Super-resolution processing method of face image integrating global feature with local information

Abstract

The invention relates to a super-resolution processing method of a face image integrating a global feature with local information, in particular to a method for processing a low-resolution face image by an image pyramid to obtain a high-resolution face image. The method comprises the following steps: (1) establishing a controllable pyramid database of high-resolution face images; (2) solving a global high-resolution image of a low-resolution face image; (3) establishing a high-resolution residual face image database; (4) calculating and inputting global high-resolution images of the low-resolution face image; (5) calculating and inputting high-resolution residual face images of the low-resolution face image; and (6) calculating and inputting the final high-resolution image of the low-resolution face image. The method has the advantages of easy realization, strong robustness, high precision, etc., and can provide real-time reliable feature information for face identification.

Description

technical field [0001] The invention relates to a method for super-resolution processing of a human face image by fusing global features and local information, in particular to a method for processing a low-resolution human face image to obtain a high-resolution human face image by using an image pyramid. It belongs to the field of image processing. Background technique [0002] Human face research has been one of the classic problems in the fields of pattern recognition, computer vision and computer graphics. Since the human face is an extremely complex, multi-dimensional pattern, it is also a typical non-rigid pattern with a very complex physiological structure, and people are familiar with and sensitive to the human face, so the research on the human face is relatively rare. It is difficult to say. The first step in any face processing system is to detect the location of a face in an image. However, face detection from an image is an extremely challenging task due to i...

AI Technical Summary

Problems solved by technology

Baker and Kanade proposed the idea of ​​"Face Fantasy" in the paper Hallucinating faces of the IEEEInternational Conference on Automatic Face and Gesture Recognition (International Conference on Automatic Face and Gesture Recognition). They chose the horizontal and vertical directions of the Gaussian pyramid of the face image Derivative and Laplacian (Laplacian) pyramid, the result is better than Freeman and Hertzmann, but the face image obtained by super-resolution has larger noise in some parts

Liu et al.'s paper "A two-step approach to hallucinating faces: global parametric model and local nonparametric model (a two-part face fantasy method: Global parameter model and local non-parametric model)” proposed a two-step face super-resolution algorithm, using the idea of ​​combining the global parameter model and local non-parametric model, the result is smoother than Baker’s, but due to the use of the main Component Analysis (Principal Components Analysis, PCA) calculates the global model, and the result of super-resolution is not similar to the original high-resolution image

Images