Image denoising method and device based on generative adversarial network

Abstract

The invention discloses an image denoising method and device based on a generative adversarial network. The method comprises the following steps: acquiring original image data; performing segmentation processing on the original image data to obtain first image data; generating second image data from the first image data through an image denoising model; and outputting the second image data. The image denoising method and device solve the technical problems that an image denoising method in the prior art usually only conducts denoising operation on image data through a fixed denoising rule and does not conduct rule change and improvement according to image input and output historical data so as to be more suitable for a current use scene, and the image denoising effect is improved.

Description

Technical field [0001] The present invention relates to the field of image data processing, and in particular, a method and apparatus based on an image denier of the generating counterfeit network. Background technique [0002] With the continuous development of intelligence, people in today's society use intelligent methods to increase learning, work, quality and efficiency, through intelligent means, can bring different technical effects of traditional treatment methods. [0003] At present, when the image data is denied noise, the image data acquired by the image acquisition device is usually used to remove the image data collected by the image acquisition device, and the image quality of the noise will be significantly improved. The user performs subsequent image analysis, but the traditional image denier method often makes the image data to noise operation only by the fixed denier rule, without the change and improvement of the rule of the image input and output historical d...

AI Technical Summary

Problems solved by technology

[0005] Embodiments of the present invention provide an image denoising method and device based on a generative adversarial network, to at least solve the problem that image denoising methods in the prior art often only perform denoising operations on image data through fixed denoising rules, while The rules will not be changed and improved according to the historical data of image input and output, so as to better adapt to the current usage scenario and increase the effect of image denoising.

Images