Regularization-Based Region Learning for Radar HRRP Open Set Recognition

In real-world scenarios, radar systems need to recognize both known and unknown classes, referred to as open set recognition (OSR). Prototype-like methods have shown significant potential in addressing this challenging task. However, prototype learning overlooks the directional relationship between feature embeddings and prototypes, leading to open space risk. In addition, strong learning for known classes can weaken the sensitivity to sample variations, impacting the OSR performance of the model. To address these issues, we propose the regularization-based region learning (RBRL) method. First, from the perspective of gradient propagation, RBRL demonstrates that $l_{2}$ -regularization can enhance the sensitivity of logit vector directions to sample variations. Unlike in closed-set scenarios, $l_{2}$ -regularization can effectively improve the OSR performance even with sufficient training data. Second, RBRL minimizes the known space in both distance and angular domains by leveraging the spatial geometric relationship between prototypes and feature embeddings, mitigating open space risk. Furthermore, decision-level fusion is achieved by computing the area of the fan-shaped region in RBRL. The effectiveness of RBRL is validated on datasets comprising both measured and simulated high-resolution range profiles (HRRPs). RBRL achieves state-of-the-art performance across various OSR metrics.