Facial image reconstruction method and device and face recognition system

Abstract

The embodiment of the invention discloses a facial image reconstruction method, a facial image reconstruction device and a face recognition system, and relates to the pattern recognition technology. The facial image reconstruction method comprises the following steps of: performing pixel level alignment on a target image; judging the illumination type of the aligned target image; acquiring a reconstruction model of the illumination type of the aligned target image; reconstructing the aligned target image according to the reconstruction model, and acquiring a standard illumination image of thealigned target image; and correcting the standard illumination image of the aligned target image into an original shape, and acquiring a standard illumination image with the same shape as the target image. The facial image reconstruction method, the facial image reconstruction device and the face recognition system are mainly used for face recognition.

Description

technical field [0001] The invention belongs to pattern recognition technology, in particular to a face image reconstruction method, device and face recognition system. Background technique [0002] As an important identification technology, face recognition has broad application prospects in security, finance, human-computer interaction, information, education and many other fields. Since the collected lighting environment cannot be controlled, the lighting conditions of the template image and the test image are always different, which makes it difficult for the current image-based face recognition system to cope with the impact of lighting changes. [0003] The improvement of the face recognition system under illumination changes is mainly based on two categories: (1) illumination template difference: the face image under illumination is decomposed into face illumination invariants and illumination templates; (2) illumination subspace: simulated from the illumination image...

AI Technical Summary

Problems solved by technology

[0004] Under the first type of method, since the illumination template is the projection of the light source on the face image along the normal direction of the face, it is not independent of the illumination invariant, and the illumination invariant image obtained by direct template difference makes The face structure and texture information are greatly weakened, often only the contour information is left, and it is difficult to describe the overall image features; under the second type of method, since the illumination subspace of the target image is trained from the reference image, the reconstructed image of the target image is also Depends on the reference image, which leads to image distortion, and this type of method directly aligns the image with the eye position. Due to the differences between different individual images, the image reconstruction distortion that is not accurately aligned is more serious

Images