Biometric Authentication Using Vascular Point Detection and Template Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current biometric authentication systems using eye images face challenges in accurately and efficiently comparing images due to variations in image quality and environmental conditions, leading to potential misidentification or false rejection of authorized users.
Innovation Solution
The implementation of an image sharpening technique combined with Vascular Point Detection (VPD) and Pattern Histograms of Extended Multi-Radii Local Binary Patterns (PH-EMR-LBP) and Center Symmetric Local Binary Patterns (PH-EMR-CS-LBP) for feature extraction and matching, which enhances the detection of vascular points and surrounding image descriptions, and uses outlier detection to improve matching efficiency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional image comparison methods are used for biometric authentication, then the system is simpler to implement, but the accuracy and reliability of authentication decreases due to image quality variations and environmental conditions
Solution Approach 1:
The patent segments the eye image into distinct anatomical regions (iris, pupil, sclera, eyelids, eyelashes) and processes each region separately with region-specific algorithms. This segmentation allows the system to focus computational resources on relevant features while ignoring irrelevant areas, thereby improving authentication accuracy without proportionally increasing overall system complexity
Solution Approach 2:
The patent performs preliminary image processing steps including sharpening, contrast enhancement, and region segmentation before feature extraction and comparison. By pre-processing images to enhance quality and extract relevant regions beforehand, the system compensates for environmental variations and image quality issues, improving authentication accuracy while maintaining efficient processing
2Measurement precision
If image sharpening and advanced feature extraction techniques are applied, then the detection accuracy of vascular points improves, but the processing time and computational complexity increase
Solution Approach 1:
The patent applies different processing techniques and levels of sharpening to different regions of the eye image based on their specific characteristics. For example, the iris region receives different treatment than the sclera region. This local quality approach enhances vascular point detection accuracy in each region while avoiding unnecessary processing elsewhere, thereby reducing overall processing time
Solution Approach 2:
The patent applies image sharpening and enhancement techniques selectively to only the regions where vascular points are expected to be found, rather than processing the entire image at maximum quality. This partial action approach maintains detection accuracy in critical areas while reducing computational load and processing time in less critical regions
3Adaptability or versatility
If multiple reference images are stored for template updating, then the system becomes more adaptable to behavioral and environmental variations, but the storage requirements and template management complexity increase
Solution Approach 1:
The patent merges multiple reference images into a consolidated template by extracting features from each image and combining them into a unified representation. This merging process allows the system to capture behavioral and environmental variations across multiple images while storing a single compact template, thereby increasing adaptability without proportionally increasing storage requirements
Solution Approach 2:
The patent implements a template update mechanism where outdated or low-quality reference images are discarded and replaced with newer, higher-quality images that better represent current behavioral and environmental conditions. This selective discarding and recovery of templates allows the system to adapt to variations over time while maintaining efficient storage by removing redundant or obsolete data
Data Source
Figure 1A
Figure 1B
Figure 2A~2F
AI summary
In a feature extraction and pattern matching system, image sharpening can enable vascular point detection (VPD) for detecting points of interest from visible vasculature of the eye. Pattern Histograms of Extended Multi-Radii Local Binary Patterns and/or Pattern Histograms of Extended Multi-Radii Center Symmetric Local Binary Patterns can provide description of portions of images surrounding a point of interest, and enrollment and verification templates can be generated using points detected via VPD and the corresponding descriptors. Inlier point pairs can be selected from the enrollment and verification templates, and a first match score indicating similarity of the two templates can be computed based on the number of inlier point pairs and one or more parameters of a transform selected by the inlier detection. A second match score can be computed by applying the selected transform, and either or both scores can be used to authenticate the user. An enrollment template can be a collection of interest points such as vascular points (VPD) and corresponding features such as Enhanced Multi-Radii Local Binary Patterns (EMR- LBP), Pattern Histograms of Enhanced Multi-Radii Local Binary Patterns (PH-EMR-LBP), Pattern histograms of Enhanced Multi-Radii Center-Symmetric Local Binary Patterns (PH- EMR-CS-LBP), and Enhanced Multi-Radii Center-Symmetric Local Binary Patterns (EMR- CS-LBP). In some implementations, an enrollment template can be created only if the acquired image exceeds a certain threshold based on ratio of VPD points to that of size of segmented scleral region. More than one enrollments are possible for a single user. Enrollment templates can be updated to accommodate behavioral and/or environmental variations affecting the acquired scans. Updating the enrollment templates using verification can be based on quality of a candidate verification template, match score, and/or other image and exposure similarity measures.