Segmentation as postprocessing for hyperspectral image classification

Hyperspectral imaging is a technique in remote sensing that collect hundreds of images at differents wavelength values in the same area of the Earth. For instance, the Airborne Visible Infra-Red Imaging Spectrometer (AVIRIS) sensor of NASA capable to obtain 224 spectral channels in a wavelength range between 400 and 2500 nanometers. As a result, each pixel of the image can be represented as a spectral signature. Image segmentation is the process of dividing a digital image into groups of pixels or objects. Hyperspectral image classification is an important and active area dedicated to identifying each pixel in the image with an exclusive material/object class. Several efforts had been done in this field using spectral and spatial information separately or simultaneously due to improve the performance of the classification techniques. In previous works have been used similar techniques using spectral and spatial information separately or simultaneously. In this work we have focused on a region-growing segmentation algorithm, applied as postprocessing on the standard classification, produces by a SVM classifier, in order to improve the performance of the classification technique. Experimental results with a real hyperspectral data set over the city of Pavia are included.