A Quality of Images Fusion for Remote Sensing Applications

Abstract

Image fusion is a useful tool for integrating low spatial resolution multispectral (MS) images with a high spatial resolution panchromatic (PAN) image, thus producing a high resolution multispectral image for better understanding of the observed earth surface. It has has become an important issue for various remotes sensing (RS) problems such as land classification, change detection, object identification, image segmentation, map updating, hazard monitoring, and visualization purposes. When applied to remote sensing images, a common problem associated with existing fusion methods has been the color distortion, or degradation in the spectral quality. The main proposed of this research is to evaluate the quality of fused images for object identification. We examine the effectiveness of the following techniques multi-resolution analysis (MRA) and component substitution (CS) fusion. In order to improve this situation, the second purpose of this work is to establish an automatic and reliable way for the evaluation of the fused images, on the basis of both qualitative and quantitative metrics. In this result, It is found that the CS fusion method provides better performance than the MRA scheme. Quantitative analysis shows that the CS-based method gives a better result in terms of spatial quality (sharpness), whereas the MRA-based method yields better spectral quality, i.e., better color fidelity to the original MS images.