Camera Image Quality Compensation for Face Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Facial recognition systems in PCs, tablets, and phones are unreliable due to sub-optimal image quality caused by varying ambient lighting conditions, leading to failed face detection and recognition, especially in low light or bright environments.
Innovation Solution
The system overrides camera firmware to adjust exposure, gamma correction, and gain settings using Intel Integrated Performance Primitives image processing libraries, running in a feedback loop to enhance face detection and recognition across all lighting conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If cameras are calibrated for best photographic quality, then image quality for general photography is improved, but face detection reliability in adverse lighting conditions deteriorates
Solution Approach 1:
The patent applies local quality by adjusting camera parameters specifically for face detection regions rather than optimizing the entire image for general photography. The system modifies exposure, gamma, and gain settings locally to enhance face visibility in adverse lighting conditions, thereby improving face detection reliability without completely sacrificing overall photographic quality.
Solution Approach 2:
The patent employs parameter changes by dynamically adjusting camera control parameters (exposure, gamma, gain) based on detected lighting conditions. The system monitors ambient light levels and modifies these parameters in real-time to optimize face detection performance across varying lighting environments, resolving the contradiction between general photographic quality and face detection reliability.
2Use of energy by moving object
If screen brightness is reduced by auto-tuning in low lighting conditions, then energy consumption is reduced, but face illumination effectiveness deteriorates
Solution Approach 1:
The patent introduces an intermediary approach by using camera parameter adjustments (exposure, gamma, gain) as a mediator between screen brightness control and face illumination requirements. Instead of directly increasing screen brightness to illuminate the face, the system uses image processing parameters to compensate for low light conditions, thereby maintaining low energy consumption while achieving effective face illumination for detection.
Solution Approach 2:
The patent replaces the mechanical/optical approach of increasing screen brightness with a digital/image processing approach. By substituting physical illumination (screen brightness) with digital enhancement (camera parameter adjustment), the system achieves face illumination effectiveness without the energy cost of high screen brightness in low lighting conditions.
3Reliability
If camera parameters are adjusted for face detection in adverse lighting, then face detection reliability is improved, but overall image quality may deteriorate
Solution Approach 1:
The patent applies dynamics by making camera parameters adjustable and adaptive rather than fixed. The system dynamically modifies exposure, gamma, and gain settings based on real-time lighting conditions and face detection requirements. This dynamic adjustment allows the system to optimize face detection reliability in adverse lighting while minimizing impact on overall image quality through controlled, conditional parameter changes.
Data Source
AI summary
A user authentication system and method. The user authentication system includes a camera and a processor connected to the camera. The processor receives images from the camera, searches for a user feature in the images, determines if the images require correction, adjusts camera controls in a pre-defined order to provide desired corrections, applies the desired corrections to subsequent images and authenticates the user based on the user feature in the corrected images.