|
|
|
| Occluded SAR Target Recognition Based on Multi-View Mutual Learning Network |
| REN Haohao1， CUI Shan2， JIANG Xinyu1， LIANG Shuyi1， ZHOU Yun1 |
| 1. Information and Communication Engineering， University of Electronic Science and Technology of China，
Chengdu 611731， Sichuan， China； 2. Shanghai Electro-Mechanical Engineering Institute， Shanghai 201109， China |
|
|
|
|
Abstract For the problem of SAR target recognition in occluded scenarios, this paper proposes a novel method called the multi-view mutual learning network. The proposed method is a mutual learning framework consisting of a multi-view complementary feature learning network and a recognition network, which aims to improve the recognition model's feature extraction ability through knowledge interaction at the feature level. Specifically, to enhance the recognition network's feature extraction ability, this study developed an attribute-scattering-center-guided hierarchical feature extraction method. In view of the implementation challenges with target features in occluded scenarios, a multi-view complementary learning method was employed to comprehensively characterize target features by leveraging complementary features across different SAR images at adjacent azimuth angles. Contrastive experimental results on the MSTAR dataset show that the proposed method performs well across varying levels of occlusion.
|
|
Received: 31 July 2025
Published: 13 January 2026
|
|
|
|
|
|
|
|
|
2025, Vol. 8 
