Design and Analysis of Neuro Fuzzy Color Image Restoration

Color image is degraded by a blur function and corrupted by random noise. It is necessary to find the best approximation of original image so that the corrupted image may be used for future use .The problem is to accurately remove the noise degradation and blurring without damaging the intelligence. New method based on neuro fuzzy techniques has been proposed in this paper, to address the filtering of Gaussian and random valued impulse noise. Here color images are classified into RGB subspace. Then each subspace can be regarded as a gray image space and the fuzzy techniques can be applied to all the subspace. The system can be robust by introducing neuro fuzzy techniques. The emphasis of technique is on the preservation of total visual quality of the image and simultaneously taking into account on computational overhead, improved performance in terms of signal to noise ratio with less number of neurons.