The Parameter Estimation And Image Restoration Of Uniform Linear Motion-blurring

Motion-blurring is caused by the relative motion between the camara and the object which is screened while the image is collected. The research on motion-blurring image restoration is quite necessary.The purpose of image restoration is to resume the distortion image to be similar with the original image. So the quality of restored image should be measured. As the images in real life are always blurred by noise, so we should study on the filtration of familiar noise, which is preparative for the restoration of motion-blurring image.The key of motion-blurring image restoration is to get the motion-direction and the motion-scales. The motion-direction is estimated by the method combined binarization, edge detection and hough transfom in frequency-domain. As the motion-direction is got, the original image could be translated to horizontal motion-blurring, and the motion-blurring could be estimated by counting the number of the dark stripes in spectrum image or using differential and autocorrelation.When the parameters of the motion-blurring are confirmed, blurred image could be resumed by Wiener filtering method and Lucy-Richardson iterative algorithm. The result should be measured and so we could research how to weaken ringing effect.