Method for rebuilding super-resolution human face image by position block nonlinear mapping

Abstract

The invention relates to a method for rebuilding a super-resolution human face image by position block nonlinear mapping. The method comprises the following steps of: segmenting the human face image for training and testing high resolution and low resolution into corresponding high-resolution position block images and low-resolution position block images which are provided with overlapped regions from structural information and detail information of a roughly-aligned human face image; rebuilding super-resolution position block images corresponding to testing low-resolution position block images at each corresponding position by using a nonlinear mapping relation which is between the high-resolution position block images and the low-resolution position block images and learned from training data; and splicing the super-resolution position block images to obtain the final super-resolution human face image. According to experiments on a standard database, the method provided by the invention has a good visual effect and high objective estimation quality.

Description

technical field [0001] The invention belongs to image processing technology, and in particular relates to a face image super-resolution reconstruction method based on nonlinear mapping of position blocks. Background technique [0002] Image super-resolution (Super Resolution, SR) refers to the process of obtaining a high-resolution (High Resolution, HR) image from one or a series of low-resolution images (Low Resolution, LR). In recent years, video surveillance has been widely used in important places such as banks and airports. But in many cases, the resolution of face images obtained by monitoring equipment is too low to be recognized directly, so the research on face image super-resolution has practical significance. [0003] The general natural image super-resolution algorithm can give a lot of inspiration to the face image super-resolution algorithm. But the face image is a special kind of image, which has its specific global structure. Therefore, there are certain l...

AI Technical Summary

Problems solved by technology

Therefore, there are certain limitations in extending the general natural image super-resolution algorithm that does not consider the global structural information to the face image super-resolution problem, and there is still room for further improvement in the reconstruction effect.

However, the least squares method used by Ma when recovering detailed information only considers the linear correlation components between high and low resolution position blocks, while ignoring the nonlinear correlation components. In order to improve the performance of super-resolution reconstruction, it is necessary to find a better method to fix the problem here

Images