Research on the blind restoration algorithm of motion-blurred image

The restoration of motion-blurred image is one of important subjects of image restoration. This paper mainly researched on the restoration methods for motion-blurred image based on estimated motion blurring angle and motion blurring length. First, the motion blurring angle is estimated by Hough transform algorithm. In order to reduce the impact of concrete structure on the detection, the edge detection of the blurred image was done by Canny before the detection of the movement direction. Next, the motion blurring length is estimated by differential-autocorrelation. Finally, motion-blurred license plate image is restored by Lucy-Richardson based on motion blurring angle and length. Experiment results show the restoration image of the motion-blurred license plate is very clear.