A Face Recognition Method Based on Laplacian Logarithmic Face and Convolutional Neural Network

Abstract

The present invention provides a face recognition method based on Laplacian logarithmic face and convolutional neural network, comprising: S1, obtaining and preprocessing the face image to be recognized; S2, judging whether the number of face images in the database reaches a predetermined value value, if not reached, execute S3, otherwise execute S4; S3, use the Laplacian logarithmic face algorithm to extract face features from the face image to be recognized, and then calculate the extracted face features and each face image in the database The chi-square distance between the corresponding face features, and output the face image with the smallest chi-square distance; S4, using the pre-trained convolutional neural network to extract face features from the pre-processed face image to be recognized; then Calculate the cosine distance between the extracted face features and the face features corresponding to each face image in the database, and output the face image with the smallest cosine distance. The invention can realize rapid face recognition, has high recognition accuracy, and has important significance for monitoring, anti-terrorism and the like.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to a face recognition method based on Laplacian logarithmic face and convolutional neural network. Background technique [0002] As an important way of human-computer interaction, biometric identification technology uses the inherent attributes or behavioral characteristics of the human body to realize human-computer interaction and identification through computer processing and analysis. It is not easy to forge, portable, and easy to use. Unique, highly reliable and stable verification path. In the field of biometrics, technologies that are widely researched and applied include: fingerprint and palmprint recognition, iris recognition, face recognition, behavior recognition and voice recognition. Among them, the recognition accuracy of fingerprint, palmprint and iris recognition is high, but this recognition method is active contact, requires the active cooperation of the ...

AI Technical Summary

Problems solved by technology

Among them, image noise has a great impact on face feature extraction, which will seriously reduce the face recognition rate.

For active face recognition systems, the influence of occlusion, expression and facial posture can be eliminated to a certain extent by means of manual intervention, but the controllability of illumination changes in face image acquisition is low, which is often used in face recognition systems. problems

Images