Eye Feature Identification Using Iterative Multi-Step Filtering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing eye tracking technologies rely on infrared imaging, which limits their application to specialized hardware and is not compatible with general-purpose computing devices like smartphones and laptops, making it difficult to achieve robust and accurate eye feature detection on these platforms.

Innovation Solution

A multi-step filter approach using RGB images captured by webcams or smartphones, which includes noise reduction and contrast stretching operations, applied iteratively with varying parameters to refine pupil detection, enabling accurate eye tracking on conventional computing hardware without specialized equipment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If infrared imaging is used for eye tracking, then measurement precision is improved, but device complexity increases and adaptability decreases

Engineering Contradiction:

Improveeye feature detection accuracyVSAvoidhardware compatibility

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent uses visible light images as a copy or alternative to infrared images for eye tracking. Instead of requiring specialized infrared sensors, the system processes standard visible light camera images through enhanced image processing algorithms to achieve eye feature detection, thereby copying the functionality of infrared eye tracking using conventional hardware.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the infrared optical system with a visible light optical system. By substituting the infrared imaging mechanism with standard visible light capture and compensating through software-based image processing enhancements, the system eliminates the need for specialized infrared hardware while maintaining eye tracking capability.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If infrared imaging is used for eye tracking, then measurement precision is improved, but device complexity increases

Engineering Contradiction:

Improveeye feature detection accuracyVSAvoidhardware requirements

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent changes the operating parameters of the imaging system by using visible light wavelengths instead of infrared wavelengths. This parameter change allows the use of standard camera hardware rather than specialized infrared sensors, reducing device complexity while maintaining detection accuracy through software-based image enhancement techniques.

Inventive Principle:

Principle #35Parameter changes

3Manufacturing precision

If multi-step filter with multiple iterations is applied, then manufacturing precision is improved, but loss of time increases

Engineering Contradiction:

Improvepixel classification accuracyVSAvoidprocessing time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary noise reduction filtering to the visible light images before performing eye feature detection. By pre-processing the images to reduce noise and enhance relevant features, the system improves subsequent pixel classification accuracy while potentially reducing the number of iterative steps needed, thus balancing processing time with detection precision.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10043075B2Eye feature identification

Publication Date: 2018.08.07 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

An image of an eye is obtained via a camera. A multi-step filter is applied to the image for multiple iterations. Applying the multi-step filter includes, for each iteration, performing one or more pixel merge operations on the image. The pixel merge operations are controlled based on one or more input parameters to control whether or not the iteration classifies pixels of the image as corresponding to a feature of the eye. The one or more input parameters vary from at least one iteration to another. The iterations each output a provisional output, in which some pixels of the image are deemed as corresponding to the feature of the eye. The provisional outputs provide diverse definitions of the eye feature, and may be combined in various ways to yield a refined output, in which some pixels of the image are deemed as corresponding to the feature of the eye.