Biometric video playback attack detection method based on gray scale change

A technology of grayscale change and detection method, applied in the field of biometrics, can solve problems such as usability and reliability need to be improved, difficulty in meeting practical application requirements, poor user experience, etc.

Active Publication Date: 2021-05-04
TIANJIN UNIV OF SCI & TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these methods have their own shortcomings. Some require a large number of complex calculations, some require special equipment support, and some methods have poor user experience and are difficult to meet the actual application requirements of various complex occasions. The ease of use and reliability need to be improved.

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
  • Biometric video playback attack detection method based on gray scale change
  • Biometric video playback attack detection method based on gray scale change
  • Biometric video playback attack detection method based on gray scale change

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0047] Take the video playback attack detection in the face recognition system as an example, the frame rate is 15fps. The specific detection process is as follows:

[0048] Step 1. Extract the relevant data of each gray level

[0049] First, the position of the sample in the video is determined by using the corresponding method in the original video, such as face detection in complex background and other technologies. From the video where the position of the face has been located, the feature points of the face are located and tracked, and the feature vector is extracted according to the coordinates of the located and tracked feature points.

[0050] The specific steps for extracting the classifiable signals above are as follows:

[0051] 1. The initial positioning of the recognized object in the video. For example, in face recognition, it is first necessary to detect faces in complex backgrounds.

[0052] 2. Select a frame of image, calculate the global average gray value...

Embodiment 2

[0060] The light source in Example 1 is changed to 808nm near-infrared light, the video acquisition device uses a common usb network camera, and other parameters and methods are the same as those in Example 1 to achieve the same recognition effect.

Embodiment 3

[0062] Using the same video sampling and lighting conditions as in Example 1, change the second step in Example 1 to calculate the global average gray value of each frame in 10 consecutive frames of pictures, denoted as G1 ~ G10, and calculate these 10 The average value of the value is G, and the average gray value of the face ROI in each frame of the 10 consecutive frames is calculated, and the average value of these 10 values ​​is calculated as F1, and the variance of the G1~G10 sequence is calculated as Δ, set Threshold d=Δ*2. Similarly, after changing the lighting conditions, delay for 500 milliseconds, count the average gray value of the face ROI in each frame of 10 consecutive frames, and calculate the average of these 10 values ​​as F2. Calculate the factor f according to G, F1, and F2, and compare it with the threshold d to determine the video playback attack.

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 present invention relates to a biometric video playback attack detection method based on grayscale changes. Its main technical features are as follows: a video signal including a face area is collected by a video acquisition device, and additional light irradiation is applied to the recognized face area. Obtain the change of the average gray level of the face area from the video signal as the identification basis; according to the ratio of the average gray level change of the face area to the average gray value of the scene as a standard parameter, name it as the gray change factor and set Set a threshold, classify and judge according to whether the gray scale change factor exceeds the set threshold, and judge whether it is a live human face or a video playback. The invention is based on the detection of the change of the average gray level of the face area under changing lighting conditions, and has high reliability; the detection function can be realized by using ordinary video acquisition equipment, and has the characteristics of low cost and simple and easy-to-implement algorithm, and can meet different actual requirements of the occasion.

Description

technical field [0001] The invention belongs to the technical field of biometrics, and relates to video playback attack detection in face recognition, in particular to a biometric video playback attack detection method based on grayscale changes. Background technique [0002] With the development of science and technology and the gradual maturity of some technologies in the field of biometric technology in recent years, biometric (authentication) technology has gradually been widely used. Biometric technology refers to a technology based on some biological characteristics of the human body, including physiological characteristics and behavioral characteristics, to identify and distinguish individual identities. At present, the application of this technology mainly relies on the recognition of physiological characteristics, usually including face recognition, finger (palm) print recognition, iris recognition, etc. The uniqueness of biological characteristics has been discove...

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): G06K9/00
CPCG06V40/166G06V40/168G06V40/172G06V40/45
Inventor 刘建征杨巨成杨华易赵婷婷陈亚瑞
Owner TIANJIN UNIV OF SCI & TECH
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