Iris Spoofing Detection via Eye Movement Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current biometric systems, particularly iris recognition, are vulnerable to spoofing attacks using printed images of human irises, as existing mechanisms to detect such attacks can be bypassed by reflecting light from the cornea, compromising security.
Innovation Solution
The use of eye movement cues to detect spoofing attacks by measuring and analyzing distortions in eye movement signals, which differ between natural and printed iris images, employing oculomotor plant characteristics and complex eye movement patterns to assess whether an iris print-attack is occurring.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If pupil reflection detection mechanism is used to detect spoofing attacks, then security against printed iris images is improved, but the mechanism can be bypassed by making a hole that allows light reflection from the cornea
Solution Approach 1:
The patent introduces eye movement tracking as an intermediary verification layer between the iris recognition system and the authentication decision. Instead of relying solely on pupil reflection detection, the system monitors corneal reflex movements and compares them against expected physiological eye movement patterns. This intermediary mechanism detects spoofing attempts by identifying the absence of natural eye movement characteristics, even when pupil reflection detection is bypassed through corneal hole attacks.
Solution Approach 2:
The patent transitions from static pupil reflection detection to dynamic eye movement analysis. By tracking the temporal patterns of corneal reflex movements during natural eye scanning, the system captures dynamic behavioral characteristics that cannot be replicated by static printed images or simple optical illusions. The system evaluates whether eye movements follow natural physiological patterns, providing adaptability to detect various spoofing techniques while maintaining security.
2Measurement precision
If iris recognition systems require users to stand very still and very close to the image capturing device, then identification accuracy is improved, but user convenience deteriorates
Solution Approach 1:
The patent employs dynamic eye movement tracking to maintain identification accuracy while allowing natural user movement. Instead of requiring users to remain perfectly still, the system continuously monitors corneal reflex movements and adapts to natural eye scanning behavior. This dynamic approach captures physiological eye movement patterns that serve as additional authentication factors, maintaining precision while significantly improving ease of operation by eliminating rigid positioning constraints.
Solution Approach 2:
The system leverages the user's own natural eye movements as the authentication mechanism rather than requiring controlled, still positioning. Users naturally scan their environment and move their eyes during normal interaction, and the system captures these self-generated movements for verification. This self-service approach eliminates the need for users to consciously control their head or eye positioning, making the authentication process as convenient as natural viewing while maintaining high identification accuracy through physiological pattern recognition.
Data Source
AI summary
A method of assessing the possibility of an iris print-attack includes measurement of the eye movement of a person. One or more values (e.g., a feature vector) are determined based on the measured eye movements. Based on the determined values, an assessment is made of whether or not the person is engaging in an iris print-attack.