Relation-Attention Networks for Remote Sensing Scene Classification
Remote sensing (RS) scene classification plays an important role in a wide range of RS applications. Recently, convolutional neural networks (CNNs) have been applied to the field of scene classification in RS images and achieved impressive performance. However, to classify RS scenes, most of the exi...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9652121/ |
| _version_ | 1798034798834352128 |
|---|---|
| author | Xin Wang Lin Duan Chen Ning Huiyu Zhou |
| author_facet | Xin Wang Lin Duan Chen Ning Huiyu Zhou |
| author_sort | Xin Wang |
| collection | DOAJ |
| description | Remote sensing (RS) scene classification plays an important role in a wide range of RS applications. Recently, convolutional neural networks (CNNs) have been applied to the field of scene classification in RS images and achieved impressive performance. However, to classify RS scenes, most of the existing CNN methods either utilize the high-level features from the last convolutional layer of CNNs, missing much important information existing at the other levels, or directly fuse the features at different levels, bringing redundant and/or mutually exclusive information. Inspired by the attention mechanism of the human visual system, in this article, we explore a novel relation-attention model and design an end-to-end relation-attention network (RANet) to learn powerful feature representations of multiple levels to further improve the classification performance. First, we propose to extract convolutional features at different levels by pretrained CNNs. Second, a multiscale feature computation module is constructed to connect features at different levels and generate multiscale semantic features. Third, a novel relation-attention model is designed to focus on the critical features whilst avoiding the use of redundant and even distractive ones by exploiting the scale contextual information. Finally, the resulting relation-attention features are concatenated and fed into a softmax layer for the final classification. Experiments on four well-known RS scene classification datasets (UC-Merced, WHU-RS19, AID, and OPTIMAL-31) show that our method outperforms some state-of-the-art algorithms. |
| first_indexed | 2024-04-11T20:49:23Z |
| format | Article |
| id | doaj.art-b90707a1c8b74e218849d679aa241636 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-04-11T20:49:23Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-b90707a1c8b74e218849d679aa2416362022-12-22T04:03:54ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352022-01-011542243910.1109/JSTARS.2021.31355669652121Relation-Attention Networks for Remote Sensing Scene ClassificationXin Wang0https://orcid.org/0000-0003-0203-9964Lin Duan1Chen Ning2https://orcid.org/0000-0002-3026-2496Huiyu Zhou3https://orcid.org/0000-0003-1634-9840College of Computer and Information, Hohai University, Nanjing, ChinaCollege of Computer and Information, Hohai University, Nanjing, ChinaSchool of Computer and Electronic Information, Nanjing Normal University, Nanjing, ChinaSchool of Informatics, University of Leicester, Leicester, U.K.Remote sensing (RS) scene classification plays an important role in a wide range of RS applications. Recently, convolutional neural networks (CNNs) have been applied to the field of scene classification in RS images and achieved impressive performance. However, to classify RS scenes, most of the existing CNN methods either utilize the high-level features from the last convolutional layer of CNNs, missing much important information existing at the other levels, or directly fuse the features at different levels, bringing redundant and/or mutually exclusive information. Inspired by the attention mechanism of the human visual system, in this article, we explore a novel relation-attention model and design an end-to-end relation-attention network (RANet) to learn powerful feature representations of multiple levels to further improve the classification performance. First, we propose to extract convolutional features at different levels by pretrained CNNs. Second, a multiscale feature computation module is constructed to connect features at different levels and generate multiscale semantic features. Third, a novel relation-attention model is designed to focus on the critical features whilst avoiding the use of redundant and even distractive ones by exploiting the scale contextual information. Finally, the resulting relation-attention features are concatenated and fed into a softmax layer for the final classification. Experiments on four well-known RS scene classification datasets (UC-Merced, WHU-RS19, AID, and OPTIMAL-31) show that our method outperforms some state-of-the-art algorithms.https://ieeexplore.ieee.org/document/9652121/Convolutional neural network (CNN)relation-attention network (RANet)remote sensing (RS)scene classification |
| spellingShingle | Xin Wang Lin Duan Chen Ning Huiyu Zhou Relation-Attention Networks for Remote Sensing Scene Classification IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Convolutional neural network (CNN) relation-attention network (RANet) remote sensing (RS) scene classification |
| title | Relation-Attention Networks for Remote Sensing Scene Classification |
| title_full | Relation-Attention Networks for Remote Sensing Scene Classification |
| title_fullStr | Relation-Attention Networks for Remote Sensing Scene Classification |
| title_full_unstemmed | Relation-Attention Networks for Remote Sensing Scene Classification |
| title_short | Relation-Attention Networks for Remote Sensing Scene Classification |
| title_sort | relation attention networks for remote sensing scene classification |
| topic | Convolutional neural network (CNN) relation-attention network (RANet) remote sensing (RS) scene classification |
| url | https://ieeexplore.ieee.org/document/9652121/ |
| work_keys_str_mv | AT xinwang relationattentionnetworksforremotesensingsceneclassification AT linduan relationattentionnetworksforremotesensingsceneclassification AT chenning relationattentionnetworksforremotesensingsceneclassification AT huiyuzhou relationattentionnetworksforremotesensingsceneclassification |