Biometric Authentication via Multi-Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric authentication systems face challenges in varying light conditions and 'spoofing' techniques, making it difficult to reliably recognize biometric traits like faces and irises, especially in environments with changing lighting and potential fraudulent attempts.
Innovation Solution
A method that acquires visible, infrared, and depth images of a biometric trait, selects the best image based on ambient lighting conditions, fuses the detected traits, and verifies authenticity using the depth image, ensuring robust detection and authentication across different lighting scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single 2D or 3D camera is used for biometric authentication, then the device complexity is low and ergonomics are good, but the reliability of biometric recognition deteriorates under varying lighting conditions
Solution Approach 1:
The patent combines multiple camera types (2D visible light camera, 3D structured light camera, and infrared camera) into a single biometric authentication system. Each camera captures different aspects of the biometric trait, and their data is fused to achieve reliable recognition across varying lighting conditions while maintaining acceptable device complexity through integrated processing.
Solution Approach 2:
The system employs multi-functional optical sensors that can operate across different lighting conditions. The visible light camera provides detailed texture information, the 3D camera provides depth and spatial structure, and the infrared camera provides illumination-independent detection. This multi-functionality ensures reliable biometric authentication regardless of ambient lighting.
2Measurement precision
If exposure is adjusted to normalize brightness of a detected region of interest, then the biometric feature detection precision improves, but other regions of the image become over- or under-exposed
Solution Approach 1:
The patent segments the biometric authentication task across multiple independent camera systems, each optimized for specific conditions. The visible light camera handles well-lit environments with high detail, the 3D camera handles various lighting with depth information, and the infrared camera handles low-light conditions. This segmentation allows each region to be processed optimally without compromising others.
Solution Approach 2:
The system changes the operational parameters by switching between different camera types based on ambient lighting conditions. The processing means adapts the acquisition parameters dynamically, selecting which camera(s) to activate and how to process the data based on the current lighting environment, thereby maintaining optimal detection precision without information loss.
3Adaptability or versatility
If multiple camera types are used to cover various lighting conditions, then the adaptability to lighting environments improves, but the device complexity and difficulty of processing increase
Solution Approach 1:
The patent implements dynamic adaptation by having the processing means automatically select and switch between different camera types based on real-time lighting condition detection. The system dynamically adjusts which sensors are active and how the data is processed, optimizing performance for each lighting environment while managing complexity through automated decision-making.
4Reliability
If 2D or 3D images are used for authentication, then the ease of operation is good with wide field of view, but the reliability against spoofing techniques deteriorates
Solution Approach 1:
The patent adds another dimension to biometric authentication by incorporating depth information from 3D structured light imaging alongside traditional 2D visible light images. This dimensional addition enables the system to detect three-dimensional characteristics of the subject's face, making it difficult for two-dimensional spoofs (masks, photos) to deceive the system while maintaining ease of operation through automated multi-dimensional analysis.
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 enhances the reliability and effectiveness of biometric authentication by leveraging the strengths of each image type under various lighting conditions, reducing false positives and spoofing attempts, thereby improving overall performance and security.
Implementation Method 1
an infrared image from data acquired by second optical acquisition means of the terminal
Implementation Method 2
a depth image from data acquired by third optical acquisition means of the terminal
Data Source
Figure 1
Figure 2
Figure 3a~3d
AI summary
Method for authenticating or identifying an individual. The present invention relates to a method for authenticating or identifying an individual, characterized in that it comprises the implementation by data processing means (11) of a terminal (1) of the following steps: (a) Obtaining a visible image, an infrared image and a depth image on each of which appears a biometric trait of said individual; (b) Selecting at least one of said visible image, infrared image and depth image according to the ambient lighting conditions; (c) Detecting said biometric trait of the individual in each selected image; (d) Merging the detected biometric trait(s); (e) Authenticating or identifying said individual on the basis of the result of the merging of the detected biometric trait(s).