A face image super-resolution method based on dictionary asymptotic update

Abstract

The invention discloses a facial image super-resolution method based on dictionary asymptotic updating. In the training stage, super-resolution reconstruction is carried out on low-resolution facial images of a low-resolution facial image training set with a one removing method to obtain a low-resolution intermediate dictionary; the low-resolution intermediate dictionary serves as a new low-resolution facial image training set for inputting, and reconstruction is carried out to obtain a new low-resolution intermediate dictionary; the process is repeated, and finally multiple low-resolution intermediate dictionaries are obtained. In the testing stage, according to the input low-resolution facial images, the prior low-resolution intermediate dictionary and a high-resolution facial image training set, super-resolution reconstruction is carried out on the input low-resolution facial images to obtain pre-estimated high-resolution facial images; the process is repeated, and finally the high-resolution facial images are reconstructed. By means of the facial image super-resolution method, the reconstruction effect which is high in quality and close to a true condition can be achieved.

Description

technical field [0001] The invention relates to the field of image super-resolution, in particular to a face image super-resolution method based on dictionary asymptotic update. Background technique [0002] Face image super-resolution technology refers to using the training library composed of low-resolution face sample images and high-resolution face sample images to learn the relationship between low-resolution images and high-resolution images, and use the learned Relation The process of predicting a high-resolution face image given an input low-resolution face image. It has a wide application background in the fields of intelligent video surveillance, digital entertainment, face synthesis and recognition, etc. [0003] Since the image super-resolution technology was proposed in 1984, it has attracted extensive attention from scholars in the fields of computer vision and machine learning. Freeman et al. (Document 1: W. Freeman, E. Pasztor, and O. Carmichael. Learning l...

AI Technical Summary

Problems solved by technology

[0005] Whether using the least squares representation method, sparse representation method, local constraint representation method or iterative neighborhood embedding method, they all try to maintain the low-resolution manifold structure to the reconstructed high-resolution space, and do not fully explore the The geometric structure information of the original high-resolution manifold space is affected by the degradation process, which makes the reconstruction results unreliable

Images