Self-supervised spectral matching network for hyperspectral target detection

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.

Self-Supervised Learning in Remote Sensing: A review

“…Change detection [93,120,151,98,230,199,215,186,186] is usually also a pixellevel task which utilizes multitemporal information to detect changing pixels. In hyperspectral image analysis, most of the tasks are based on pixel level, including hyperspectral image classification 4 [106,100,101], image denoising [233], spectral unmixing [99], target detection [232], image restoration [102] and super-resolution [105,104]. Other pixel-level tasks include depth estimation [158,95] and SAR despeckling [110,109].…”