Image Deblurring Algorithm by using Conditional Generation Adversarial Network

Deblurring from a motion blurred image has been studied in recent years. Recently, convolution neural network (CNN) has been used widely, which can be used on finding the blur kernel or the latent sharp edge of a blurred image. The last five years, the generative adversarial network (GAN) performs well on style transformation. It is often difficult to maintain a relatively static state between shooting equipment and the photographed object in real life and industrial production, so motion blur is easy to appear in the image taken. To solve this problem, we can use DeblurCGAN to deblur motion blur images. The model can effectively remove the motion blur in moving images in the experimental process. At the same time, compared with the traditional blur kernel, the algorithm not only can get a clearer image but also to restore the texture and detail in the image. Both in the image evaluation index peak signal-to-noise ratio and structural similarity measurement can be verified this method can achieve a better deblurring effect.