Spectral unmixing through Gaussian synapse ANNs in hyperspectral images

The work presented here is concerned with the application of Gaussian Synapse based Artificial Neural Networks to the spectral unmixing process when analyzing hyperspectral images. This type of networks and their training algorithm will be shown to be very efficient in the determination of the abundances of the different endmembers present in the image using a very small training set that can be obtained without any knowledge on the proportions of endmembers present. The Networks are tested using a benchmark set of artificially generated hyperspectral images containing five endmembers with spatially diverse abundances and finally verified on a real image.