Facial Recognition Liveness Detection via Corneal Glint
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Facial recognition systems are vulnerable to spoofing attempts, where unauthorized users can gain access by presenting printed or digital images of authorized users, leading to erroneous authentication.
Innovation Solution
Implementing anti-spoofing techniques that emit light to detect corneal glint and analyze facial landmarks for signs of liveness, such as eye reflections, pitch, and yaw angles, to differentiate between real and fake attempts, and requiring specific gestures like blinking or head movements to authenticate users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If facial recognition systems authenticate users based on facial images, then user access is enabled, but the system becomes vulnerable to spoofing attacks using printed or digital images
Solution Approach 1:
The system performs liveness detection before final authentication by analyzing corneal glint characteristics in real-time captured images. This preliminary check verifies that the subject is a live human eye rather than a printed or digital image, preventing spoofing attacks before authentication occurs.
Solution Approach 2:
The patent replaces traditional mechanical or manual verification methods with optical analysis of corneal glint patterns. By using light reflection properties of the cornea, the system automatically detects liveness without requiring physical contact or complex mechanical verification devices.
2Measurement precision
If anti-spoofing techniques emit light to detect corneal glint, then spoofing detection accuracy is improved, but device complexity and energy consumption increase
Solution Approach 1:
The system uses the device's existing flash or light source, which serves dual purposes: providing illumination for the camera to capture facial images and generating corneal glint for liveness detection. This eliminates the need for separate dedicated light-emitting components, reducing device complexity while maintaining detection accuracy.
Solution Approach 2:
The system utilizes the camera flash or ambient light to create corneal glint, and the same camera sensor detects the glint pattern. The existing optical components serve multiple functions, and the system self-verify liveness using resources already present in the device without requiring additional specialized hardware.
3Reliability
If multiple facial images are captured and analyzed for liveness gestures, then authentication reliability is improved, but processing time and energy consumption increase
Solution Approach 1:
The system captures multiple facial images during a single authentication attempt, analyzing sequences of images to detect liveness gestures such as blinking or head movements. By processing a series of images rather than requiring separate authentication attempts, the system verifies liveness more reliably without significantly increasing total processing time.
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
Significantly reduces the chances of unauthorized access by accurately distinguishing between real and fake facial images, enhancing security and conserving computing resources by minimizing unnecessary facial recognition processes.
Implementation Method 1
emit light to detect corneal glint and analyze facial landmarks for signs of liveness, such as eye reflections
Data Source
Figure 1A~1B
Figure 2
Figure 3
AI summary
An example method includes receiving a first image and a second image of a face of a user, where one or both images have been granted a match by facial recognition. The method further includes detecting a liveness gesture based on at least one of a yaw angle of the second image relative to the first image and a pitch angle of the second image relative to the first image, where the yaw angle corresponds to a transition along a horizontal axis, and where the pitch angle corresponds to a transition along a vertical axis. The method further includes generating a liveness score based on a yaw angle magnitude and/or a pitch angle magnitude, comparing the liveness score to a threshold value, and determining, based on the comparison, whether to deny authentication to the user with respect to accessing one or more functionalities controlled by the computing device.