Pupil Detection Using Dual Imaging and Red-Eye Intensity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional pupil detection methods using color saturation for red-eye detection face accuracy issues when applied to gray scale images using near-infrared light, leading to decreased detection accuracy and difficulties in line of sight detection.
Innovation Solution
A pupil detection apparatus and method that selectively outputs detection results based on red-eye occurrence intensity and correlation characteristics, using two imaging configurations with different separation distances between imaging and illumination sections to improve accuracy, independent of color saturation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If color saturation is used as red-eye detection evaluation value, then red-eye detection can be performed, but detection accuracy falls when applied to gray scale images using near-infrared light
Solution Approach 1:
The patent changes the detection parameter from color saturation (RGB-based) to brightness intensity ratios. Specifically, it calculates the ratio of brightness in the pupil region to brightness in the peripheral region, which is applicable to both color and grayscale images including near-infrared images. This parameter transformation resolves the contradiction by making the detection method universally applicable while maintaining accuracy.
Solution Approach 2:
The patent divides the eye region into multiple segments: the pupil region and the peripheral region surrounding it. By separately analyzing the brightness characteristics of each segment and comparing their ratios, the method achieves accurate pupil detection without relying on color information, thus working effectively with grayscale and near-infrared images.
2Productivity
If pupil detection is performed using conventional methods in low illuminance conditions, then detection can be attempted, but red-eye phenomenon occurs reducing detection stability
Solution Approach 1:
The patent converts the harmful red-eye phenomenon into a useful detection feature. In low illuminance conditions, the red-eye phenomenon causes the pupil region to appear brighter than the periphery. The patent leverages this brightness difference by calculating the brightness ratio between pupil and peripheral regions, transforming the previously problematic effect into a reliable detection criterion that works particularly well in low-light scenarios.
3Ease of operation
If pupil detection accuracy falls, then line of sight detection can still be performed, but accuracy of line of sight detection processing also falls
Solution Approach 1:
The patent implements a feedback mechanism where the brightness ratio calculation results directly inform the pupil detection outcome, which in turn provides accurate input for line of sight detection. By using the ratio of pupil brightness to peripheral brightness as the detection criterion, the system ensures high-quality pupil detection results that reliably support subsequent line of sight detection processing, maintaining accuracy throughout the detection chain.
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
Enables high-accuracy pupil detection and subsequent line of sight detection even in low illuminance conditions using near-infrared light, by optimizing the selection of detection results based on red-eye occurrence intensity and illuminance coefficients.
Implementation Method 1
a gray scale image using near-infrared light
Data Source
Figure 1~2
Figure 3~4
Figure 5~7
AI summary
A pupil detection apparatus and pupil detection method are provided that enable a high-accuracy detection result to be selected and output even when a gray scale image using near-infrared light is used. In a pupil detection apparatus (100), based on a calculated value of red-eye occurrence intensity that is relative brightness of brightness within a first pupil image detected by a pupil detection section (103) with respect to brightness of a peripheral image outside the first pupil image, and a correlation characteristic of red-eye occurrence intensity and a pupil detection accuracy value, a switching determination section (105) selectively outputs a detection result of the first pupil image or a detection result of a second pupil image detected by a pupil detection section (104). The pupil detection apparatus (100) has a first imaging pair comprising an imaging section (111) and an illumination section (112) separated by separation distance d1, and a second imaging pair whose separation distance d2 is greater than that of the first imaging pair. One pupil detection section (103) uses an image captured by the first imaging pair, and the other pupil detection section (104) uses an image captured by the second imaging pair.