Hand Vein Biometric Authentication Posture Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric authentication systems face challenges in maintaining high genuine acceptance rates when there are posture differences between registration and matching, particularly with hand vein patterns, due to local deformations caused by finger postures, which existing correction methods fail to adequately address.
Innovation Solution
A biometric authentication apparatus that captures and corrects hand vein patterns by registering and matching images of hands in different postures, using shape data and posture indices to adjust for positional and rotational differences, thereby minimizing false acceptance rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If biometric information is captured in different postures at registration and matching, then adaptability is improved, but measurement precision deteriorates due to local deformation
Solution Approach 1:
The hand biometric information is divided into multiple regions (fingertip region, middle phalanx region, proximal phalanx region) along the finger length direction. Each region is processed independently with region-specific correction parameters, allowing localized deformation correction while preserving overall adaptability to different postures.
Solution Approach 2:
Different correction parameters are applied to different regions of the hand biometric information. The correction amount varies by region, with the fingertips region receiving different correction than the proximal phalanx region. This local quality approach maintains high matching accuracy despite posture variations.
2Measurement precision
If posture correction is applied to biometric information, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system changes physical parameters (correction amounts) based on measurable indices (finger opening angles) rather than complex geometric transformations. By establishing a relationship between finger opening angle and correction amount, the system achieves accurate posture correction through simple parameter adjustments rather than complex image processing.
3Adaptability or versatility
If multiple registered postures are stored, then adaptability is improved, but loss of information increases due to local deformation variations
Solution Approach 1:
The system performs preliminary correction by calculating correction parameters based on the detected finger opening angle before conducting the matching process. This preliminary action ensures that biometric information from any posture is pre-adjusted to match the registered posture, preventing information loss and maintaining consistency without storing multiple posture variations.
Data Source
AI summary
A biometric authentication apparatus includes a storage unit which stores first shape data representing a shape of biometric information of a registered user's hand with fingers at a first posture and second shape data representing a shape of biometric information of the hand with the fingers at a second posture; a posture specification unit which calculates an index representing a third posture of fingers of a user's hand in a biometric image; a biometric information extraction unit which generates third shape data representing a shape of biometric information of the user's hand in the biometric image; and a correction unit which obtains corrected shape data by correcting the first or the second shape data to cancel a shape difference of the biometric information due to a difference between the third posture and the first or the second posture based on the index for matching.