自监督解耦动态分类器的小样本类增量SAR图像目标识别  

Few-Shot Class-Incremental SAR Image Target Recognition using Self-supervised Decoupled Dynamic Classifier

作　　者：赵琰 赵凌君 张思乾 计科峰[1] 匡纲要[1] ZHAO Yan;ZHAO Lingjun;ZHANG Siqian;JI Kefeng;KUANG Gangyao(College of Electronic Science and Technology,National University of Defense Technology,Changsha 410073,China)

机构地区：[1]国防科技大学电子科学学院,长沙410073

出　　处：《电子与信息学报》2024年第10期3936-3948,共13页Journal of Electronics & Information Technology

摘　　要：为提升基于深度学习(DL)的合成孔径雷达自动目标识别(SAR ATR)系统在开放动态的非合作场景中对新类别目标的持续敏捷识别能力,该文研究了SAR ATR的小样本类增量学习(FSCIL)问题,并提出了自监督解耦动态分类器(SDDC)。针对FSCIL中“灾难性遗忘”和“过拟合”本质难点和SAR ATR领域挑战,根据SAR图像目标信息的部件化与方位角敏感性特点,于图像域构建了基于散射部件混淆与旋转模块(SCMR)的自监督学习任务,以提升目标表征的泛化性与稳健性。同时,设计了类印记交叉熵(CI-CE)损失并以参数解耦学习(PDL)策略对模型动态微调,以对新旧知识平衡判别。实验在由MSTAR和SAR-AIRcraft-1.0数据集分别构建的覆盖多种目标类别、观测条件和成像平台的FSCIL场景上验证了该算法开放动态环境的适应能力。To power Deep-Learning(DL)based Synthetic Aperture Radar Automatic Target Recognition(SAR ATR)systems with the capability of learning new-class targets incrementally and rapidly in openly dynamic non-cooperative situations,the problem of Few-Shot Class-Incremental Learning(FSCIL)of SAR ATR is researched and a Self-supervised Decoupled Dynamic Classifier(SDDC)is proposed.Considering solving both the intrinsic Catastrophic forgetting and Overfitting dilemma of the FSCIL and domain challenges of SAR ATR,a self-supervised learning task powered by Scattering Component Mixup and Rotation(SCMR)is designed to improve the model’s generalizability and stability for target representation,leveraged by the partiality and azimuth dependence of target information in SAR imagery.Meanwhile,a Class-Imprinting Cross-Entropy(CI-CE)and a Parameter Decoupled Learning(PDL)strategy are designed to fine-tune networks dynamically to identify old and new targets evenly.Experiments on various FSCIL scenarios constructed by the MSTAR and the SAR-AIRcraft-1.0 datasets covering diverse target categories,observing environments,and imaging payloads,verify the method’s adaptability to openly dynamic world.

关 键 词：SAR目标识别 小样本类增量学习 自监督学习 深度学习 

分 类 号：TN957[电子电信—信号与信息处理] TP751.2[电子电信—信息与通信工程]