Pupil Detection Device Using Circle Fitting for Obscured Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional pupil detection technologies face challenges in accurately detecting the pupil center when it is partially obscured by corneal reflection, leading to reduced precision in visual line detection.
Innovation Solution
A pupil detection device comprising an identification unit, an extractor, a selector, a center calculator, and a pupil detector that identifies the pupil region, extracts its contour, selects points on the contour, calculates the center of a circle passing through these points, and detects the pupil center, enabling accurate detection even when the pupil is partially hidden.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional pupil detection methods are used, then the detection process is simple, but the measurement precision deteriorates when the pupil is partially obscured by corneal reflection
Solution Approach 1:
The detection process is segmented into distinct functional units: identification unit for pupil region detection, extraction unit for contour isolation, selection unit for point sampling, center calculation unit for circle fitting, and detection unit for final pupil center determination. This segmentation allows each unit to specialize in a specific task, improving overall measurement precision while maintaining manageable system complexity through modular design.
Solution Approach 2:
The invention transitions from direct 2D pupil center detection to 3D circle fitting by selecting multiple points on the pupil contour and calculating the center of the circumscribed circle. This dimensional approach (using multiple points to define a circle in 2D space) provides robustness against partial obscuration, as the circle fitting algorithm can accurately determine the pupil center even when some contour points are obscured by corneal reflection.
2Reliability
If the pupil is partially hidden by corneal reflection, then the detection robustness should improve by using contour information, but the measurement precision deteriorates due to incomplete pupil visibility
Solution Approach 1:
The system uses feedback from the circle fitting process to iteratively refine pupil center detection. By calculating the circumscribed circle from selected contour points and using its center as the pupil center, the system creates a self-correcting mechanism that compensates for partial obscuration. The circle fitting algorithm provides feedback that helps identify the true pupil center even when the visible pupil contour is incomplete or distorted by corneal reflection.
Solution Approach 2:
The invention changes the detection parameter from direct pixel intensity analysis to geometric circle fitting parameters. By transforming the problem from detecting the darkest region to fitting a circle through selected contour points, the system changes the mathematical parameters used for detection. This parameter transformation makes the detection more reliable under partial obscuration conditions, as circle fitting is inherently more robust to missing or distorted contour data than direct intensity-based methods.
Data Source
AI summary
A pupil detection device includes an identification unit that identifies a pupil region from a captured image of an eye, an extractor that extracts a contour of the pupil region identified by the identification unit, a selector that selects a plurality of points on the contour of the pupil region extracted by the extractor, a center calculator that calculates a center of a circle passing through the plurality of points selected by the selector, and a pupil detector that detects a center of the pupil region from the center of a circle calculated by the center calculator.