Face feature extraction method of self-adaptive extension cross type local binary pattern

Abstract

The invention discloses a face feature extraction method of a self-adaptive extension cross type local binary pattern. An existing method is not stable enough in feature extraction result and not strong in robustness due to fixed threshold value or single and unbalanced key point sampling. The method comprises the following steps: firstly, preprocessing an input face image to obtain a standard single face grayscale image; determining a local neighborhood and a central pixel; the method comprises the following steps: selecting four inner-layer pixels and four outer-layer pixels, calculating anadaptive threshold according to a normalized linear weighting mode, taking any inner-layer pixel in eight pixels as a starting point, sequentially taking all pixels from the inner-layer pixels to theouter-layer pixels at intervals, comparing with the adaptive threshold, encoding according to a comparison result, and converting into a decimal encoding value; and traversing all the pixels to obtaina local binary pattern coding feature map of the face. According to the method, under various interferences such as illumination, postures, expressions and shielding, the extracted features show goodrobustness, and the method has practical application value for face recognition.

Description

technical field [0001] The invention belongs to the technical field of biometric feature recognition and information security, and relates to a face feature extraction method of self-adaptive extended cross-type local binary mode. Background technique [0002] In recent years, face recognition has been widely used in real life because of its stability, non-contact and easy access, such as subway security check, bank identity verification, security monitoring and so on. However, the actual application scene environment is complex and changeable, and changes in lighting, expression, occlusion, posture and other conditions will significantly affect the performance of face recognition. [0003] Face recognition is a biometric recognition technology that uses computer vision to find faces in images or videos and identify their true identities. Face recognition mainly has the following steps: face detection, face representation and face matching, etc. [0004] The face feature e...

AI Technical Summary

Problems solved by technology

However, due to the fixed threshold or the single and unbalanced sampling of key points in the existing methods, the feature extraction results are not stable enough and the robustness is not strong.

Images