Enhanced hyperspectral image segmentation using wavelets transform

Remote sensing technologies are very useful due to their wide practical applications. The latest new technology, hyperspectral imaging, could be a valuable tool to study and evaluate many economic and environmental issues. It offers an advanced monitoring technique in more detailed spectral information than previous remote sensing systems. The present work is devoted to the study of segmentation of hyperspectral imaging. The proposed approach depends on the principal component analysis to select the best features of the HS image. Then applying the wavelet transform to get different scales of the scene. Finally using the mean shift clustering technique we get the segmented image. The developed method is evaluated and proved result improvement over related work; it proved also that the segmented regions of the proposed approach are very smooth.