Restoration And Quality Evaluation For Motion-blurred Image

Motion-blurring is caused by the relative motion between the camera and the object while the camera is recording. The purpose of restoration is to compensate for image degradation, so that the restored image is as close as possible the original image. So the quality of restored image should be evaluated. Motion blur parameters are priori knowledge necessary for building two-dimensional point spread function. The direction of motion blur is estimated by using the method combined Gauss - Laplace edge detection and Radon transform based on frequency-domain features. The measurement method of blur extent identification based on wave equations and autocorrelation function is used. On this basis, the Richardson-Lucy iterative algorithm and Wiener filtering method are utilized for motion-blurred image restoration under different parameters. For the restored images, Objective assessment to restoration of motion-blurred images based on traveling wave equation is adopted to evaluate the images quality. The criterion's validity, reliability and consistency with subjective evaluation are proven, and the effect of parameters selection of the Wiener filter method to restored image quality is determined through the comparison with the evaluation index of mean square error.