Eye Gaze Detection Using Single Light Source Corneal Reflection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting developmental disorders in young children during medical checkups lack accuracy and efficiency due to the shortage of psychiatrists and lengthy interview times, necessitating an objective and precise diagnostic assistance apparatus.
Innovation Solution
A detection apparatus using a single positioned light source, such as an LED, to accurately calculate the corneal curvature center and detect eye gaze by illuminating the subject's eyeball with near-infrared light, allowing for three-dimensional coordinate calculation of the pupil and corneal reflection, and employing diagnostic images with specific determination areas to compare gazing times between person and geometric pattern images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single light source is used for corneal reflection imaging, then device complexity and cost are reduced, but measurement precision may deteriorate due to insufficient illumination angles
Solution Approach 1:
The patent combines multiple functions into a single light source by equipping it with both a diffuse reflection imaging unit and a corneal reflection imaging unit. This allows one light source to perform what traditionally required multiple light sources, reducing device complexity while maintaining measurement precision through computational geometry calculations of the corneal curvature center.
2Measurement precision
If traditional gaze detection methods are used, then detection speed is adequate, but detection accuracy deteriorates due to inability to compensate for head movement and individual differences
Solution Approach 1:
The patent performs preliminary measurement of the distance between the pupil center and corneal curvature center before actual gaze detection. This preliminary action establishes a reference value that compensates for individual differences and head movements during subsequent gaze measurements, improving accuracy without requiring multiple time-consuming measurements.
Solution Approach 2:
The system uses feedback from the measured distance between pupil center and corneal curvature center to dynamically adjust and compensate for head movement during gaze detection. The corneal curvature center calculation incorporates this distance information to correct gaze point positions, maintaining high accuracy even when the subject moves their head.
3Measurement precision
If comprehensive eye tracking is performed to account for head movement, then detection accuracy improves, but device complexity and processing time increase
Solution Approach 1:
The single light source is designed to serve multiple functions: diffuse reflection imaging for overall eye structure, corneal reflection imaging for gaze detection, and providing geometric reference points for movement compensation. This multi-functionality allows comprehensive tracking without requiring separate dedicated systems for each function.
Solution Approach 2:
The system uses the corneal reflection and pupil images themselves to generate the reference information needed for compensation. The geometric relationships extracted from these images (distances, angles, positions) serve as self-generated feedback that automatically compensates for head movement, eliminating the need for external tracking devices.
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
Enhances detection accuracy by minimizing measurement errors and providing a more precise evaluation of gaze points, distinguishing between neurotypical and developmental disorder subjects with higher sensitivity and specificity, while also simplifying the apparatus design and reducing costs by using a single light source.
Implementation Method 1
A detection apparatus using a single positioned light source, such as an LED, to accurately calculate the corneal curvature center and detect eye gaze by illuminating the subject's eyeball with near-infrared light
Implementation Method 2
calculating a corneal curvature center position based on a line that connects the LED light source with a corneal reflection center position
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A detection apparatus includes a display (101), an imaging unit (102) an eye gaze detector (354), an output controller (356), and a gaze point detector (355). The imaging unit (102) images a subject. The eye gaze detector (354) detects an eye gaze direction of the subject based on a captured image captured by the imaging unit (102). The output controller (356) displays a diagnostic image that includes an own person image including a face image and a geometric pattern image, onto the display (101). The gaze point detector (355) detects a first gaze point as a gaze point of the subject in the person image based on the eye gaze direction and detects a second gaze point as a gaze point of the subject in a region that includes at least one of a center of a geometric pattern, a characteristic portion of the geometric pattern, a portion in which density of lines forming the geometric pattern is higher than other portions, and a portion in which the geometric pattern changes, in the geometric pattern image.