Date of Publication :27th December 2017

Abstract: For the classification of hyperspectral images (HSIs), this work presents a novel framework for an automatic system to segment the hyperspectral images. In the HSI, each pixel can be regarded as a shape-adaptive region, which consists of a number of spatial neighbouring pixels with very similar spectral characteristics. First, the proposed methodology adopts an over segmentation algorithm to cluster the HSI into many superpixels. Then, feature extraction is employed for the utilization of the spectral information, as well as spatial information, within and among superpixels. Finally, the hybridized machine learning algorithm is incorporated for the hyperspectral classification. This work introduces particle swarm optimization based support vector machine classifier for the classification. The Pavia database images are collected and simulated on MATLAB R2014a and the exposed results are showing the effectiveness of the proposed methodology

Reference :