The Restoration of Blind Images Based on Neural Networks
基于神经网络的盲图像恢复

Based on the part prior knowledge of the original image, a blind restoration method for the degraded image is presented in this paper. The approach includes two main steps, namley, the estimation of The point spread function of the degraded image and the restoration by using Hopfield neural networks. A cycling procedure is used for the estimation and the restoration in order to obtain the restoration image of the higher quality. The simulation results indicate that the method is effective for the badly degraded image with less quantisation noise.