IncSAR: A Dual Fusion Incremental Learning Framework for SAR Target Recognition

TL;DR

IncSAR is a novel incremental learning framework for SAR target recognition that combines dual-branch architecture, noise reduction, and feature enhancement techniques to effectively learn new targets without forgetting previous ones.

Contribution

The paper introduces IncSAR, a dual fusion incremental learning framework integrating ViT and CNN, with noise reduction and feature decorrelation, to improve SAR target recognition in dynamic scenarios.

Findings

Achieves 99.63% accuracy on benchmark datasets.

Reduces performance drop to 0.33%, an 89% improvement in retention.

Outperforms state-of-the-art methods significantly.

Abstract

Deep learning techniques have achieved significant success in Synthetic Aperture Radar (SAR) target recognition using predefined datasets in static scenarios. However, real-world applications demand that models incrementally learn new information without forgetting previously acquired knowledge. The challenge of catastrophic forgetting, where models lose past knowledge when adapting to new tasks, remains a critical issue. In this paper, we introduce IncSAR, an incremental learning framework designed to tackle catastrophic forgetting in SAR target recognition. IncSAR combines the power of a Vision Transformer (ViT) and a custom-designed Convolutional Neural Network (CNN) in a dual-branch architecture, integrated via a late-fusion strategy. Additionally, we explore the use of TinyViT to reduce computational complexity and propose an attention mechanism to dynamically enhance feature representation. To mitigate the speckle noise inherent in SAR images, we employ a denoising module based on a neural network approximation of Robust Principal Component Analysis (RPCA), leveraging a simple neural network for efficient noise reduction in SAR imagery. Moreover, a random projection layer improves the linear separability of features, and a variant of Linear Discriminant Analysis (LDA) decorrelates extracted class prototypes for better generalization. Extensive experiments on the MSTAR, SAR-AIRcraft-1.0, and OpenSARShip benchmark datasets demonstrate that IncSAR significantly outperforms state-of-the-art approaches, achieving a 99.63\% average accuracy and a 0.33\% performance drop, representing an 89\% improvement in retention compared to existing techniques. The source code is available at https://github.com/geokarant/IncSAR.