Feature Augmentation Improves Anomalous Change Detection for Human Activity Identification in Synthetic Aperture Radar Imagery

TL;DR

This paper demonstrates that augmenting feature space with local spatial information significantly improves anomalous change detection in SAR imagery for human activity identification, especially in low-dimensional data scenarios.

Contribution

The study introduces a feature augmentation approach that enhances ACD performance in SAR imagery by incorporating local spatial information, addressing low-dimensional data challenges.

Findings

Feature augmentation improves ACD accuracy in SAR imagery.

Increased feature dimensionality outperforms simple differencing methods.

Local spatial information is crucial for detecting human activity in SAR data.

Abstract

Anomalous change detection (ACD) methods separate common, uninteresting changes from rare, significant changes in co-registered images collected at different points in time. In this paper we evaluate methods to improve the performance of ACD in detecting human activity in SAR imagery using outdoor music festivals as a target. Our results show that the low dimensionality of SAR data leads to poor performance of ACD when compared to simpler methods such as image differencing, but augmenting the dimensionality of our input feature space by incorporating local spatial information leads to enhanced performance.