PSO-based estimation for Gaussian blur in blind image deconvolution problem

This study focuses on Gaussian blur estimation for blind image deconvolution (BID) problem. In BID problem, it only uses blurred image and less information of point spread function (PSF) to restore the received the blurred image. Due to restore the received image, the first step is to identify the proper PSF model. The received image does not uniquely define the PSF. Nevertheless these are many applications where the received image have been blurred either by an unknown or a partially know PSF. Therefore, this paper choose Gaussian blur image for further research, which utilized the particle swarm optimization algorithm to search for the unknown PSF. The objective function for searching the parameters of PSF is based on edge detection and image morphology. It can identify the parameters of PSF exactly. Finally, the feasibility and validity of proposed algorithm are demonstrated by several simulations.