Pattern recognition in hyperspectral imagery using stable distribution analysis and one dimensional fringe-adjusted joint transform correlation

In hyperspectral imaging applications, the background generally exhibits a clearly non-Gaussian impulsive behavior, where valuable information stays in the tail. In this paper, we proposed a new technique where the background is modeled using stable distributions for robust detection of outliers. The outliers of the distribution can be considered as potential anomalies or regions of interest (ROI). To further decrease the false alarm rate, it may be necessary to compare the ROI with the given reference using a simple method. In this paper, we applied one dimensional fringe-adjusted joint transform correlation technique, which can detect both single and multiple objects in constant time while accommodating the in-plane and out-of-plane distortions. Simulation results using real life hyperspectral image data are presented to verify the effectiveness of the proposed technique.