Face silence living body detection method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a face silence living body detection method and device, electronic equipment and a storage medium, and belongs to the technical field of image processing, and the face silence living body detection method comprises the steps of obtaining a normalized face image; performing feature extraction on the normalized face image by using a convolutional neural network, calculating a living body judgment score according to the extracted features, wherein cavity convolution and multi-scale multi-stage supervised training are used during feature extraction; and predicting a living body probability according to the living body judgment score. According to the embodiment of the invention, the cavity convolution is used for replacing a traditional convolution method, the scale of the feature map is not reduced, meanwhile, the receptive field information of the feature map can be increased, the in-vivo detection features with better generalization can be obtained, a multi-stage supervised method is used, the generalization of the features is further enhanced, and a better in-vivo detection effect is obtained. According to the embodiment of the invention, the cost is low, the generalization is good, and the detection effect on various prostheses is good.

Description

Technical field [0001] Technical Field The present invention relates to machine learning, and particularly to a living body face silence detecting method, apparatus, electronic apparatus and a storage medium. Background technique [0002] Application Face payments, Face Unlock technology in the financial, security and other fields, making the password required to face more and more, especially in the face of anti-false question. Face mainly refers to the use of anti-fake facial prosthesis system attacks the face, and then to fool the system to give the object related permissions. Prosthetic face there are three, namely print category of the prosthesis, prosthetic and 3D video type mask prosthesis. Print category prosthesis refers to the face after another remake of a paper print out the results, referring to the video category prosthesis is obtained prosthesis remake the image of the video image, and 3D masks prosthesis is referring to the wearing 3D simulation collect prosthesis...

AI Technical Summary

Problems solved by technology

The data production of this scheme is relatively complicated and the detection effect on 3D mask attacks is not good

Images