Superpixel-guided Discriminative Low-rank Representation of Hyperspectral Images for Classification

TL;DR

This paper introduces SP-DLRR, a novel hyperspectral image classification method that combines superpixel segmentation with discriminative low-rank representation, effectively utilizing local spatial and spectral information for improved accuracy.

Contribution

The paper presents a new classification scheme integrating superpixel segmentation and low-rank representation, enhancing discriminability and robustness in hyperspectral image classification.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Effective with very limited training pixels.

Improves intra-class similarity and inter-class discriminability.

Abstract

In this paper, we propose a novel classification scheme for the remotely sensed hyperspectral image (HSI), namely SP-DLRR, by comprehensively exploring its unique characteristics, including the local spatial information and low-rankness. SP-DLRR is mainly composed of two modules, i.e., the classification-guided superpixel segmentation and the discriminative low-rank representation, which are iteratively conducted. Specifically, by utilizing the local spatial information and incorporating the predictions from a typical classifier, the first module segments pixels of an input HSI (or its restoration generated by the second module) into superpixels. According to the resulting superpixels, the pixels of the input HSI are then grouped into clusters and fed into our novel discriminative low-rank representation model with an effective numerical solution. Such a model is capable of increasing the intra-class similarity by suppressing the spectral variations locally while promoting the inter-class discriminability globally, leading to a restored HSI with more discriminative pixels. Experimental results on three benchmark datasets demonstrate the significant superiority of SP-DLRR over state-of-the-art methods, especially for the case with an extremely limited number of training pixels.