Out-of-focus blur estimation using isotropic step responses and its application to image restoration

Abstract

In this paper, we propose an out-of-focus blur estimation using isotropic step responses and its application to image restoration. The proposed algorithm can make in-focused image by using only digital image processing techniques, and it required neither infrared light or ultrasound nor focusing lens assembly operated by electrically powered movement of focusing lens.To increase accuracy in estimating the Point-spread-function (PSF) of the defocus image, the proposed algorithm finds true and linear edges by using Canny edge detector, which is optimal edge detector and has good localization, estimates the step response across the edge for each pixel, computes the one-dimensional step response by averaging the step responses, estimates the two-dimensional PSF from the averaged step response, and finally provides in-focused image by image restoration filter based on the estimated PSF.There is a limit of the amount of out-of-focus, which can be recovered by the proposed algorithm. Moreover, the proposed algorithm is operating under assumption that an input image contains at least one piece-wise linear boundary between an object and background. In spite of above-mentioned limitations, the proposed algorithm can make acceptable quality of focused image by using only digital image processing.