Unsupervised Unmixing of Hyperspectral Imagery

This paper presents an approach for simultaneous determination of end members and their abundances in hyperspectral imagery using a constrained positive matrix factorization. The algorithm presented here solves the constrained PMF using Gauss-Seidel method. This algorithm alternates between the end members matrix updating step and the abundance estimation step until convergence is achieved. Preliminary results using a subset of the Enrique Reef image data are presented. These results show the potential of the method to solve the unsupervised unmixing problem.