Pupil Detection via Luminance Gradient Vectors

Solution Overview

Problem

Conventional pupil detection devices rely on detecting corneal reflection images to locate pupils, which can fail due to low camera resolution or large distances, leading to unstable pupil location detection.

Innovation Solution

The proposed pupil detection device processes image data from a camera to calculate luminance gradient vectors and evaluation values for individual image units in the eye area, allowing for pupil location detection without relying on corneal reflection images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If corneal reflection image detection is used for pupil location detection, then detection accuracy can be achieved under ideal conditions, but system reliability deteriorates when camera resolution is low or distance is large

Engineering Contradiction:

Improvepupil location detection accuracyVSAvoidpupil location detection reliability

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent changes the detection parameter from corneal reflection image-based methods to luminance gradient vector-based methods. By calculating luminance gradient vectors for each image unit and using evaluation values derived from these vectors, the system achieves pupil location detection without relying on corneal reflection images, thereby improving reliability under various camera resolutions and distances while maintaining accuracy

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent substitutes the optical-based corneal reflection detection mechanism with an image processing-based luminance gradient analysis mechanism. Instead of detecting physical corneal reflections, the system processes image data to calculate luminance gradients and evaluation values, achieving more reliable pupil detection across different imaging conditions

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If corneal reflection image detection is used, then pupil location can be detected under ideal imaging conditions, but detection capability deteriorates when camera resolution is low

Engineering Contradiction:

Improvepupil location detection accuracyVSAvoidcorneal reflection image detection difficulty

Core Design Contradiction:

Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The patent changes the detection approach from relying on corneal reflection images to using luminance gradient vectors calculated from image data. By computing evaluation values based on luminance gradients for each image unit, the system can detect pupil locations accurately even when camera resolution is low, as the method does not depend on the presence or quality of corneal reflection images

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If corneal reflection image detection is used, then pupil location detection is possible under ideal conditions, but detection reliability decreases when distance between camera and occupant head is large

Engineering Contradiction:

Improvepupil location detection accuracyVSAvoidpupil location detection reliability

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent changes the detection parameter from corneal reflection image-based methods to luminance gradient vector-based methods. By calculating luminance gradient vectors for each image unit and using evaluation values derived from these vectors, the system achieves pupil location detection without relying on corneal reflection images, thereby improving reliability under various camera resolutions and distances while maintaining accuracy

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12277728B2Pupil detection device, line-of-sight detection device, occupant monitoring system, and pupil detection method

Publication Date: 2025.04.15 MITSUBISHI ELECTRIC CORP

AI summary

A pupil detection device includes an eye area image obtaining unit that obtains image data representing an eye area image in a captured image obtained by a camera; a luminance gradient calculating unit that calculates luminance gradient vectors corresponding to respective individual image units in the eye area image, using the image data; an evaluation value calculating unit that calculates evaluation values corresponding to the respective individual image units, using the luminance gradient vectors; and a pupil location detecting unit that detects a pupil location in the eye area image, using the evaluation values.