An image super-resolution reconstruction method based on multi-scale generation countermeasure network

Abstract

The invention discloses an image super-resolution reconstruction method based on a multi-scale generation antagonistic network, which adopts a multi-scale SENet module as a generator and finally formsa multi-scale generation antagonistic network. Then the target loss function is determined and the pre-training process of the generator is completed so as to improve the convergence speed of the multi-scale generation countermeasure network. The feature extraction of high-frequency information of LR input image is realized by the multi-scale generation countermeasure network, then the HR image is reconstructed with the result of bi-cubic interpolation; the discriminator discriminates the true and false of the reconstructed image, and weighs the loss of mean square error and the loss of antagonism as the final adjustment objective function; after adjusting the output HR image by the objective function, the whole image reconstruction process is completed. The invention can extract more high-frequency information details of the LR input image, thereby enabling the HR image with better display effect to be generated after the image super-resolution reconstruction.

Description

technical field [0001] The invention relates to an image super-resolution reconstruction method based on a multi-scale generation confrontation network. Background technique [0002] The image super-resolution (Single image super-resolution, SISR) reconstruction task refers to learning more image details from a low-resolution (Low-resolution, LR) image input to generate a high-resolution (High-resolution, HR) image. Since HR images can learn important details of the original images, SISR technology is widely used, including video surveillance, medical diagnosis, face recognition, etc. Traditional image super-resolution methods are mainly interpolation algorithms, such as bicubic interpolation and nearest neighbor interpolation, which estimate the value of unknown pixels in SR images by using fixed or structure-adapted kernel functions. Although the interpolation algorithm is more efficient, the result obtained is blurred and the edge details are smoother. [0003] In recen...

AI Technical Summary

Problems solved by technology

[0005] However, the above-mentioned methods are poor in extracting high-frequency information from low-resolution input images, so that the display effect of high-frequency information details on low-resolution input images is not good after image super-resolution reconstruction.

Images