Dictionary learning for clustering on hyperspectral images
This incremental method addresses clustering challenges for remote sensing applications where labeled data are scarce.
The authors tackled the problem of clustering hyperspectral image pixels by using sparse coefficients from a dictionary as features, showing it works more effectively than clustering on original pixels and sometimes outperforms PCA and NMF-based features.
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.