Infrared and Laser Engineering, Volume. 50, Issue 6, 20200531(2021)
SAR target recognition using feature fusion by 2D compressive sensing with multiple random projection matrices
Figures & Tables(8)
Fig. 1. Recognition method of joint sparse representation based on multi-layer 2D compressive sensing features
Fig. 3. Recognition results of each class by the proposed method under SOC
Table 1. Describtion of SOC
View tableView in ArticleTable 1. Describtion of SOC
Target class Training set （17°） Test set （15°） Configuration Number of samples Configuration Number of samples BMP2 9563 233 9563 195 9566 196 c21 196 BTR70 c71 233 c71 196 T72 132 232 132 196 812 195 s7 191 T62 A51 299 A51 273 BRDM2 E-71 298 E-71 274 BTR60 7532 256 7532 195 ZSU23/4 d08 299 d08 274 D7 13015 299 13015 274 ZIL131 E12 299 E12 274 2S1 B01 299 B01 274
Table 2. Results under SOC
View tableView in ArticleTable 2. Results under SOC
Method Recognition accuracy Recognition efficiency/ms Proposed 99.04% 76.2 2DPCA 96.12% 96.4 Mono 97.83% 102.9 2DCS 95.84% 64.3 CNN 98.12% 136.5
Table 3. Describtion of the depression angle variance
View tableView in ArticleTable 3. Describtion of the depression angle variance
Target class Training set Test set Depression angle/(°) Number of samples Depression angle/(°) Number of samples 2S1 17 299 30 288 45 303 BDRM2 17 298 30 287 45 303 ZSU23/4 17 299 30 288 45 303
Tools
Get Citation
Copy Citation Text
Jianbo Wu, Zhengwu Lu, Yurong Guan, Qingdong Wang, Guosong Jiang. SAR target recognition using feature fusion by 2D compressive sensing with multiple random projection matrices[J]. Infrared and Laser Engineering, 2021, 50(6): 20200531
Paper Information
Category: Image processing
Received: Dec. 16, 2020
Accepted: --
Published Online: Aug. 19, 2021
The Author Email: Jiang Guosong (jixiang_sng@163.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20