A novel joint dictionary framework for sparse hyperspectral unmixing incorporating spectral library

Abstract

The dictionary-aided sparse representation has recently become a promising method in hyperspectral unmixing. Under the assumption of linear mixture model, sparse unmixing aims at selecting a small subset of spectral signals in the spectral library known in advance to represent the whole hyperspectral image. Unfortunately, the high mutual coherence of spectral libraries, along with their ever-growing dimensionality, strongly limits the performance of sparse unmixing. For this purpose, we propose a joint dictionary framework for sparse unmixing (JDSU) to tackle such a limitation in this paper. The proposed approach combines the learning capacity and priori information to improve the performance of sparse unmixing by incorporating the spectral library into the dictionary learning method. A multi-dictionary learning model has been developed based on cluster analysis, and it takes advantage of the collaborative effect of endmembers in the local hyperspectral image to learn several local dictionaries which comprise the joint dictionary. Moreover, JDSU can act as a dictionary pruning algorithm which provides a possibility that sparse unmixing algorithms could have higher accuracy and efficiency. Experimental results illustrate that, under certain conditions, JDSU can recover the optimal endmembers from the spectral library. The effectiveness of the proposed approach is extensively validated on both simulated and real hyperspectral data sets.