Adaptive training library-based block domain face super-resolution reconstruction method

Abstract

The invention provides an adaptive training library-based block domain face super-resolution reconstruction method. The new method based on learning comprises the following steps of: firstly, uniformly dividing high resolution images and low resolution training images into blocks, and clustering in a block set for which position is considered as the unit; and then, adaptively selecting corresponding position training block set for each block of the input images, screening the adaptive training set according to neighboring selection and adaptive class selection, and recovering high frequency details in a low resolution block by executing principal component analysis coefficient projection in the adaptive training set. By the method, the problem of noise in super-resolution recovery of the face image (such as a monitoring image) with serious noise points is solved or relieved.

Description

technical field [0001] The invention relates to the field of image processing (mainly image restoration), and aims at the demand for face image restoration in low-quality surveillance video, and specifically relates to a block-domain face super-resolution reconstruction method based on an adaptive training library. Background technique [0002] In recent years, with the rapid development of security monitoring systems, monitoring and forensics has played an increasingly important role in the fields of security prevention and crime forensics, among which face image forensics is one of the important concerns of monitoring and forensics. However, due to the serious blur and noise caused by the long distance between the camera and the target face, bad weather (rain, fog, etc.), poor lighting conditions, etc. in the surveillance video, the usable pixels of the face image captured in the surveillance video are extremely low, and the image recovery , identification is often severel...

AI Technical Summary

Problems solved by technology

[0004] The face super-resolution problem is a problem with infinite solutions, because a low-quality image may correspond to many different high-quality images

Good results can be obtained when dealing with general face super-resolution problems, but when the image quality is very low, the pixels will be severely damaged and aliased, and redundant training information will seriously interfere with the super-resolution recovery results, and the effect is not impressive. people are satisfied

Images