Liveness detection method

A technology of living body detection and living body, which is applied in the field of recognition of live face images, and can solve the problems of face recognition video or photo attacks, and the inability to accurately identify faces in live face images, etc.

Active Publication Date: 2022-03-08
成都智汇脸卡科技有限公司
View PDF13 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0031] The purpose of the present invention is to provide a living body detection method to solve the problem of being vulnerable to video or photo attacks in the current face recognition process, so that it is impossible to accurately identify live human face images and fake human faces

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Liveness detection method
  • Liveness detection method
  • Liveness detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0093] The living body detection method of the embodiment of the present invention comprises the following steps:

[0094]Step 1. Collect a sample and set the image tag value of the sample. The sample includes multiple live face images and multiple non-live face images. The corresponding picture tag values ​​of the live face images are all 1, and the non-live face images The corresponding image label values ​​are all 0;

[0095] Step 2, randomly select some samples collected and use them as a training set, and calculate the Laplacian eigenvalues ​​and Tamura texture eigenvalues ​​of each sample in the training set;

[0096] Step 3, splicing the Laplacian eigenvalues ​​and the Tamura texture eigenvalues ​​of each sample in the calculated training set respectively to obtain the splicing result, and then proceed to step 4;

[0097] Step 4. Use the splicing results to train a model, wherein the model is preferably a neural network model, and the neural network model is stable and...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a living body detection method, which relates to the field of face recognition. The present invention solves the problem that the current face recognition process is vulnerable to video or photo attacks, so that it is impossible to accurately identify real faces and fake faces. The Laplace eigenvalues ​​and Tamura texture eigenvalues ​​of each sample in the training set are stitched together, and then the model is trained using the splicing results, and the initial output range of the model is set, and then the Laplace eigenvalues ​​of each sample in the verification set are calculated separately. Stitch eigenvalues ​​and Tamura texture eigenvalues ​​and stitch them, and transfer the stitching results to the model for calculation, and judge whether the calculation results are within the initial output range, and if so, record the relevant parameters of the model at this time , if not, calculate the loss value of the calculation result and reversely transmit the loss value to the model, and adjust the relevant parameters of the model according to the loss value.

Description

technical field [0001] The invention relates to face recognition technology, in particular to how to more accurately recognize live human face images in access control systems or advertising machines. Background technique [0002] The access control system is a system that controls the entrance and exit passages. The early access control systems are usually called electronic locks, mainly electronic magnetic card locks and electronic combination locks, but the information of magnetic card locks is easy to copy, the wear between the card and the card reader is large, the failure rate is high, and the safety factor is low. The problem of combination lock is that password is easy to reveal, and has no way of checking again, and safety factor is very low. With the development of inductive IC card technology and biometric technology, the access control system has developed by leaps and bounds and has entered a mature stage. There are various technical systems, and they have the...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06V40/40G06V40/16G06V10/774G06V10/80G06V10/82G06K9/62
CPCG06V40/168G06V40/45G06F18/253G06F18/214
Inventor 姜尧岗孙晓刚林云康鑫李泽原万磊解至煊谢文吉
Owner 成都智汇脸卡科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap