Iris Spoof Detection via Dual Exposure Reflection Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Iris-based biometric systems are susceptible to spoofing attacks, with existing methods causing user discomfort and safety concerns due to varying illumination levels, which can exceed safety thresholds.
Innovation Solution
The method involves obtaining two iris images with different exposure settings, comparing reflection patterns, and determining authenticity based on consistent or inconsistent variations in image characteristics such as size, intensity, and pixel saturation, without altering illumination levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If illumination levels are varied to detect spoofing, then spoof detection capability is improved, but user comfort and eye safety deteriorate due to intensity exceeding safety thresholds
Solution Approach 1:
The patent changes the parameter being varied from illumination intensity to exposure time. By taking images at different exposure times (e.g., first exposure time t1, second exposure time t2) while keeping illumination constant, the system achieves spoof detection without exposing the eye to harmful light levels. The reflection patterns differ authentically between exposure times for real eyes but not for spoofs.
Solution Approach 2:
The patent replaces the mechanical/optical approach of varying light source intensity with a temporal approach of varying exposure duration. Instead of modulating the illumination source strength, the system uses the camera's exposure time setting to capture different amounts of reflected light, achieving the same detection goal without the harmful side effects of high-intensity illumination.
2Loss of information
If high resolution printed images are presented for imaging, then spoofing attack success is improved, but system security deteriorates
Solution Approach 1:
The patent uses periodic variation in exposure time to detect spoofing. By alternating between different exposure settings (first exposure time, then second exposure time) and analyzing the periodic response of reflection patterns, the system can distinguish authentic eyes from high-resolution printed spoofs. Real eyes show consistent periodic response patterns while spoofs do not.
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 approach effectively differentiates between authentic and counterfeit irises while ensuring user safety and comfort by using exposure control instead of varying illumination, thereby enhancing spoof detection in iris recognition systems.
Implementation Method 1
comparing the incidence of reflection patterns within the first and second iris images
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
The invention provides methods, systems and computer program products for spoof detection in iris based biometric recognition. In an embodiment, the invention comprises (i) obtaining a first image of an iris at an imaging apparatus configured to a first exposure setting (202) (ii) obtaining a second image of the iris at the imaging apparatus configured to a second exposure setting (204) (iii) comparing incidence of reflection patterns within the first and second iris images (206) and (iv) determining authenticity of the imaged iris based on a result of comparing the incidence of reflection patterns within the first and second iris images (208).