Target And Background Separation in Hyperspectral Imagery for Automatic Target Detection

TL;DR

This paper introduces a novel method for separating known targets from backgrounds in hyperspectral images by modeling the image as a sum of low-rank background and sparse target components, enabling effective target detection.

Contribution

It presents a new approach that uses low-rank and sparse decomposition for target-background separation in hyperspectral imagery, with two strategies evaluated for detection.

Findings

Effective separation of targets from background demonstrated on synthetic data.

Method successfully applied to real hyperspectral images for target detection.

Two strategies provide flexible options for different scenarios.

Abstract

In this paper, we propose a method for separating known targets of interests from the background in hyperspectral imagery. More precisely, we regard the given hyperspectral image (HSI) as being made up of the sum of low-rank background HSI and a sparse target HSI that contains the known targets based on a pre-learned target dictionary specified by the user. Based on the proposed method, two strategies are outlined and evaluated independently to realize the target detection on both synthetic and real experiments.