Bottom-Up Scattering Information Perception Network for SAR target recognition

TL;DR

The paper targets robust SAR target recognition by better capturing scattering information in SAR imagery. It introduces a bottom-up scattering information perception network that uses a localized scattering perceptron as the backbone and an unsupervised scattering part feature extractor to model target scattering components. By aggregating information from target parts, the method yields a complete, interpretable target description with improved discriminative power. Experiments on FAST-Vehicle and SAR-ACD datasets demonstrate enhanced performance and robustness over baseline approaches, highlighting practical benefits for SAR interpretation and recognition.

Abstract

Deep learning methods based synthetic aperture radar (SAR) image target recognition tasks have been widely studied currently. The existing deep methods are insufficient to perceive and mine the scattering information of SAR images, resulting in performance bottlenecks and poor robustness of the algorithms. To this end, this paper proposes a novel bottom-up scattering information perception network for more interpretable target recognition by constructing the proprietary interpretation network for SAR images. Firstly, the localized scattering perceptron is proposed to replace the backbone feature extractor based on CNN networks to deeply mine the underlying scattering information of the target. Then, an unsupervised scattering part feature extraction model is proposed to robustly characterize the target scattering part information and provide fine-grained target representation. Finally, by aggregating the knowledge of target parts to form the complete target description, the interpretability and discriminative ability of the model is improved. We perform experiments on the FAST-Vehicle dataset and the SAR-ACD dataset to validate the performance of the proposed method.