An Improved Image Deblurring Algorithm Based on GAN

In order to solve the problem that the limited number of original data sets cannot be fully close to the real data distribution, thus affecting the effect of image deblurring. This paper proposes the method of generating adversarial network image deblurring based on data set expansion. Firstly, the generated adversarial network is used to expand data sets; Secondly, in order to solve the problem, the expanded data set is expanded again by traditional methods to generate kuochong data set; finally, the GoPro data set, kuochong data set and Kohler data set are input into the network model for training and testing. The experimental results show that the peak SNR PSNR value increased by 19.8%, and the structural similarity SSIM value increased by 2.5%.