Angular Filtering for 3D Finger Vein Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing finger vein authentication systems using 2D projection images are prone to errors due to the complex 3D structure of vein patterns, which changes significantly with minor changes in finger position, leading to erroneous recognition.
Innovation Solution
A light-detecting apparatus estimates the global and direct components of light from a scattering scene to reconstruct a 3D pattern of finger veins, using angular samples to separate and analyze light intensity, allowing for the calculation of a 3D shape of the vein pattern.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If 2D projection images are used for vein pattern recognition, then the authentication system is simple to implement, but the recognition accuracy deteriorates due to sensitivity to finger position changes
Solution Approach 1:
The patent transitions from 2D projection imaging to 3D vein pattern reconstruction by capturing light intensity at multiple angles and using angular filtering to separate direct and scattered light components. This dimensional transition enables accurate vein pattern recognition that is invariant to finger position changes, resolving the contradiction between implementation simplicity and recognition accuracy.
2Reliability
If 3D vein pattern reconstruction is performed using angular filtering and light component separation, then recognition accuracy is improved, but device complexity increases
Solution Approach 1:
The patent segments the light field into angular components using an angular filter array, separating light rays based on their incident angles. This segmentation allows the system to isolate and process direct and scattered light components independently, achieving accurate 3D reconstruction while managing computational complexity through structured angular sampling.
Solution Approach 2:
The patent introduces an angular filter array as an intermediary optical element between the imaging lens and sensor. This intermediary device performs angular filtering to separate direct and scattered light components, enabling 3D vein pattern reconstruction without requiring complex computational algorithms or multiple imaging systems.
3Measurement precision
If angular samples are taken to separate direct and scattered light components, then high-frequency details and edge cues are restored, but measurement and processing difficulty increases
Solution Approach 1:
The patent employs periodic angular sampling by arranging filters at regular angular intervals around the optical axis. This periodic angular discretization transforms the continuous angular filtering problem into a manageable set of discrete measurements, enabling efficient separation of direct and scattered light components while preserving high-frequency spatial details.
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 provides more accurate vein pattern recognition by restoring high-frequency details and providing strong edge cues, reducing errors associated with minor finger position changes and enabling 3D reconstruction using visible-wavelength light sources.
Implementation Method 1
Some of that light is scattered, and some is absorbed by hemoglobin in veins. An image sensor, such as a CCD (charge-coupled device) camera, captures an image.
Implementation Method 2
Some of that light is scattered, and some is absorbed by hemoglobin in veins.
Data Source
AI summary
In an exemplary implementation of this invention, light from a scattering scene passes through a spatial light attenuation pattern and strikes a sensor plane of a camera. Based on said camera's measurements of the received light, a processing unit calculates angular samples of the received light. Light that strikes the sensor plane at certain angles comprises both scattered and directly transmitted components; whereas light that strikes at other angles comprises solely scattered light. A processing unit calculates a polynomial model for the intensity of scattered-only light that falls at the latter angles, and further estimates the direct-only component of the light that falls at the former angles. Further, a processing unit may use the estimated direct component to calculate a reconstructed 3D shape, such as a 3D shape of a finger vein pattern, using an algebraic reconstruction technique.