Multispectral Face Detection for Spoof-Resistant Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Facial recognition systems are vulnerable to spoofing, where a nefarious user can gain unauthorized access by presenting a fake face, such as a printed image or three-dimensional model, fooling the system.
Innovation Solution
A multispectral camera captures image data in two electromagnetic spectrum regions, such as infrared and visible, to detect differences characteristic of human skin that are unreproducible by spoofing objects, using a computing device to analyze the image data and determine if it corresponds to a real or spoofed face.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If facial recognition systems use standard image capture, then the system is easy to operate and quick to implement, but the system becomes vulnerable to spoofing attacks
Solution Approach 1:
The patent transitions from standard visible light imaging to multispectral imaging by adding infrared spectral dimension. The camera captures images in both visible and infrared spectrum regions, creating a multidimensional feature space where spoofed faces can be differentiated from real faces based on spectral characteristics that are impossible to replicate with conventional printing or modeling materials.
Solution Approach 2:
The system changes the physical parameters of image capture by utilizing different spectral regions (visible and infrared). Real human skin exhibits distinct spectral reflection and absorption characteristics in these regions, while spoofing materials do not. By analyzing feature distances across these spectral parameters, the system achieves reliable spoof detection.
2Measurement precision
If the system captures images in multiple spectral regions, then spoofing detection accuracy improves, but the energy consumption and processing time increase
Solution Approach 1:
The system performs preliminary action by pre-computing and storing reference feature distances for authentic faces in the database during system initialization. When authentication is required, the system only needs to compute feature distances for the test image and compare against pre-established thresholds, significantly reducing real-time processing energy consumption while maintaining high accuracy.
Solution Approach 2:
The system creates feature distance representations (mathematical copies) of spectral characteristics rather than storing and comparing entire multispectral images. By computing and comparing feature distances in a reduced-dimensional feature space, the system achieves accurate authentication with minimal energy expenditure during the verification phase.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system provides robust facial recognition resistant to spoofing, enhancing security by reducing the risk of identity theft and unauthorized access.
Implementation Method 1
A multispectral camera captures image data in two electromagnetic spectrum regions, such as infrared and visible
Implementation Method 2
captures image data in two electromagnetic spectrum regions, such as infrared and visible
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Examples are disclosed herein that relate to detecting spoofed human faces. One example provides a computing device comprising a processor configured to compute a first feature distance between registered image data of a human face in a first spectral region and test image data of the human face in the first spectral region, compute a second feature distance between the registered image data and test image data of the human face in a second spectral region, compute a test feature distance between the test image data in the first spectral region and the test image data in the second spectral region, determine, based on a predetermined relationship, whether the human face to which the test image data in the first and second spectral regions corresponds is a real human face or a spoofed human face, and modify a behavior of the computing device.