Unlock instant, AI-driven research and patent intelligence for your innovation.

A multi-view face liveness detection method and system

A liveness detection and multi-view technology, applied in the field of multi-view human face liveness detection method and system, can solve the problems of hardware cost reduction, hardware dependence, and active cooperation dependence, etc., to achieve hardware cost reduction, big Lu Stickiness, experience-enhancing effects

Active Publication Date: 2022-07-22
CHONGQING UNIV OF POSTS & TELECOMM
View PDF11 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] The present invention aims to overcome the problems of the aforementioned existing human face liveness detection technology that is highly dependent on human active cooperation and hardware, and provides a new multi-view face liveness detection method and system, through multiple The information modeling of the key points of the face under the viewpoint can make it possible to identify whether the object is a real person without active cooperation. It is also very robust to the deflection of the face, which improves the experience in the face recognition process, and only Ordinary camera, hardware cost will be greatly reduced

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
  • A multi-view face liveness detection method and system
  • A multi-view face liveness detection method and system
  • A multi-view face liveness detection method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0062] Due to the problems that the existing human face liveness detection technology has a strong dependence on the active cooperation of people and a large dependence on hardware, this embodiment provides a multi-view face liveness detection method, such as figure 1 shown, the specific steps are as follows:

[0063] Step A: Collect face pictures from the front, left and right angles at the same time, use the face detection algorithm in the Dlib library to detect the pictures taken by all cameras, detect the length of the face in the input picture, and cut out a living body The range of the face for the preliminary judgment condition. In this embodiment, the value of the top function and the value of the bottom function in the detection result of the Dlib library are used as the length of the face, and the length of the face that meets the range specified by the acquisition module is between 258 and 125, and when it is greater or less than this range, the prompt module is pas...

Embodiment 2

[0087] like image 3 As shown in the figure, this embodiment provides a multi-view face live detection method, including: a collection device and a data processing device, and the data processing device includes: a preliminary judgment device, a key point coordinate acquisition device, a first calculation device, and a preprocessing device , sorting device and output device.

[0088] The collection device simultaneously collects face pictures from three angles of the front, the left and the right;

[0089] The preliminary judgment device performs preliminary judgment on the collected face pictures, and selects the face pictures whose face pictures collected from three angles meet the preliminary judgment conditions of the living body;

[0090] The prompt module is used to prompt the detected object to adjust the position, including the face, when the preliminary judgment conditions of the living body are not met.

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 discloses a method and system for detecting a face living body from multiple viewpoints. First, face pictures are collected from multiple angles at the same time, and preliminary judgment of living body is performed on the collected face pictures. If the face pictures collected from multiple angles meet the preliminary judgment of living body condition, then obtain the coordinate position of the face key points in each face image, calculate the face key point distance vector in each face image based on the coordinate position, and preprocess the face key point distance vector to obtain the feature key point Distance vector; arbitrarily select a set of feature key point distance vectors and input them into the trained classification model for classification; output whether the face in the currently collected face photo is from a real person according to the classification result. Simultaneous acquisition from multiple angles has great robustness to the deflection of the face, improves the experience in the face recognition process, and greatly reduces the hardware cost. It uses the feature key point distance vector input classification model for classification, and the amount of calculation is low. , the detection reliability is high.

Description

technical field [0001] The invention relates to the technical field of biometric identification, in particular to a method and system for multi-view face living detection. Background technique [0002] Face liveness detection is to analyze the incoming face image, so as to determine whether the current incoming face image is collected from a real person or a remade face image. It is usually a part of face recognition and is often used in the security field to prevent face recognition verification by retaking photos. [0003] The prior art face liveness detection methods can be roughly divided into two categories: [0004] (1) The method of active cooperation, which requires people to make corresponding actions according to the instructions of the computer, and the computer judges whether the current picture is a real person according to whether the action is standard or not. [0005] (2) The method of silent recognition, which can recognize a real person without active coo...

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): G06V40/16G06V40/40
CPCG06V40/165G06V40/171G06V40/172G06V40/45
Inventor 米建勋乔冬陈涛向菲钱基业江金洋
Owner CHONGQING UNIV OF POSTS & TELECOMM