Dictionary learning for clustering on hyperspectral images

TL;DR

This paper introduces a novel clustering method for hyperspectral images using dictionary learning and sparse coding, which outperforms traditional pixel clustering and other feature extraction techniques in certain scenarios.

Contribution

The paper proposes a new unsupervised clustering approach based on sparse coefficients from learned dictionaries, tailored for high-dimensional hyperspectral data.

Findings

Outperforms clustering on original pixels

Surpasses PCA and NMF feature-based clustering in certain cases

Suitable for large-scale high-dimensional data

Abstract

Dictionary learning and sparse coding have been widely studied as mechanisms for unsupervised feature learning. Unsupervised learning could bring enormous benefit to the processing of hyperspectral images and to other remote sensing data analysis because labelled data are often scarce in this field. We propose a method for clustering the pixels of hyperspectral images using sparse coefficients computed from a representative dictionary as features. We show empirically that the proposed method works more effectively than clustering on the original pixels. We also demonstrate that our approach, in certain circumstances, outperforms the clustering results of features extracted using principal component analysis and non-negative matrix factorisation. Furthermore, our method is suitable for applications in repetitively clustering an ever-growing amount of high-dimensional data, which is the case when working with hyperspectral satellite imagery.