Biometric Eye Region Template Merging for Recognition Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric person recognition methods based on the eye region face significant challenges in moderate and poor lighting conditions, with recognition rates being considerably lower compared to facial recognition, necessitating an improvement in accuracy and robustness.
Innovation Solution
A method that combines a first image of one eye region with a mirrored version of the second eye region, creating an overlaid image in position or feature space to generate a template for person recognition, which enhances recognition rates by creating a more robust template and addresses issues related to mirroring in image capture.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If person recognition is performed based on the eye region only, then recognition speed is improved and computational complexity is reduced, but recognition accuracy deteriorates significantly in moderate and poor lighting conditions
Solution Approach 1:
The patent combines the left and right eye regions into a single merged eye region template. This merging process integrates information from both eyes, creating a more comprehensive biometric representation that maintains the speed benefits of eye-region-only recognition while improving accuracy by capturing more distinctive facial characteristics through the combined eye regions.
2Measurement precision
If facial recognition is performed using the entire face, then recognition accuracy is improved, but computational complexity and processing time increase
Solution Approach 1:
The patent extracts and focuses on the eye region as the critical biometric feature for recognition. By isolating the eye region from the entire face, the system achieves high recognition accuracy through the distinctive characteristics of the eyes while significantly reducing the computational complexity compared to processing the complete facial image.
3Device complexity
If image mirroring is not accounted for in eye region recognition, then processing simplicity is maintained, but recognition reliability deteriorates due to mismatched eye orientations
Solution Approach 1:
The patent addresses image mirroring by creating asymmetric handling of the eye regions. The left and right eye regions are processed and combined in a way that accounts for their asymmetric positions and orientations. This asymmetric approach allows the system to reliably recognize individuals regardless of whether the captured image is mirrored or not, significantly improving recognition reliability without requiring complex mirroring correction algorithms.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The method according to the invention is based on a first image of a first eye region of a person and a second image of a second eye region of the person, wherein the first eye region contains one of the eyes of the person, for example the right eye, and the second eye region contains the other eye of the person, for example the left eye; one of the images is mirrored, and the mirrored and the non-mirrored image are combined in the position space and/or in the feature space, in order to generate a template of an overlaid image. The template contains biometric features for person recognition.