Spatial resolution enhancement of hyperspectral image by negative abundance oriented spectral unmixing

Abstract

Hyperspectral images possess high spectral resolution, and are capable to give spectral information of different elements in the scene. The spatial resolution of hyperspectral image is very low. This leads to mixing of pixels. Spectral Unmixing algorithms are the most extensively used hyperspectral image analysis techniques for decomposing the mixed pixels. However spectral unmixing does not enhance the spatial resolution. In order to enhance the spatial resolution of hyperspectral image we propose a Hyperspectral-Multispectral image fusion method. The image fusion is based on a negative abundance oriented hyperspectral unmixing algorithm. The algorithm is tested with images from Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data.