Multimodal Biometric Scanner Using Infrared Surface and Subcutaneous Feature Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional biometric systems are limited by the use of single modalities, such as images of veins or surface features, which can be inaccurate and prone to artifacts, and complex, costly systems are required to acquire and process multimodal images effectively.
Innovation Solution
A system that acquires and processes multimodal images including both surface and subcutaneous features using a single scanner with infrared light, employing a machine learning-based multimodal embedding module to generate representative embedding data for identity assertion, which includes intersection and XOR features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional single modality biometric systems are used, then the system complexity is low, but the identification accuracy is poor and prone to artifacts
Solution Approach 1:
The patent combines multiple modalities (surface features and subcutaneous vein patterns) into a single biometric recognition system. The scanner integrates multiple sensing capabilities to capture both surface geometry and subsurface vascular structures simultaneously, merging what were traditionally separate biometric modalities into one unified system that improves accuracy without proportionally increasing complexity
Solution Approach 2:
The scanner is designed with multi-functionality to acquire different types of biometric data (surface features and subcutaneous features) using the same device. This universal scanner can perform multiple biometric functions, capturing various modalities through different imaging modes or processing techniques, thereby improving identification accuracy while avoiding the need for multiple separate specialized devices
2Measurement precision
If complex multimodal systems are used to acquire and process images, then identification accuracy improves, but hardware cost and complexity increase
Solution Approach 1:
The system merges multiple biometric modalities into a single scanning operation. By capturing surface and subcutaneous features simultaneously or in sequence using one scanner, the system avoids the need for multiple separate imaging devices, reducing hardware costs while maintaining the accuracy benefits of multimodal recognition
Solution Approach 2:
The system creates multiple representations (embeddings) of biometric features from a single image source. Through machine learning processing, the scanner generates both surface feature embeddings and subcutaneous feature embeddings from the same captured image, effectively creating multiple biometric views without requiring multiple physical imaging systems
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 improves biometric identification accuracy by decorrelating features from multiple modalities, reducing the need for complex and costly hardware while enhancing the ability to distinguish between individuals and detect artifacts.
Implementation Method 1
A scanner captures optical images of a user's hand
Data Source
AI summary
A biometric identification system acquires a multimodal image. The system is trained to determine three embeddings: a first embedding associated with features present in a first modality, an intersection embedding associated with features present in both the first and a second modality, and an XOR embedding associated with features that are not shared between the first modality and the second modality. Once trained, the system may process a multimodal query image to determine query embedding data. Additional information, such as minutiae depicted in the multimodal query image may also be determined. The query embedding data, and in some implementations the additional information, may be compared with enrolled embedding data associated with a previously enrolled user. If the comparison exceeds a threshold value, an identity may be asserted.