Sparse Signal Models for Data Augmentation in Deep Learning ATR

TL;DR

This paper introduces a sparse modeling-based data augmentation technique for SAR ATR that enhances CNN training with limited data by synthesizing new images using domain-specific sparsity characteristics.

Contribution

It presents a novel sparse signal modeling approach for SAR data augmentation, improving ATR performance with scarce training data.

Findings

Significant accuracy gains in ATR with limited training data

Effective synthesis of new SAR images at unseen poses and translations

Enhanced CNN generalization performance through proposed augmentation

Abstract

Automatic Target Recognition (ATR) algorithms classify a given Synthetic Aperture Radar (SAR) image into one of the known target classes using a set of training images available for each class. Recently, learning methods have shown to achieve state-of-the-art classification accuracy if abundant training data is available, sampled uniformly over the classes, and their poses. In this paper, we consider the task of ATR with a limited set of training images. We propose a data augmentation approach to incorporate domain knowledge and improve the generalization power of a data-intensive learning algorithm, such as a Convolutional neural network (CNN). The proposed data augmentation method employs a limited persistence sparse modeling approach, capitalizing on commonly observed characteristics of wide-angle synthetic aperture radar (SAR) imagery. Specifically, we exploit the sparsity of the scattering centers in the spatial domain and the smoothly-varying structure of the scattering coefficients in the azimuthal domain to solve the ill-posed problem of over-parametrized model fitting. Using this estimated model, we synthesize new images at poses and sub-pixel translations not available in the given data to augment CNN's training data. The experimental results show that for the training data starved region, the proposed method provides a significant gain in the resulting ATR algorithm's generalization performance.