Motion Image Restoration

A fundamental issue in image restoration is blur removal in the presence of observation noise. Blur may be introduced by relative motion between a camera and the object, by a camera that is out of focus, and the other factors. Image restoration is an important task of the image processing, there have a broadly market and applying in practice.We proposed two methods which apply different command and different to restoration blur image which was introduced by relative motion between the camera and the object scene.First, It is an important question that how to restore the blurred image quickly in a large number of applications. The solution proposed here identifies important parameters with which to characterize the point spread function (PSF) of the blur, given only the blurred image itself. The method proposed here identifies the extent of the PSF of the blurred image and the movement direction is horizontal. When the direction of the line movement which we already know or we can get the direction by using other methods is not horizontal, we can use line transformation to transform the horizontal direction. Usually, the exposure time of camera is very short, so we can think the motion is constant line movement in the exposure time in a large number of applications. As we know, correct identification of the PSF parameters permits fast high resolution restoration of the blurred image. Using the method, we can quickly restore the blurred image on line, and when the noise of the image is bigger, it can perform very well to identifity the blurred distance.Second, we can not identifies the PSF of the blurred image, that is to say, we only have insufficient information of the blurred image. How to restore the blurred image in this case? This is blind image restoration. In this paper,we proposed an EMM (error minimum and maximum A posteriori estimation) method to restore the blurred image. The new method effectively utilizes the piecewise smoothness of the PSF. It attempts to minimize a cost function consisting of a restoration error and the one regularization terms subject to other hard constraints. A scale problem inherent to the cost function is identified, which, if not properly treated, may hinder the minimization blind restoration process. Alternating minimization cost function and maximization poster probability is proposed to restore the blurred image. The method is divided two step, one step is EM step, it is to estimate the PSF by using minimimum-error criteria, the other step is M step, we use MAP criteria to restore the blurred image. Good performance is observed with motion blurred image with using the EMM method.