ABSTRACT
This chapter presents an overview of image fusion methods for the case where the input channels are blurred, noisy, and geometrically different. Image fusion has been used in many application areas. In remote sensing and in astronomy, multisensor fusion is used to achieve high spatial and spectral resolutions by combining images from two sensors, one of which has high spatial resolution and the other, high spectral resolution. Analysis and interpretation of degraded images is the key problem in real applications, because the degradations are, in principle, inevitable. A very promising approach to image quality enhancement is to fuse several channels with different degradations together in order to extract as much useful information as possible. The chapter presents an alternating minimization algorithm for multichannel blind deconvolution and demonstrates that it is a powerful tool for image fusion in the case of uniformly blurred channels.
