Off-Axis Iris Recognition via Elliptical Grid Unwrapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional iris biometric systems struggle with accurate iris recognition when the user's gaze is off-axis relative to the camera's optical axis, leading to distortion and errors in recognition/authentication due to cornea refraction and elliptical iris shapes, particularly in VR, AR, and MR applications.
Innovation Solution
An image acquisition system that models and corrects distortions in iris texture by transforming iris and pupil ellipses into circular shapes, interpolating pixel values along an iris grid to account for cornea refraction, and selects suitable enrolment templates based on pupil geometric parameters to improve recognition accuracy in off-axis conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the system uses conventional circular iris segmentation, then the processing is simple, but the segmentation accuracy deteriorates for off-axis gaze images with high eccentricity
Solution Approach 1:
The system changes the segmentation parameter from a fixed circular shape to an adaptive elliptical shape. By calculating the eccentricity of the iris from the image and adjusting the segmentation ellipse parameters accordingly, the system maintains both simplicity and accuracy across different gaze angles without requiring complex multi-step processing.
Solution Approach 2:
The segmentation approach transitions from a static circular model to a dynamic elliptical model that adapts to the specific characteristics of each off-axis iris image. The system dynamically adjusts the ellipse parameters based on the calculated eccentricity, enabling accurate segmentation for varying gaze conditions while keeping the overall process straightforward.
2Area of stationary object
If the system sets the circle radius to the major axis of the elliptic iris shape, then the coverage area is maximized, but false inclusion of sclera and periocular region increases
Solution Approach 1:
Instead of using a fixed circular radius, the system dynamically adjusts the segmentation ellipse parameters to match the actual iris shape in each image. By calculating the correct ellipse parameters from the image data, the system achieves precise boundary definition that prevents both under-segmentation and over-inclusion of non-iris regions.
Solution Approach 2:
The system replaces the mechanical approach of using fixed circular radii with a computational geometry approach. By calculating the eccentricity and fitting an ellipse to the iris boundary points, the system achieves more precise segmentation that adapts to the specific geometric characteristics of each off-axis iris image.
3Productivity
If the system uses a fixed circular segmentation radius, then the processing speed is maintained, but the false acceptance rate increases due to mismatched iris templates
Solution Approach 1:
The system efficiently calculates the eccentricity parameter and adjusts the segmentation ellipse accordingly, maintaining processing speed while improving accuracy. The computational steps required to determine the ellipse parameters are minimal and can be performed quickly, allowing the system to adapt to different gaze angles without significant performance penalty.
Solution Approach 2:
The system performs preliminary calculation of the eccentricity and ellipse parameters during the segmentation stage, preparing accurate iris templates in advance. This preliminary action ensures that when recognition occurs, the templates are already optimized for the specific gaze condition, reducing the need for additional processing and maintaining high recognition accuracy.
4Adaptability or versatility
If the system operates with off-axis gaze conditions, then the adaptability to VR/AR/MR applications improves, but the cornea refraction distortion increases
Solution Approach 1:
Instead of treating cornea refraction distortion as a problem to be eliminated, the system converts it into a useful indicator. By calculating the eccentricity from the distorted iris shape, the system actually benefits from the distortion information to determine the correct segmentation ellipse parameters and create gaze-appropriate templates, turning the harmful distortion into a useful measurement.
Solution Approach 2:
The system changes the approach from trying to correct the distortion to utilizing it for parameter calculation. By measuring the eccentricity caused by cornea refraction and using this value to adjust the segmentation ellipse, the system adapts to off-axis conditions naturally, expanding applicability to VR/AR/MR while managing the distortion effect constructively.
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
The system effectively compensates for cornea refraction distortions and reduces false acceptance/rejection rates by accurately interpolating iris texture and selecting appropriate enrolment templates, enabling reliable iris recognition in off-axis gaze conditions with limited computational effort.
Implementation Method 1
an image of a portion of a face of the user looking towards a target displayed on a display in a sequence of different locations is captured by a lens assembly and an image sensor
Implementation Method 2
distortions in iris texture due to cornea refraction
Data Source
AI summary
An image acquisition system determines first and second sets of points defining an iris-pupil boundary and an iris-sclera boundary in an acquired image; determines respective ellipses fitting the first and second sets of points; determines a transformation to transform one of the ellipses into a circle on a corresponding plane; using the determined transformation, transforms the selected ellipse into a circle on the plane; using the determined transformation, transforms the other ellipse into a transformed ellipse on the plane; determines a plurality of ellipses on the plane for defining an iris grid, by interpolating a plurality of ellipses between the circle and the transformed ellipse; moves the determined grid ellipses onto the iris in the image using an inverse transformation of the determined transformation; and extracts an iris texture by unwrapping the iris and interpolating image pixel values at each grid point defined along each of the grid ellipses.