Self-supervised spectral matching network for hyperspectral target detection

TL;DR

This paper introduces a self-supervised spectral matching network for hyperspectral target detection, effectively handling data imbalance and complex backgrounds to improve pixel-level recognition accuracy.

Contribution

It proposes a novel spectral similarity based matching network with a pair-based loss and background separation, enhancing feature discrimination in hyperspectral data.

Findings

Outperforms existing detectors on three real datasets

Effectively handles imbalanced and complex background data

Improves discriminative feature learning

Abstract

Hyperspectral target detection is a pixel-level recognition problem. Given a few target samples, it aims to identify the specific target pixels such as airplane, vehicle, ship, from the entire hyperspectral image. In general, the background pixels take the majority of the image and complexly distributed. As a result, the datasets are weak annotated and extremely imbalanced. To address these problems, a spectral mixing based self-supervised paradigm is designed for hyperspectral data to obtain an effective feature representation. The model adopts a spectral similarity based matching network framework. In order to learn more discriminative features, a pair-based loss is adopted to minimize the distance between target pixels while maximizing the distances between target and background. Furthermore, through a background separated step, the complex unlabeled spectra are downsampled into different sub-categories. The experimental results on three real hyperspectral datasets demonstrate that the proposed framework achieves better results compared with the existing detectors.