A Dual-Model Architecture with Grouping-Attention-Fusion for Remote Sensing Scene Classification
Remote sensing images contain complex backgrounds and multi-scale objects, which pose a challenging task for scene classification. The performance is highly dependent on the capacity of the scene representation as well as the discriminability of the classifier. Although multiple models possess bette...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/3/433 |
| _version_ | 1827597652259241984 |
|---|---|
| author | Junge Shen Tong Zhang Yichen Wang Ruxin Wang Qi Wang Min Qi |
| author_facet | Junge Shen Tong Zhang Yichen Wang Ruxin Wang Qi Wang Min Qi |
| author_sort | Junge Shen |
| collection | DOAJ |
| description | Remote sensing images contain complex backgrounds and multi-scale objects, which pose a challenging task for scene classification. The performance is highly dependent on the capacity of the scene representation as well as the discriminability of the classifier. Although multiple models possess better properties than a single model on these aspects, the fusion strategy for these models is a key component to maximize the final accuracy. In this paper, we construct a novel dual-model architecture with a grouping-attention-fusion strategy to improve the performance of scene classification. Specifically, the model employs two different convolutional neural networks (CNNs) for feature extraction, where the grouping-attention-fusion strategy is used to fuse the features of the CNNs in a fine and multi-scale manner. In this way, the resultant feature representation of the scene is enhanced. Moreover, to address the issue of similar appearances between different scenes, we develop a loss function which encourages small intra-class diversities and large inter-class distances. Extensive experiments are conducted on four scene classification datasets include the UCM land-use dataset, the WHU-RS19 dataset, the AID dataset, and the OPTIMAL-31 dataset. The experimental results demonstrate the superiority of the proposed method in comparison with the state-of-the-arts. |
| first_indexed | 2024-03-09T03:36:22Z |
| format | Article |
| id | doaj.art-50d6772456fd4f098b332a03f5a2a559 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T03:36:22Z |
| publishDate | 2021-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-50d6772456fd4f098b332a03f5a2a5592023-12-03T14:47:38ZengMDPI AGRemote Sensing2072-42922021-01-0113343310.3390/rs13030433A Dual-Model Architecture with Grouping-Attention-Fusion for Remote Sensing Scene ClassificationJunge Shen0Tong Zhang1Yichen Wang2Ruxin Wang3Qi Wang4Min Qi5Unmanned System Research Institute, Northwestern Polytechnical University, Xi’an 710072, ChinaUnmanned System Research Institute, Northwestern Polytechnical University, Xi’an 710072, ChinaUnmanned System Research Institute, Northwestern Polytechnical University, Xi’an 710072, ChinaNational Pilot School of Software, Yunnan University, Kunming 650504, ChinaUnmanned System Research Institute, Northwestern Polytechnical University, Xi’an 710072, ChinaSchool of Electronics and Information, Northwestern Polytechnical University, Xi’an 710072, ChinaRemote sensing images contain complex backgrounds and multi-scale objects, which pose a challenging task for scene classification. The performance is highly dependent on the capacity of the scene representation as well as the discriminability of the classifier. Although multiple models possess better properties than a single model on these aspects, the fusion strategy for these models is a key component to maximize the final accuracy. In this paper, we construct a novel dual-model architecture with a grouping-attention-fusion strategy to improve the performance of scene classification. Specifically, the model employs two different convolutional neural networks (CNNs) for feature extraction, where the grouping-attention-fusion strategy is used to fuse the features of the CNNs in a fine and multi-scale manner. In this way, the resultant feature representation of the scene is enhanced. Moreover, to address the issue of similar appearances between different scenes, we develop a loss function which encourages small intra-class diversities and large inter-class distances. Extensive experiments are conducted on four scene classification datasets include the UCM land-use dataset, the WHU-RS19 dataset, the AID dataset, and the OPTIMAL-31 dataset. The experimental results demonstrate the superiority of the proposed method in comparison with the state-of-the-arts.https://www.mdpi.com/2072-4292/13/3/433remote sensingdual-model architecturegrouping-attention-fusionscene classification |
| spellingShingle | Junge Shen Tong Zhang Yichen Wang Ruxin Wang Qi Wang Min Qi A Dual-Model Architecture with Grouping-Attention-Fusion for Remote Sensing Scene Classification Remote Sensing remote sensing dual-model architecture grouping-attention-fusion scene classification |
| title | A Dual-Model Architecture with Grouping-Attention-Fusion for Remote Sensing Scene Classification |
| title_full | A Dual-Model Architecture with Grouping-Attention-Fusion for Remote Sensing Scene Classification |
| title_fullStr | A Dual-Model Architecture with Grouping-Attention-Fusion for Remote Sensing Scene Classification |
| title_full_unstemmed | A Dual-Model Architecture with Grouping-Attention-Fusion for Remote Sensing Scene Classification |
| title_short | A Dual-Model Architecture with Grouping-Attention-Fusion for Remote Sensing Scene Classification |
| title_sort | dual model architecture with grouping attention fusion for remote sensing scene classification |
| topic | remote sensing dual-model architecture grouping-attention-fusion scene classification |
| url | https://www.mdpi.com/2072-4292/13/3/433 |
| work_keys_str_mv | AT jungeshen adualmodelarchitecturewithgroupingattentionfusionforremotesensingsceneclassification AT tongzhang adualmodelarchitecturewithgroupingattentionfusionforremotesensingsceneclassification AT yichenwang adualmodelarchitecturewithgroupingattentionfusionforremotesensingsceneclassification AT ruxinwang adualmodelarchitecturewithgroupingattentionfusionforremotesensingsceneclassification AT qiwang adualmodelarchitecturewithgroupingattentionfusionforremotesensingsceneclassification AT minqi adualmodelarchitecturewithgroupingattentionfusionforremotesensingsceneclassification AT jungeshen dualmodelarchitecturewithgroupingattentionfusionforremotesensingsceneclassification AT tongzhang dualmodelarchitecturewithgroupingattentionfusionforremotesensingsceneclassification AT yichenwang dualmodelarchitecturewithgroupingattentionfusionforremotesensingsceneclassification AT ruxinwang dualmodelarchitecturewithgroupingattentionfusionforremotesensingsceneclassification AT qiwang dualmodelarchitecturewithgroupingattentionfusionforremotesensingsceneclassification AT minqi dualmodelarchitecturewithgroupingattentionfusionforremotesensingsceneclassification |