Pupil Extraction Using Thresholding and Circle Fitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current pupil recognition systems face inefficiencies due to high computing requirements for extracting the pupil from image data, making the process time-consuming.
Innovation Solution
A system and method that utilize an image obtainer, illuminator, and pupil extractor to obtain and process image data with controlled lighting, applying a thresholding process to convert eye region data into binary images and using circle fitting to accurately extract the pupil boundary, thereby reducing computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional pupil extraction methods are used, then accurate pupil recognition can be achieved, but the computing amount becomes large and the process becomes time-consuming
Solution Approach 1:
The patent applies preliminary action by performing face detection and eye region extraction before pupil extraction. The system pre-defines the eye region of interest (ROI) based on detected face and eye positions, so that when pupil extraction is needed, only the pre-identified eye region needs to be processed. This preliminary preparation significantly reduces the computing amount and processing time for the actual pupil extraction operation.
2Measurement precision
If conventional pupil extraction methods are used, then accurate pupil recognition can be achieved, but the computing complexity increases
Solution Approach 1:
The patent applies the extraction principle by isolating and focusing computational resources on the eye region of interest rather than processing the entire face image. By extracting and defining the eye ROI based on detected eye boundaries, the system reduces the data volume and computational complexity required for pupil extraction, while maintaining accuracy by concentrating processing power on the relevant region.
3Measurement precision
If lighting strength is increased to improve image quality, then pupil extraction accuracy improves, but energy consumption increases
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the illuminator's lighting strength based on detected image quality metrics. The system monitors image brightness and contrast parameters, and only activates or increases illuminator power when image quality falls below thresholds necessary for accurate pupil extraction. This adaptive parameter adjustment ensures sufficient lighting for accuracy while minimizing unnecessary energy consumption during already well-lit conditions.
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
Facilitates efficient and accurate pupil extraction by minimizing processing time and improving reliability without requiring complex face recognition processes.
Implementation Method 1
an illuminator configured to provide a lighting at the time of obtaining the image data
Implementation Method 2
apply a thresholding process to an eye region of interest (ROI) within the image data. The thresholding process extracts pixels having intensity having a value which is less than the predetermined reference value
Implementation Method 3
apply a circle fitting process after extracting the boundary lines and select a boundary line closest to a circle among the boundary lines to thereby determine the boundary line closest to the circle as the pupil
Data Source
AI summary
A system for extracting a pupil includes an image obtainer configured to obtain front image data of a user; an illuminator configured to provide a lighting at the time of obtaining the image data; and a pupil extractor configured to receive the image data photographed in a light strength state having a predetermined reference value or more by the illuminator and apply a thresholding process to an eye region of interest (ROI) within the image data.