Analysis and Comparison of Image Restoration Methods

The principles of inverse filtering, wiener filtering, and histogram equalization combined median filtering are introduced and studied. Three image restoration methods are compared in a variety of blur and noise conditions. Their own advantage and disadvantage are described. The simulation experiments of three methods for the motion blurred image with and without noise have been done by MATLAB. It is demonstrated that, in certain conditions, one restoration method is preferable to others.