Defocusing blur restoration in natural scene images for fractal analysis

Natural scene images contain objects at many different distances not all of which are known. In this situation, some of the objects will be out of focus and their images will be blurred. In this paper the authors present a model for defocusing blur. The authors show that the effect of the defocusing blur is to attenuate the high frequency components in the image leading to loss of edge information. The authors present a gradient descent approach to restoration using the observed image as the initial estimate. The authors present the restoration results for synthetic test images as well as natural scene images and demonstrate that the gradient descent is able to deblur the images