Biometric Fraud Detection via Optical Geometric Parameter Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing biometric authentication methods are vulnerable to 're-projection' fraud, where a pre-recorded video of an individual is displayed on a screen to bypass authentication, and current detection methods are limited in effectiveness, especially when the reacquisition is fronto-parallel, leading to false negatives.
Innovation Solution
A method that estimates and compares the geometric parameters of the optical device used during candidate image acquisition with those from the enrollment phase, detecting fraud by identifying differences in focal length, sensor pixel size, and optic center position, even in the absence of image distortion, to determine if the candidate image sequence was acquired from a pre-recorded image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional biometric authentication methods are used, then the authentication process is simple and fast, but the system is vulnerable to re-projection fraud attacks
Solution Approach 1:
The patent applies preliminary action by capturing geometric parameters of the optical device during the enrollment phase, before the actual authentication occurs. These parameters (focal length, sensor pixel size, optic center position) are stored and later used to verify the authenticity of the verification images. This pre-capture of reference data enables fraud detection without adding complexity to the core authentication flow.
Solution Approach 2:
The patent replaces traditional mechanical/fraudulent projection systems with an optical parameter analysis system. Instead of relying on physical characteristics alone, the system substitutes geometric parameter verification (optical properties) to detect whether images were captured directly or projected from pre-recorded sources. This substitution enables detection of re-projection fraud while maintaining system simplicity.
2Reliability
If distortion-based detection methods are used, then some fraud can be detected, but the method fails when the reacquisition is fronto-parallel without distortion
Solution Approach 1:
The patent changes the detection parameter from image distortion (visual characteristic) to geometric optical parameters (focal length, sensor pixel size, optic center position). These parameters remain consistent regardless of whether the reacquisition is fronto-parallel or angled, enabling the system to detect fraud in all orientations. The geometric parameters provide a distortion-independent verification mechanism that works universally.
3Reliability
If geometric parameter estimation is performed, then fraud detection accuracy improves, but the processing time and computational load increase
Solution Approach 1:
The patent applies preliminary action by computing and storing the geometric parameters during the enrollment phase, before authentication is needed. This pre-computation eliminates the need for complex real-time calculations during verification, reducing processing time while maintaining high detection accuracy. The system only needs to compare stored parameters with new measurements, not perform full geometric estimations from scratch.
Data Source
Figure 1
Figure 2~3
Figure 4a~4b
AI summary
The invention proposes a method for detecting fraud (220) during the biometric authentication of an individual, comprising the acquisition (210), by an optical device, of a sequence of candidate images comprising at least two candidate images of an individual, in order to compare said sequence with a reference image sequence comprising at least one reference image of an individual acquired during a prior enrollment step (100) by the same optical device, the method comprising the steps of: - estimating (221), from the sequence of candidate images, the geometric parameters of the optical device that acquired said sequence of candidate images on the individual, and - comparing (222) said geometric parameters with the geometric parameters of the optical device that acquired the reference image sequence, and - from the result of the comparison, determining (223,224) if the verification image sequence was acquired from at least one pre-recorded image of the individual.