Super-resolution image restoration method based on generative adversarial network

Abstract

The invention discloses a super-resolution image restoration method based on a generative adversarial network, and the method specifically comprises the steps: 1, collecting and arranging real image data, processing the collected images into images with the same size, and enabling the processed images with the same size to form a training data set; 2, constructing a generative adversarial networkmodel, wherein the generative adversarial network model comprises a generator and a discriminator; 3, marking each image in the training data set with a mask with the same size to obtain a to-be-restored image, importing the to-be-restored image into the generative adversarial network model, and training the generative adversarial network model in a weak supervised learning mode to obtain a trained image; and 4, inputting the training image into the U-Net generation network to repair the trained image to be repaired, and outputting a repaired high-resolution image. The problems that in the prior art, the resolution ratio is low after large-area image missing is repaired, and the visual effect is not coherent are solved.

Description

technical field [0001] The invention belongs to the technical field of computer vision and image processing, and in particular relates to a super-resolution image restoration method based on a generative confrontation network. Background technique [0002] In recent years, with the rapid development of computer graphics and computer vision, there are various methods of image restoration. Image restoration technology appeared as early as the Renaissance. There are works of art that are damaged due to improper preservation or other different reasons, and craftsmen do some filling work to maintain the integrity of the crafts through their mastered skills. With the rapid development of computer vision technology, digital image processing technology has gradually replaced manual image restoration technology. [0003] At present, image restoration is an important research content in the field of computer vision. Its purpose is to automatically restore the lost content according t...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a super-resolution image repair method based on a generative confrontation network, which solves the problems of low resolution and incoherent visual effects existing in the prior art after large-area missing images are repaired

Images