Biometric Video Authentication Using Random Light Frequency Cues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing biometric authentication methods for remote access are vulnerable to frauds such as injection and presentation attacks, particularly inconvenient for users with disabilities, and require complex image analysis.
Innovation Solution
A method using a terminal with a light source and camera that applies a time-varying light signal with invariant parameters to authenticate users by analyzing the response of the captured object, employing spectral analysis and neural networks to verify the authenticity of the acquired video, without requiring specific user alignment or complex image analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If random gestures are required for authentication, then fraud detection capability is improved, but user convenience deteriorates
Solution Approach 1:
The patent applies periodic action by using a time-varying illumination signal with a specific frequency (e.g., 10 Hz) that modulates the light source. This periodic modulation creates a characteristic temporal signature in the captured video that is automatically detected by analyzing the time derivative of pixel values. The system does not require user gestures but instead relies on the periodic lighting pattern to distinguish real faces from fraudulent representations, thus maintaining user convenience while achieving fraud detection.
2Reliability
If complex image analysis is performed, then fraud detection accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent extracts only the essential temporal information from the video stream by computing the time derivative of pixel values in response to the periodic illumination. Instead of performing complex full-image analysis, the system extracts a simplified temporal signature that captures the key fraud-detection information. This extraction approach maintains high detection accuracy while significantly reducing computational complexity, as it focuses only on the temporal variation caused by the periodic lighting rather than analyzing entire image frames.
Solution Approach 2:
The patent changes the parameter being analyzed from spatial image features to temporal derivative features. By computing the time derivative of pixel values in response to periodic illumination, the system transforms the problem into detecting temporal patterns rather than analyzing complex spatial relationships. This parameter transformation simplifies the computational requirements while maintaining or improving fraud detection accuracy, as the temporal signature is a more direct indicator of authenticity.
3Measurement precision
If alignment with dedicated events is required, then measurement precision is improved, but ease of operation deteriorates
Solution Approach 1:
The patent uses periodic illumination with a known frequency to create a predictable temporal pattern in the captured video. This periodic action allows the system to estimate parameters (such as face presence and authenticity) without requiring precise temporal alignment or synchronization events. The periodic nature of the lighting ensures that the temporal signature repeats consistently, enabling robust parameter estimation even without dedicated alignment events, thus improving ease of operation while maintaining measurement precision.
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
This method effectively secures remote access by detecting frauds, ensuring user convenience and adaptability, while minimizing data exchange and computational load, and is applicable to various devices and environments.
Implementation Method 1
the terminal comprising a light source and a camera directed towards said user
Implementation Method 2
video acquisition by the camera
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The present invention relates to a method for securing access to a data source from a device (1) by a user of said device, the device comprising a light source (5) and a camera (3) directed towards said user, said method comprising steps of: - video acquisition by the camera (3); - control, during the acquisition step, of the light source (5) according to a time-varying signal of imposed and randomly fixed fundamental frequency; - evaluation, from a first series of images of the acquisition, of the presence of fraud by injection or recapture by: - estimation of the fundamental frequency of the observed time signal; - authorization to continue access to the data source according to the imposed and estimated fundamental frequency.