Adaptive Eye Illumination for Pose-Robust Biometric Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional eye tracking and biometric authentication systems face challenges in capturing high-quality images due to varying lighting conditions, device positioning, and optical imperfections, leading to inefficiencies and inaccuracies in gaze tracking and authentication processes.
Innovation Solution
Implementing flexible illumination methods that dynamically adjust lighting configurations based on user pose, environmental conditions, and device changes, using pre-generated lighting configurations and lookup tables to optimize image capture for biometric authentication and gaze tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional eye tracking systems focus on specular reflections or glints for gaze tracking, then gaze tracking can be performed, but image quality and contrast on the iris and other regions of interest deteriorate under varying lighting conditions
Solution Approach 1:
The system dynamically adjusts lighting configurations based on detected user pose and environmental conditions. Multiple LED groups can be independently controlled to change illumination patterns in real-time, transforming the static lighting system into a dynamic one that adapts to varying conditions to maintain optimal image quality
Solution Approach 2:
The system changes physical parameters of illumination by selecting from pre-generated lighting configurations that specify different LED combinations, intensities, and patterns. This allows transformation of the lighting state to optimize both gaze tracking and iris imaging under different conditions
2Manufacturing precision
If lighting configurations are fixed for specific poses, then image capture can be optimized for known positions, but the system cannot adapt to device positioning changes or unexpected user positions
Solution Approach 1:
The system performs preliminary actions by pre-generating lighting configurations for a comprehensive set of poses before actual operation. These pre-computed configurations are stored and can be quickly retrieved when needed, allowing the system to prepare for various scenarios in advance
Solution Approach 2:
The lighting system achieves multi-functionality by using the same illumination hardware to serve multiple purposes: optimizing for known poses through pre-generated configurations, adapting to unexpected positions through real-time selection, and maintaining performance across diverse operating conditions
3Productivity
If multiple lighting configurations are pre-generated and stored for different poses, then the system can quickly select optimal lighting for known positions, but device complexity and memory requirements increase
Solution Approach 1:
The system creates simplified representations by pre-computing and storing only the essential lighting configuration parameters (which LEDs to activate and their intensities) rather than storing complete lighting scenarios. This copying approach reduces memory requirements while maintaining the ability to quickly retrieve and apply optimal configurations
4Loss of time
If the system uses default initial lighting configuration, then the authentication process can start immediately, but image quality may be insufficient for accurate biometric authentication under suboptimal conditions
Solution Approach 1:
The system implements feedback by capturing initial images with default lighting, evaluating their quality, and using this evaluation to determine whether to switch to alternative lighting configurations. This closed-loop approach ensures that authentication accuracy is maintained while minimizing the time penalty for adaptive lighting changes
Data Source
AI summary
Methods and apparatus for flexible illumination that improve the performance and robustness of an imaging system are described. Multiple different lighting configurations for the imaging system are pre-generated. Each lighting configuration may specify one or more aspects of lighting. A lookup table may be generated via which each pose is associated with a respective lighting configuration. A user may put on, hold, or otherwise use the device. A process may be initiated in which different lighting configurations may be selected by the controller to capture images of the user's eye, periorbital region, or face at different poses and in different conditions for use by a biometric authentication or gaze tracking process.


