A New Image Fusion Technique to Improve the Quality of Remote Sensing images

Image fusion is a process of producing a single fused image from a set of input images. In this paper a new fusion technique based on the use of independent component analysis (ICA) and IHS transformation is proposed. A comparison of this new technique with PCA, IHS, and ICA-based fusion techniques is given. Quick Bird data are used to test these techniques, the output was evaluated using subjective comparison, statistical correlation, information entropy, mean square error, and standard deviation. The results of the proposed technique show higher performance compared to the other techniques.