Comparing Quadtree Region Partitioning Metrics for Hyperspectral Unmixing
This paper presents an approach for unmixing using quadtree region partitioning. Images are partitioned in low spectral variability regions using quadtree region partitioning. Unmixing is performed in each individual region using the positive matrix factorization. The extracted endmembers are the clustered in endmembers classes which account for the variability of spectral endmembers across the scene. The proposed method lends itself to an unsupervised approach. In the paper, we will examine the effect of different metrics for spectral variability and the type of scene being analyzed. Hyperspectral imagery from different sources is used in the experiments.