Using of the Moore-Penrose inverse matrix in image restoration

Abstract

A method for digital image restoration, based on the Moore-Penrose inverse matrix, has many practical applications. We apply the method to remove blur in an image caused by uniform linear motion. This method assumes that linear motion corresponds to an integral number of pixels. Compared to other classical methods, this method attains higher values of the Improvement in Signal to Noise Ratio (ISNR) parameter and of the Peak Signal-to-Noise Ratio (PSNR), but a lower value of the Mean Square Error (MSE). We give an implementation in the MATLAB programming package.