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

Face anti-cheating method

An anti-spoofing, facial technology, applied in the direction of spoofing detection, neural learning methods, computer parts, etc., can solve the problems of 3D camera dependence, end-user unfriendly, time-consuming, etc.

Pending Publication Date: 2021-11-23
BLACK SESAME TECH CO LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

3D cameras rely on time-of-flight data and will incur additional costs, and utilizing predefined facial movements will be unfriendly to the end user and take extra time

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
  • Face anti-cheating method
  • Face anti-cheating method
  • Face anti-cheating method

Examples

Experimental program
Comparison scheme
Effect test

no. 1 example

[0034] A first embodiment utilizes a near-infrared (NIR) camera and a red-blue-green (RGB) camera to implement face anti-spoofing based on a combined mixed-channel input routed through a deep neural network. The NIR camera input can provide images invariant to lighting conditions, while the RGB camera can provide face color information. The first embodiment is based on a data set of 300 real subjects and 1000 spoofed subjects to obtain test results with enhanced accuracy (true positive rate (TPR) greater than 99.9%, false acceptance rate (FAR) = 10e-3.5 ).

[0035]A NIR camera consists of a NIR light source, a NIR transmission lens and a NIR responsive sensor. NIR cameras detect light with a spectrum of near-infrared wavelengths from 700 nm to 1400 nm, and are typically filtered with narrow NIR bandpass filters. Electronic versions of spoofing attacks such as photos or videos displayed on a phone, tablet or computer screen are rejected by NIR cameras because these spoofing a...

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

An exemplary face anti-cheating method includes: receiving a near-infrared face image having a near-infrared channel; receiving a red-green-blue face image having a red channel, a green channel, and a blue channel; generating a synthesized three-channel image based on the near-infrared channel, the red channel, the green channel and the blue channel; and training a deep neural network based on the synthesized three-channel image.

Description

technical field [0001] The invention relates to a face anti-spoofing method of a dual camera (ie a red-blue-green (RGB) camera and a near-infrared (NIR) camera). Background technique [0002] Existing face anti-spoofing methods extract texture features from images to train classifiers to distinguish real faces from spoofed faces. For example, local binary pattern (LBP) and image distortion analysis (IDA) enable training and testing on images taken under similar imaging conditions. However, the LBP / IDA approach will be sensitive to camera and lighting differences and will have poor generalization ability. [0003] Other anti-spoofing methods rely on depth information from 3D cameras, or utilize user collaboration to perform predefined facial movements. 3D cameras rely on time-of-flight data and would incur additional costs, and utilizing predefined facial movements would be unfriendly to the end user and take additional time. Contents of the invention [0004] An exempla...

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
IPC IPC(8): G06K9/00G06K9/32G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06V40/164G06V40/166G06V40/40G06V10/82G06V10/143G06V40/168H04N5/33H04N23/45
Inventor 刘晓旻张磊顾群胡子龙
Owner BLACK SESAME TECH CO LTD