Physically Explainable CNN for SAR Image Classification

TL;DR

This paper introduces a novel physically explainable CNN for SAR image classification that incorporates electromagnetic physics knowledge to improve interpretability and performance, especially with limited labeled data.

Contribution

The paper proposes a new physics-guided and injected learning framework (PGIL) that integrates physics knowledge into CNNs for SAR image classification, enhancing explainability and accuracy.

Findings

PGIL outperforms traditional CNNs with limited labeled data.

Physics explanations improve interpretability and physical consistency.

Hybrid dataset evaluation confirms effectiveness across SAR data types.

Abstract

Integrating the special electromagnetic characteristics of Synthetic Aperture Radar (SAR) in deep neural networks is essential in order to enhance the explainability and physics awareness of deep learning. In this paper, we first propose a novel physically explainable convolutional neural network for SAR image classification, namely physics guided and injected learning (PGIL). It comprises three parts: (1) explainable models (XM) to provide prior physics knowledge, (2) physics guided network (PGN) to encode the knowledge into physics-aware features, and (3) physics injected network (PIN) to adaptively introduce the physics-aware features into classification pipeline for label prediction. A hybrid Image-Physics SAR dataset format is proposed for evaluation, with both Sentinel-1 and Gaofen-3 SAR data being experimented. The results show that the proposed PGIL substantially improve the classification performance in case of limited labeled data compared with the counterpart data-driven CNN and other pre-training methods. Additionally, the physics explanations are discussed to indicate the interpretability and the physical consistency preserved in the predictions. We deem the proposed method would promote the development of physically explainable deep learning in SAR image interpretation field.