Eye Tracking Exposure Control via Infrared ROI Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Eye tracking systems face challenges in capturing well-exposed images of the eye region, leading to unreliable eye detection due to over-exposure or under-exposure, especially under varying lighting conditions, as existing exposure control methods do not specifically optimize for the eye area.
Innovation Solution
An eye-based exposure control method that adjusts the image sensor's exposure settings based on the intensity values of the eye region and its surroundings, using infrared illuminators to ensure the eye is well-exposed, regardless of other image parts, and continuously adapts to changing light conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If exposure control is applied to the entire image, then overall image quality is improved, but the eye region may still be over-exposed or under-exposed
Solution Approach 1:
The patent divides the image into multiple regions, specifically identifying the eye region as a separate area of interest. The exposure control is then applied independently to the eye region rather than uniformly to the entire image, allowing optimized exposure settings for eye detection while maintaining acceptable quality in other areas.
Solution Approach 2:
The patent implements local exposure control by calculating exposure metrics specifically for the eye region and applying corrections only to that local area. This allows the eye region to have optimal exposure quality for detection purposes, while other parts of the image may have different exposure characteristics that are acceptable for their respective purposes.
2Measurement precision
If exposure settings are optimized for the eye region, then eye detection accuracy is improved, but other parts of the image may become over-exposed or under-exposed
Solution Approach 1:
The patent segments the image processing into distinct regions, with separate exposure control algorithms applied to the eye region versus the rest of the image. This segmentation allows the eye region to receive optimized exposure settings for maximum detection accuracy, while other regions maintain their own exposure characteristics.
Solution Approach 2:
The patent applies exposure control selectively only to the extent necessary for eye detection. By focusing exposure optimization purely on the eye region and leaving other areas with standard or minimal exposure adjustment, the system achieves sufficient eye detection accuracy without unnecessarily compromising the overall image exposure balance.
3Stability of the object's composition
If standard exposure control algorithms are used, then general image quality is maintained, but eye tracking performance suffers under varying lighting conditions
Solution Approach 1:
The patent implements dynamic exposure control that continuously adapts to changing lighting conditions by monitoring the eye region in real-time. The exposure settings are adjusted dynamically based on the detected lighting environment and eye region characteristics, allowing the system to maintain optimal performance across varying lighting conditions rather than relying on fixed exposure parameters.
Solution Approach 2:
The patent employs feedback mechanisms where the eye detection system continuously monitors the quality and exposure of the eye region, and this information is fed back to adjust the exposure control algorithm. This closed-loop feedback allows the system to adapt to changing lighting conditions and maintain optimal eye detection performance by continuously refining the exposure settings based on actual performance metrics.
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 significantly improves the reliability and accuracy of eye detection by optimizing exposure settings specifically for the eye region, enhancing eye tracking performance across different lighting conditions and ensuring more accurate gaze tracking.
Implementation Method 1
an image sensor device associated with an eye tracking system, the image sensor device being sensitive to IR light... captured under active illumination by at least one infrared, IR, illuminator
Implementation Method 2
obtaining, in the eye tracking system, a first image captured under active illumination by at least one infrared, IR, illuminator... using the image sensor device
Data Source
AI summary
There is provided a method, system, and non-transitory computer-readable storage medium for controlling an eye tracking system (200) to obtain a first image (500) captured under active illumination by at least one infrared, IR, illuminator (112, 113) associated with the eye tracking system (200) and at a current exposure setting, using the image sensor device (110); and, if at least one eye (100, 120) is depicted in the first image (500): define at least one region of interest, ROI, (501, 502, 503, 504) from the first image (500) comprising a group of pixels in the first image (500) representing at least a part of the depicted at least one eye (100, 120); determine a respective intensity value for each of the at least one ROI (501, 502, 503, 504); determine a second exposure setting by adjusting at least one exposure parameter of the image sensor device (110) based on the determined intensity value, or values, for the at least one ROI (501, 502, 503, 504); and set the current exposure setting to the second exposure setting.