3D Fingerprint Imaging via Multi-Directional Optical Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current biometric fingerprint acquisition methods face challenges such as low recognition rates due to insufficient finger surface scanning, contamination issues, and inaccuracy in touchless imaging, particularly in automated systems, which are inadequate for secure border control and large database searches.
Innovation Solution
A method involving touchless imaging of the finger surface in multiple directions, followed by geometric pre-warping and stereoscopic analysis to create a three-dimensional model, which enhances accuracy and fraud resistance by minimizing errors and allowing complete finger surface imaging without contamination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If flat finger impressions are used with simple sensors, then the acquisition process is simple and fast, but the recognition rate is only 80% to 85% due to insufficient finger surface coverage
Solution Approach 1:
The patent transitions from 2D flat finger impressions to 3D unrolled fingerprint imaging. By capturing the finger surface in three dimensions and unrolling it into a comprehensive 2D representation, the system achieves complete surface coverage while maintaining automated acquisition. This dimensional transformation allows all finger characteristics to be captured without requiring manual rolling, resolving the contradiction between acquisition simplicity and recognition accuracy.
2Reliability
If live-scanners with glass plates are used to roll fingers, then the recognition rate increases to more than 98%, but the system requires manual operation and frequent cleaning due to grease marks and contamination
Solution Approach 1:
The patent replaces the mechanical rolling process on glass plates with a touchless optical imaging system. Multiple cameras capture the finger surface from different angles, and computational algorithms reconstruct the complete fingerprint image. This substitution eliminates physical contact, removing the need for manual operation and cleaning while maintaining high recognition rates through comprehensive surface capture.
Solution Approach 2:
The system performs automated fingerprint acquisition and processing without requiring operator intervention. The touchless imaging system automatically captures, processes, and stores fingerprint data, eliminating the need for supervisors to clean glass plates or assist users. The system serves itself by handling the complete acquisition process autonomously, improving ease of operation while maintaining high reliability.
3Object-affected harmful factors
If touchless imaging is used to image the finger, then contamination is avoided, but the light efficiency is low due to polarization filters and imaging distortions are not compensated
Solution Approach 1:
The patent changes the optical parameters by using multiple imaging directions and wavelengths. Instead of relying on polarization filters that reduce light efficiency, the system captures images from multiple angles and processes them computationally. This parameter change allows touchless imaging to avoid contamination while achieving high measurement precision through multi-directional capture and distortion compensation algorithms.
Solution Approach 2:
The patent introduces computational processing as an intermediary between image capture and fingerprint analysis. The system captures distorted images from multiple directions and uses algorithms to compensate for imaging distortions and reconstruct accurate fingerprint patterns. This computational intermediary resolves the contradiction by maintaining the benefits of touchless imaging while correcting the precision issues through mathematical transformation.
4Loss of time
If the scanned surface of the finger is reduced, then the acquisition time is shortened, but the number of characteristics available for identification becomes insufficient for definite identification
Solution Approach 1:
The patent uses 3D imaging to capture the complete finger surface in a single acquisition event. By imaging in three dimensions and unrolling the surface computationally, the system captures all fingerprint characteristics simultaneously without requiring extended scanning time. This dimensional approach allows complete information capture while maintaining fast automated acquisition, resolving the contradiction between time and information completeness.
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
This approach significantly improves recognition rates and fraud resistance by providing high-accuracy biometric data acquisition, enabling precise identification and reducing false positives, suitable for secure applications like border control.
Implementation Method 1
a surface to be imaged is touchlessly and simultaneously imaged along at least three different directions into digital images
Data Source
AI summary
A method for acquiring biometric data of a finger a surface to be imaged includes touchlessly and simultaneously imaging a finger surface to be imaged along at least three different directions into digital images, wherein each point of the surface to be imaged is imaged along at least two of the directions and wherein the surface to be imaged extends from a first nail edge across a finger-ball to a second nail edge. Subsequently, a geometric basic model for at least a part of the finger is determined using at least one of the digital images. Pre-warped images are calculated by geometrically pre-warping the digital images using the geometric basic model. Additionally, a three-dimensional model of the surface to be imaged is reconstructed using a stereoscopic analysis of the pre-warped images and at least one finger characteristic is determined using the three-dimensional model.