremotesensingsceneclassification AT shiruzhang hdtffnethierarchicaldeeptexturefeaturesfusionnetworkforhighresolutionremotesensingsceneclassification AT yanwu hdtffnethierarchicaldeeptexturefeaturesfusionnetworkforhighresolutionremotesensingsceneclassification AT pengzhang hdtffnethierarchicaldeeptexturefeaturesfusionnetworkforhighresolutionremotesensingsceneclassification |