A Dual-Polarization Feature Fusion Network for Radar Automatic Target Recognition Based On HRRP Sequence

TL;DR

This paper introduces a dual-polarization feature fusion network with a novel two-stage fusion strategy and specialized loss function, significantly enhancing radar target recognition accuracy using polarimetric HRRP sequences.

Contribution

It presents a new dual-polarization feature fusion network with a two-stage fusion strategy and adaptive loss function for improved radar target recognition.

Findings

Significant performance improvement in radar target recognition tasks.

Effective multi-modal representation generation from polarimetric HRRP sequences.

Validated robustness of the proposed network through experiments.

Abstract

Recent advances in radar automatic target recognition (RATR) techniques utilizing deep neural networks have demonstrated remarkable performance, largely due to their robust generalization capabilities. To address the challenge for applications with polarimetric HRRP sequences, a dual-polarization feature fusion network (DPFFN) is proposed along with a novel two-stage feature fusion strategy. Moreover, a specific fusion loss function is developed, which enables the adaptive generation of comprehensive multi-modal representations from polarimetric HRRP sequences. Experimental results demonstrate that the proposed network significantly improves performance in radar target recognition tasks, thus validating its effectiveness. The PyTorch implementation of our proposed DPFFN is available at https://github.com/xmpan/DPFFN.