Corneal Reflection Identification for Eyeglass-Resistant Gaze Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gaze detection systems using corneal reflected images face challenges in accurately distinguishing between corneal reflections and reflections from eyeglasses, leading to reduced precision in gaze detection.
Innovation Solution
A corneal reflected image identifying device that projects light from two spaced-apart light sources onto an eye, captures images, calculates distances and angles between reflected images, and uses these features to exclude reflections from eyeglasses, identifying the corneal reflected images for accurate gaze detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a plurality of synchronized cameras are used to generate corneal reflected images, then the range of gaze detection is expanded, but the precision of gaze detection deteriorates when the subject is wearing eyeglasses due to mistaken detection of reflections from lens and frame
Solution Approach 1:
The patent divides the reflection identification task into multiple independent judgment dimensions: distance between reflected images, angle of the line connecting reflected images, and area of reflected images. Each dimension provides a separate criterion for distinguishing corneal reflections from eyeglass reflections, allowing the system to make more accurate identifications by evaluating multiple segmented aspects simultaneously
Solution Approach 2:
The patent utilizes multiple parameters (distance, angle, area) to characterize reflected images and compares them against predetermined threshold values to identify corneal reflections. By changing and evaluating multiple parameters rather than relying on a single parameter, the system achieves higher precision in distinguishing corneal reflections from eyeglass reflections while maintaining the expanded detection range
2Adaptability or versatility
If reflections from eyeglasses are mistakenly detected as corneal reflected images, then the detection coverage is maintained, but the reliability of gaze detection deteriorates
Solution Approach 1:
The patent introduces intermediary judgment criteria (distance threshold, angle threshold, area threshold) that mediate between the detected reflected images and the final identification of corneal reflections. These intermediary parameters act as filters that prevent eyeglass reflections from being mistakenly identified as corneal reflections, thereby maintaining reliability while preserving detection coverage
Solution Approach 2:
The system continuously evaluates reflected images against predetermined threshold values and uses this feedback to correct identification errors. When a reflected image fails to meet the required distance, angle, or area criteria, the system rejects it as a false positive, thereby maintaining high reliability while keeping comprehensive detection coverage
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
The system enables high-precision gaze detection even when the subject is wearing eyeglasses by accurately distinguishing between corneal and eyeglass reflections, thereby improving the reliability of gaze detection.
Implementation Method 1
using a reflected image of the light at the cornea to detect a gaze
Data Source
AI summary
A corneal reflected image identifying device and gaze detection device enabling detection of a gaze with a high precision even when a person being monitored is wearing an object covering the eyes, in particular a corneal reflected image identifying device comprising a light projecting part 110 for projecting light from two light sources 110a, 110b which are spaced apart from each other toward an eye, a camera 120 for capturing an image of the eye on which light is projected, a distance calculating part 152d for using the captured image to calculate a distance between two reflected images for each combination of two reflected images among a plurality of reflected images obtained by light projected from the light sources 110a, 110b being reflected in a vicinity of the eye, and a corneal reflected image identifying part 152g for excluding a reflected image of light projected from the light sources 110a, 110b being reflected by an object covering the eye based on predetermined features and identifying a corneal reflected image to be used for gaze detection, wherein the predetermined features include the distance which the distance calculating part 152d has calculated.