Palm Vein Image Correction via 3D Shape Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Palm vein authentication systems face challenges in accurately authenticating individuals due to posture differences between registration and matching, leading to decreased success rates, as existing methods fail to effectively correct for distortions caused by hand inclination and bending.
Innovation Solution
An image processing device with a shape estimation unit, storage unit, and correction unit that estimates the three-dimensional shape of the palm, generates a quadratic surface, and corrects the image by developing points to a planar surface, thereby normalizing the image and reducing distortion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If palm vein authentication is performed without posture correction, then the authentication process is simple and fast, but the success rate decreases due to posture differences between registration and matching
Solution Approach 1:
The system performs preliminary posture correction during the registration phase by estimating the three-dimensional shape of the palm and generating a development pattern. This preparation work is stored and later applied during authentication to compensate for posture differences, ensuring high success rates without adding complexity to the real-time authentication process
Solution Approach 2:
The invention creates a two-dimensional development pattern that copies and flattens the three-dimensional palm surface geometry. This development pattern serves as a reference model that can be applied to both registration and authentication images, enabling posture normalization through pattern matching and transformation without requiring complex real-time 3D reconstruction during authentication
2Measurement precision
If posture correction using 3D shape estimation is applied, then image normalization accuracy improves, but the processing time and computational load increase
Solution Approach 1:
The three-dimensional shape estimation and development pattern generation are performed in advance during registration. The computed geometric transformations and correction parameters are stored for reuse during authentication, eliminating the need for repeated complex calculations and reducing real-time processing time
Solution Approach 2:
The invention models the palm surface as a developable quadratic surface with specific curvature properties. By representing the complex 3D geometry as a mathematically tractable quadratic surface, the system achieves accurate posture estimation while enabling efficient computation through closed-form solutions for surface development and coordinate transformation
3Adaptability or versatility
If the palm is allowed to bend or incline during authentication, then the user experience is more flexible and convenient, but the image distortion increases and authentication accuracy decreases
Solution Approach 1:
The system creates a development pattern that copies the actual three-dimensional shape of the user's palm including any bending or inclination. This development pattern is then used to transform both registration and authentication images into a common reference coordinate system, effectively removing distortion caused by flexible hand postures while maintaining user convenience
Solution Approach 2:
The invention dynamically adjusts the development pattern parameters based on the detected palm shape. By changing the geometric parameters of the quadratic surface model to match the actual palm configuration, the system adapts to various hand postures and eliminates distortion through parameter-based image transformation rather than requiring fixed rigid posture constraints
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A shape estimation unit (311) estimates a shape (321) of a subject on the basis of an image of the subject, and a storage unit (312) stores the shape of the subject. The correction unit (313) generates a quadratic surface connecting a plurality of points on the subject on the basis of the shape of the subject, and develops each of the plurality of points to a planar surface contacting the quadratic surface. Thereafter the correction unit corrects the image of the subject on the basis of a position of each of the plurality of points developed to the planar surface.