Image Restoration using Higher Order Statistics

Most of the techniques for image restoration are based on some known degradation models. But in many situations it is difficult to accurately measure the degradation factors or noise type that is the real motivation behind the use of blind deconvolution technique for image restoration. Here the observed degraded image is restored without having any prior knowledge about the noise type. Most of the existing blind deconvolution methods for images apply for the restoration of grey scale images. In this paper a blind deconvolution technique using higher order statistics is applied for colour image restoration. Selective filtering is repeatedly applied for better results.