Investigation of unsupervised and supervised hyperspectral anomaly detection

TL;DR

This paper evaluates and compares supervised and unsupervised hyperspectral anomaly detection methods, including a hybrid ensemble approach, to improve detection accuracy and address the challenge of detecting unknown patterns.

Contribution

It introduces a hybrid ensemble method combining unsupervised algorithms with supervised classification and provides a comprehensive evaluation of these techniques on hyperspectral data.

Findings

Hybrid ensemble improves detection accuracy.

Supervised methods struggle with unknown patterns.

Evaluation offers new insights into method effectiveness.

Abstract

Hyperspectral sensing is a valuable tool for detecting anomalies and distinguishing between materials in a scene. Hyperspectral anomaly detection (HS-AD) helps characterize the captured scenes and separates them into anomaly and background classes. It is vital in agriculture, environment, and military applications such as RSTA (reconnaissance, surveillance, and target acquisition) missions. We previously designed an equal voting ensemble of hyperspectral unmixing and three unsupervised HS-AD algorithms. We later utilized a supervised classifier to determine the weights of a voting ensemble, creating a hybrid of heterogeneous unsupervised HS-AD algorithms with a supervised classifier in a model stacking, which improved detection accuracy. However, supervised classification methods usually fail to detect novel or unknown patterns that substantially deviate from those seen previously. In this work, we evaluate our technique and other supervised and unsupervised methods using general hyperspectral data to provide new insights.