Iris Authentication Boundary Detection for Patterned Contact Lenses
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing iris authentication systems fail to accurately authenticate individuals wearing patterned contact lenses due to the interference of the lens patterns with iris patterns, leading to erroneous recognition and reduced authentication accuracy.
Innovation Solution
An information processing apparatus and method that determines the inner boundary of the patterned contact lens area in an eye image and performs iris authentication based on this boundary, masking out the lens patterns to isolate the iris patterns for accurate verification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If iris authentication is performed on individuals wearing patterned contact lenses, then authentication coverage is improved, but authentication accuracy deteriorates due to pattern interference
Solution Approach 1:
The patent segments the eye image into multiple regions: contact lens region, iris region, and pupil region. By detecting the contact lens boundary and separating it from the iris area, the system processes only the clean iris portion without pattern interference, thus maintaining authentication accuracy while enabling authentication for contact lens wearers.
Solution Approach 2:
The patent extracts and removes the contact lens pattern area from the authentication process. By detecting the contact lens boundary and excluding this region from iris analysis, the system eliminates the harmful pattern interference while preserving the valid iris information for accurate authentication.
2Productivity
If the entire eye image is used for authentication, then processing speed is improved, but authentication accuracy deteriorates due to inclusion of contact lens patterns
Solution Approach 1:
The patent quickly segments the eye image by detecting the contact lens boundary using color or pattern recognition algorithms. This segmentation identifies the contact lens region and separates it from the iris region, allowing the system to process only the necessary iris portion without unnecessary computations on contact lens areas.
Solution Approach 2:
The patent applies partial action by processing only the relevant iris region rather than the entire eye image. By detecting and excluding the contact lens boundary area, the system performs authentication processing on a reduced, cleaner dataset that maintains accuracy while reducing computational overhead.
3Device complexity
If contact lens patterns are included in iris analysis, then feature extraction complexity is reduced, but authentication reliability deteriorates due to erroneous pattern recognition
Solution Approach 1:
The patent segments the image to identify and separate the contact lens region from the iris region using boundary detection algorithms. This segmentation ensures that only the pure iris features are extracted for authentication, eliminating the risk of erroneously interpreting contact lens patterns as iris features while maintaining a relatively simple processing pipeline.
Solution Approach 2:
The patent introduces an intermediary step of contact lens boundary detection that acts as a mediator between image acquisition and iris feature extraction. This intermediary process identifies and masks the contact lens area, ensuring that subsequent feature extraction operates only on valid iris data, thus preventing erroneous pattern recognition.
Data Source
AI summary
An information processing apparatus 1 comprising: a determination portion 11 that determines, in a case where an iris authentication target person is wearing a patterned contact lens including a predetermined pattern area, from an eye image including an eye of the iris authentication target person, an inner boundary of the pattern area included in the eye image; and an iris authentication portion 12 that performs iris authentication for the iris authentication target person based on the inner boundary.