Computer-implemented method of performing a comparison between an actual response to image stimuli and a response template
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing biometric authentication methods using reflex responses face challenges such as high acquisition costs, data management complexity, precision in comparison, and environmental noise interference, limiting their widespread commercial use.
Innovation Solution
A computer-implemented method using a sequence of images as reflex response stimulation, involving personal and generic images, normalizes the response data to factor out environmental noise, enabling precise comparison and re-initialization by changing the image sequence, and computes a normalization parameter to enhance discrimination between individuals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If environmental noise factors are included in the comparison basis, then the system is more robust to external variations, but the precision of detecting actual reflex response changes deteriorates
Solution Approach 1:
The patent segments the comparison basis into multiple independent factors (pupil area, pupil perimeter, iris area, iris perimeter, image intensity) and selectively weights or excludes them based on their susceptibility to environmental noise. This allows the system to maintain robustness by including stable factors while improving precision by excluding or down-weighting noise-sensitive factors.
Solution Approach 2:
Different regions of the eye (pupil vs. iris) and different measurement dimensions (area vs. perimeter vs. intensity) are treated with different quality weights. The patent applies local quality assessment to determine which specific measurement aspects are most reliable for detecting reflex responses versus which are more susceptible to environmental interference.
2Reliability
If the comparison basis is broadened to account for environmental variations, then false positives are reduced, but the ability to detect subtle reflex changes deteriorates
Solution Approach 1:
The patent dynamically changes the parameters of comparison by adjusting the weights assigned to different measurement factors based on environmental conditions and individual user characteristics. This allows the system to broaden the comparison basis when environmental variability is high (reducing false positives) while maintaining sensitivity to subtle changes when conditions are stable.
Solution Approach 2:
The comparison basis is made dynamic rather than static. The system adapts the set of factors used for comparison and their relative importance based on real-time assessment of environmental conditions, user state, and historical data, enabling it to balance robustness and precision according to current operating conditions.
3Measurement precision
If more measurement factors are included in the comparison, then discrimination between individuals is enhanced, but data management complexity increases
Solution Approach 1:
The patent segments the biometric data into distinct measurable factors (pupil area, perimeter, iris area, perimeter, intensity) that can be independently processed and analyzed. This segmentation enables more efficient data management compared to treating the reflex response as a single complex signal, as each factor can be handled separately with dedicated processing algorithms.
Solution Approach 2:
The system selectively changes which parameters are active in the comparison based on the specific application context, environmental conditions, and individual user profiles. This parameter adaptation reduces the effective data management burden while maintaining high discrimination capability by focusing on the most relevant factors for each situation.
Data Source
AI summary
The method can include acquiring response data including a digital representation of an actual physiological reflex response of a user to exposure to a stimuli sequence associated to the user ID, the stimuli sequence including personal images and generic images; providing response template data associated to the user ID, the response template data including a digital representation of a baseline physiological reflex response of a user corresponding to the user ID to exposure to the stimuli sequence; computing a normalization parameter based on a comparison between the response data associated to the generic images and the response template data associated to the generic images, and comparing the response data associated to the personal images to the response template data associated to the personal images, including factoring out a noise component based on the normalization parameter.