Updated Homogeneity Criteria based Low-dimensional Representation for Hyperspectral Unmixing

Superpixel-based approaches have been proposed for hyperspectral unmixing. The basic assumption of this approach is that the superpixel over-segmentation segments the image into small homogeneous areas. Here we present an improved superpixel-based dimensionality reduction approach that accounts for superpixel homogeneity. Homogeneous superpixels are represented by their mean but heterogeneous superpixels are represented by multiple representative signatures selected using the SVDSS column subset selection algorithm. The representative signatures for the homogeneous and heterogeneous superpixels provide an improved low-dimensional representation for the hyperspectral image that better captures the image structure. We present experiments applying the proposed and the conventional superpixel dimensionality reduction approaches to unmixing using the constrained non-negative matrix factorization (cNMF). Experimental results comparing the proposed approach with other superpixel-based unmixing approaches are presented. The proposed algorithm was performed on the hyperspectral image Fort A.P. Hill. Both the qualitative evaluation and quantitative assessment are implemented. Experimental results show that unmixing results are improved by the enhanced superpixel-based low dimensional representation.