Face recognition method and device based on gabor binary mode

Abstract

Embodiments of the present invention provide a face recognition method and device based on a Gabor binary pattern. The device includes: a threshold determination module, which adopts Fisher's criterion to obtain the discriminant factors of all first filter response images in the training image set, and determines the pixel point thresholds at each scale in each direction according to the discriminant factors; the filter processing module, to be processed The image is subjected to Jiabo filtering to obtain the second filter response image in each preset direction and scale; the determination module determines the LGBP binary image of the second filter response image according to the pixel threshold of each second filter response image The acquisition module obtains the feature vector of the image to be processed according to the LGBP binary image; the recognition module obtains the similarity between the image to be processed and the training image according to the above feature vector and the feature vector of any training image in the training image collection, and according to the similarity degree threshold to get the recognition result. The embodiments of the present invention can improve the ability to identify human faces.

Description

technical field [0001] Embodiments of the present invention relate to image processing and pattern recognition technologies, and in particular to a face recognition method and device based on a Gabor binary pattern. Background technique [0002] Because face recognition technology is intuitive and irreproducible, it is widely used in security inspection systems, access control systems, attendance systems, intelligent robot systems, and virtual game systems. The basic concept is to detect the face area from an image or video containing a face; select and extract feature descriptors with strong face discrimination; and then design a classifier according to the selected features to realize face recognition. [0003] In the prior art, facial expression recognition and illumination detection methods based on two-dimensional multi-scale local Gabor Binary Patterns (Multi-scale BlockLocal Gabor Binary Patterns, MB-LGBP) features and illumination detection are used for face recognit...

AI Technical Summary

Problems solved by technology

[0004] However, the use of the above recognition technology for face recognition may cause the loss of image texture jump features, which will eventually lead to low identification ability

Images