An Attention Cascade Global–Local Network for Remote Sensing Scene Classification
Remote sensing image scene classification is an important task of remote sensing image interpretation, which has recently been well addressed by the convolutional neural network owing to its powerful learning ability. However, due to the multiple types of geographical information and redundant backg...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-04-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/9/2042 |
| _version_ | 1797503025020928000 |
|---|---|
| author | Junge Shen Tianwei Yu Haopeng Yang Ruxin Wang Qi Wang |
| author_facet | Junge Shen Tianwei Yu Haopeng Yang Ruxin Wang Qi Wang |
| author_sort | Junge Shen |
| collection | DOAJ |
| description | Remote sensing image scene classification is an important task of remote sensing image interpretation, which has recently been well addressed by the convolutional neural network owing to its powerful learning ability. However, due to the multiple types of geographical information and redundant background information of the remote sensing images, most of the CNN-based methods, especially those based on a single CNN model and those ignoring the combination of global and local features, exhibit limited performance on accurate classification. To compensate for such insufficiency, we propose a new dual-model deep feature fusion method based on an attention cascade global–local network (ACGLNet). Specifically, we use two popular CNNs as the feature extractors to extract complementary multiscale features from the input image. Considering the characteristics of the global and local features, the proposed ACGLNet filters the redundant background information from the low-level features through the spatial attention mechanism, followed by which the locally attended features are fused with the high-level features. Then, bilinear fusion is employed to produce the fused representation of the dual model, which is finally fed to the classifier. Through extensive experiments on four public remote sensing scene datasets, including UCM, AID, PatternNet, and OPTIMAL-31, we demonstrate the feasibility of the proposed method and its superiority over the state-of-the-art scene classification methods. |
| first_indexed | 2024-03-10T03:44:35Z |
| format | Article |
| id | doaj.art-12876796abac4580abaa8947f2db0a3c |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T03:44:35Z |
| publishDate | 2022-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-12876796abac4580abaa8947f2db0a3c2023-11-23T09:09:41ZengMDPI AGRemote Sensing2072-42922022-04-01149204210.3390/rs14092042An Attention Cascade Global–Local Network for Remote Sensing Scene ClassificationJunge Shen0Tianwei Yu1Haopeng Yang2Ruxin Wang3Qi Wang4Unmanned 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, ChinaEngineering Research Center of Cyberspace, School of Software, Yunnan University, Kunming 650106, ChinaUnmanned System Research Institute, Northwestern Polytechnical University, Xi’an 710072, ChinaRemote sensing image scene classification is an important task of remote sensing image interpretation, which has recently been well addressed by the convolutional neural network owing to its powerful learning ability. However, due to the multiple types of geographical information and redundant background information of the remote sensing images, most of the CNN-based methods, especially those based on a single CNN model and those ignoring the combination of global and local features, exhibit limited performance on accurate classification. To compensate for such insufficiency, we propose a new dual-model deep feature fusion method based on an attention cascade global–local network (ACGLNet). Specifically, we use two popular CNNs as the feature extractors to extract complementary multiscale features from the input image. Considering the characteristics of the global and local features, the proposed ACGLNet filters the redundant background information from the low-level features through the spatial attention mechanism, followed by which the locally attended features are fused with the high-level features. Then, bilinear fusion is employed to produce the fused representation of the dual model, which is finally fed to the classifier. Through extensive experiments on four public remote sensing scene datasets, including UCM, AID, PatternNet, and OPTIMAL-31, we demonstrate the feasibility of the proposed method and its superiority over the state-of-the-art scene classification methods.https://www.mdpi.com/2072-4292/14/9/2042remote sensing scene classificationconvolutional neural networkneural architecture searchfeature fusion |
| spellingShingle | Junge Shen Tianwei Yu Haopeng Yang Ruxin Wang Qi Wang An Attention Cascade Global–Local Network for Remote Sensing Scene Classification Remote Sensing remote sensing scene classification convolutional neural network neural architecture search feature fusion |
| title | An Attention Cascade Global–Local Network for Remote Sensing Scene Classification |
| title_full | An Attention Cascade Global–Local Network for Remote Sensing Scene Classification |
| title_fullStr | An Attention Cascade Global–Local Network for Remote Sensing Scene Classification |
| title_full_unstemmed | An Attention Cascade Global–Local Network for Remote Sensing Scene Classification |
| title_short | An Attention Cascade Global–Local Network for Remote Sensing Scene Classification |
| title_sort | attention cascade global local network for remote sensing scene classification |
| topic | remote sensing scene classification convolutional neural network neural architecture search feature fusion |
| url | https://www.mdpi.com/2072-4292/14/9/2042 |
| work_keys_str_mv | AT jungeshen anattentioncascadegloballocalnetworkforremotesensingsceneclassification AT tianweiyu anattentioncascadegloballocalnetworkforremotesensingsceneclassification AT haopengyang anattentioncascadegloballocalnetworkforremotesensingsceneclassification AT ruxinwang anattentioncascadegloballocalnetworkforremotesensingsceneclassification AT qiwang anattentioncascadegloballocalnetworkforremotesensingsceneclassification AT jungeshen attentioncascadegloballocalnetworkforremotesensingsceneclassification AT tianweiyu attentioncascadegloballocalnetworkforremotesensingsceneclassification AT haopengyang attentioncascadegloballocalnetworkforremotesensingsceneclassification AT ruxinwang attentioncascadegloballocalnetworkforremotesensingsceneclassification AT qiwang attentioncascadegloballocalnetworkforremotesensingsceneclassification |