Method of using generative adversarial network in image restoration

Abstract

The invention discloses an image restoration model PRGAN. The image restoration model PRGAN is composed of two independent generative adversarial network modules. An image restoration network module PConv-GAN is formed by combining part of a convolutional network and an adversarial network and is used for restoring an irregular mask, and meanwhile, making the overall texture structure and color ofan image closer to those of an original image according to feedback of a discriminator. In order to solve the problems of local chromatic aberration and mild boundary artifacts of an image caused bythe deficiency of a restoration network module, the invention designs an image optimization network module. The image optimization network module Res-GAN combines a deep residual network with an adversarial network, and the image optimization network module is trained by combining the adversarial loss, the perception loss and the content loss, so that the information of the non-missing area in theimage is reserved, the consistency of the texture structure of the image in the non-missing area is kept, and the purposes of eliminating the local chromatic aberration phenomenon and solving the pseudo boundary are achieved.

Description

technical field [0001] The invention relates to the field of computer vision, specifically a two-stage image restoration system, which is composed of two independent networks. Among them, an image inpainting network is formed by improving some convolutions. In addition, an image optimization network is proposed to solve the problem of local color difference in the image obtained after the first step operation. The synergy of the two networks improves the qualitative and quantitative indicators of the experimental results, and the visual effect of the obtained images is significantly improved. Background technique [0002] Image inpainting, i.e. filling in lost pixel regions of an image, plays an important role in the field of computer vision. It is applied in many research fields such as image editing, image rendering. The core of image inpainting is how to fill in missing areas to achieve semantically reasonable and visually realistic results. [0003] The principle of e...

AI Technical Summary

Problems solved by technology

However, these methods based on deep convolutional networks usually have problems such as boundary artifacts, image distortion, or blurry textures that are inconsistent with surrounding areas when repairing irregular damage, which may be due to the relationship between the learned context information and the missing area. caused by the invalid correlation of

[0005] The partial convolutional network proposed by Liu et al. can make the result of the convolution only depend on the non-damage area. Through automatic mask update, only the effective context correlation information is left in the feature map, which makes the image texture generated by the missing area It maintains a high degree of consistency with the surrounding image texture, which solves the problems of image blur and inconsistent texture structure, but the generated image still has the problem of mild boundary artifacts and local color inconsistency

Images