A combined approach to enhancement of unknown blurred and noisy images
Hong Tang · 2002
Estimating the point-spread function of a real-world blurred image is an essential step in the restoration process. When blur is unknown and the noise varies over the neighborhood, image restoration is more difficult. Unlike the deterministic blur identification techniques that attempt to identify the PSF function, this paper introduces the AADIF, statistical, multidirectional and comparative (ASMC) sharpening approaches for dealing with unknown blurred and noisy images. The results are given to show the effectiveness and the practicality of the ASMC approach.>