SAR ATR Based on FCNN and ICAE
In recent years, Synthetic Aperture Radar (SAR) image target recognition based on the Convolutional Neural Network (CNN) has attracted a significant amount of attention. Fully CNN (FCNN) is a structural improvement of the CNN, which features a higher recognition rate than CNN, but it still requires...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
China Science Publishing & Media Ltd. (CSPM)
2018-10-01
|
| Series: | Leida xuebao |
| Subjects: | |
| Online Access: | http://radars.ie.ac.cn/fileup/HTML/R18066.htm |
| _version_ | 1797428687632596992 |
|---|---|
| author | Yu Lingjuan Wang Yadong Xie Xiaochun Lin Yun Hong Wen |
| author_facet | Yu Lingjuan Wang Yadong Xie Xiaochun Lin Yun Hong Wen |
| author_sort | Yu Lingjuan |
| collection | DOAJ |
| description | In recent years, Synthetic Aperture Radar (SAR) image target recognition based on the Convolutional Neural Network (CNN) has attracted a significant amount of attention. Fully CNN (FCNN) is a structural improvement of the CNN, which features a higher recognition rate than CNN, but it still requires a large number of labeled data in the training process. This study aims to propose a method of SAR image target recognition based on FCNN and Improved Convolutional Auto-Encoder (ICAE), where several parameters of FCNN are initialized by the parameters of the ICAE encoder. These parameters are obtained in the unsupervised training mode. Then, the FCNN is trained by the labeled training samples. The experimental results on 10 kinds of target classification based on the MSTAR datasets show that this method cannot only achieve an average of 98.14% correct recognition rate but also feature a strong anti-noise capability when the labeled training samples are unexpanded. |
| first_indexed | 2024-03-09T09:02:29Z |
| format | Article |
| id | doaj.art-bb70af9c8098457c95e451869303cff8 |
| institution | Directory Open Access Journal |
| issn | 2095-283X 2095-283X |
| language | English |
| last_indexed | 2024-03-09T09:02:29Z |
| publishDate | 2018-10-01 |
| publisher | China Science Publishing & Media Ltd. (CSPM) |
| record_format | Article |
| series | Leida xuebao |
| spelling | doaj.art-bb70af9c8098457c95e451869303cff82023-12-02T11:26:09ZengChina Science Publishing & Media Ltd. (CSPM)Leida xuebao2095-283X2095-283X2018-10-017562263110.12000/JR18066SAR ATR Based on FCNN and ICAEYu Lingjuan0Wang Yadong1Xie Xiaochun2Lin Yun3Hong Wen4①(School of Information Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China) ②(Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China)①(School of Information Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China) ②(Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China)③(School of Physics and Electronic Information, Gannan Normal University, Ganzhou 341000, China)②(Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China)②(Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China)In recent years, Synthetic Aperture Radar (SAR) image target recognition based on the Convolutional Neural Network (CNN) has attracted a significant amount of attention. Fully CNN (FCNN) is a structural improvement of the CNN, which features a higher recognition rate than CNN, but it still requires a large number of labeled data in the training process. This study aims to propose a method of SAR image target recognition based on FCNN and Improved Convolutional Auto-Encoder (ICAE), where several parameters of FCNN are initialized by the parameters of the ICAE encoder. These parameters are obtained in the unsupervised training mode. Then, the FCNN is trained by the labeled training samples. The experimental results on 10 kinds of target classification based on the MSTAR datasets show that this method cannot only achieve an average of 98.14% correct recognition rate but also feature a strong anti-noise capability when the labeled training samples are unexpanded.http://radars.ie.ac.cn/fileup/HTML/R18066.htmSynthetic Aperture Radar (SAR)Automatic target recognitionFully Convolutional Neural Network (FCNN)Convolutional Auto-Encoder (CAE)Improved Convolutional Auto-Encoder (ICAE) |
| spellingShingle | Yu Lingjuan Wang Yadong Xie Xiaochun Lin Yun Hong Wen SAR ATR Based on FCNN and ICAE Leida xuebao Synthetic Aperture Radar (SAR) Automatic target recognition Fully Convolutional Neural Network (FCNN) Convolutional Auto-Encoder (CAE) Improved Convolutional Auto-Encoder (ICAE) |
| title | SAR ATR Based on FCNN and ICAE |
| title_full | SAR ATR Based on FCNN and ICAE |
| title_fullStr | SAR ATR Based on FCNN and ICAE |
| title_full_unstemmed | SAR ATR Based on FCNN and ICAE |
| title_short | SAR ATR Based on FCNN and ICAE |
| title_sort | sar atr based on fcnn and icae |
| topic | Synthetic Aperture Radar (SAR) Automatic target recognition Fully Convolutional Neural Network (FCNN) Convolutional Auto-Encoder (CAE) Improved Convolutional Auto-Encoder (ICAE) |
| url | http://radars.ie.ac.cn/fileup/HTML/R18066.htm |
| work_keys_str_mv | AT yulingjuan saratrbasedonfcnnandicae AT wangyadong saratrbasedonfcnnandicae AT xiexiaochun saratrbasedonfcnnandicae AT linyun saratrbasedonfcnnandicae AT hongwen saratrbasedonfcnnandicae |