BLIND RESTORATION OF IMAGES DEGRADED BY SPACE-VARIANT BLURS USING ITERATIVE ALGORITHMS FOR BOTH BLUR IDENTIFICATION AND IMAGE-RESTORATION

This paper addresses the problem of digital blind restoration of image s degraded by space-variant blurs and noise. The existing Expectation- Maximization (EM) algorithm reported in the literature is extended in this paper and combined with the region adaptive technique to handle t he problem of identifying spatially variant blurs. The proposed algori thm is a two-step interative process. The expectation step of the EM a lgorithm is modified by the use of iterative image restoration. The en tire image is divided into disjointed regions and the blur is identifi ed in these regions using the proposed modified form of the EM algorit hm. The iterative Constrained Least Squares (CLS) algorithm used in sp ace-invariant image restoration is extended to restore the space varia nt blur images. Spatially adaptive algorithms for restoration are also applied. Experiments have been carried out to evaluate the performanc es of the proposed algorithms.