Restoration of degraded image with partial blurred regions based on blur detection and classification

In this paper, a new restoration algorithm for degraded image is proposed based on partial blurred regions detection and classification. At first, a segmentation based blurred regions detection algorithm is designed to detect the blurred regions from the degraded image, secondly an efficient classification algorithm of blurred regions is given to carry out the classification of blurred regions, thirdly the blur kernel of various class of blurred regions is estimated respectively so as to effectively implement the restoration of the detected various blurred regions with total variation method, finally all restored regions and the original non-blurred regions are synthesized to complete the restoration of the whole degraded image. The simulated experiment results verify the effectiveness of proposed algorithm.