Iterative composite filtering for image restoration

An algorithmic solution to the image restoration problem is proposed, under the assumptions that the image is nonstationary and the noise process is a superposition of white and impulsive noises. Separating the gross information of the image from its textural information the authors exploit the advantages of median, range, and Levinson filters. Median statistics are to estimate the image´s gross information and to filter the impulsive noise. Range statistics are used to segment the textural image into locally stationary images to be filtered by Levinson filters. The efficiency of the algorithm is illustrated by examples