Face Recognition Eye Detection for Mobile Terminals

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional face recognition technologies in mobile terminals fail to accurately detect eye positions when the face is obscured or partially captured, leading to incorrect recognition and operational issues, especially under varying illumination conditions.

Innovation Solution

A method for controlling a mobile terminal based on analysis of a user's face, which includes performing face recognition and eye detection using algorithms that identify facial contours, skin color, and texture, and compensating for partial face detection by using mirroring and template matching techniques to enhance detection accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional face recognition technologies are used to detect facial components, then face recognition can be performed, but detection accuracy deteriorates when the face is obscured or partially captured

Engineering Contradiction:
Improveface recognition reliabilityVSAvoideye position detection precision
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The face detection process is divided into multiple independent detection modules for different facial components (eyes, mouth, jaw, eyebrows). Each module operates independently and can detect its specific component even when other parts of the face are obscured, allowing the system to maintain reliability by detecting available features without requiring complete face visibility

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs partial face detection by identifying only the visible facial components in each detected region, rather than requiring complete face detection. When part of the face is captured, the system detects and processes only that partial region, which is sufficient for the intended function while avoiding the need for full face visibility

Inventive Principle:
Principle #16Partial or excessive action

2Reliability

If face detection is performed using conventional methods, then facial components can be identified, but detection accuracy deteriorates under varying illumination conditions

Engineering Contradiction:
Improveface detection reliabilityVSAvoidillumination variation impact
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The detection system applies different detection strategies and parameters to different regions of the face based on their specific characteristics and illumination conditions. Each facial component region can be processed with optimized local detection parameters that adapt to the specific illumination and visibility conditions of that region, improving overall detection reliability under varying lighting

Inventive Principle:
Principle #3Local quality

3Ease of operation

If conventional face recognition is used, then basic face detection can be performed, but the system cannot provide intuitive control without direct user input

Engineering Contradiction:
Improveterminal control easeVSAvoidautomatic control extent
Core Design Contradiction:
Ease of operationVSExtent of automation

Solution Approach 1:

The mobile terminal uses the detected facial information and eye movement data to automatically control its own functions without requiring direct user input through keypads or touchscreens. The system serves itself by translating facial expressions and eye movements into control signals for various terminal functions, enabling intuitive automated control

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP2685352B1Apparatus and method of controlling mobile terminal based on analysis of user's face
Publication Date: 2020.11.25 SAMSUNG ELECTRONICS CO LTD
  • EP2685352B1 patent drawingFigure 1
  • EP2685352B1 patent drawingFigure 2
  • EP2685352B1 patent drawingFigure 3

AI summary

An apparatus and method of controlling a mobile terminal by detecting a face or an eye in an input image are provided. The method includes performing face recognition on an input image facing and being captured by an image input unit equipped on the front face of the mobile terminal; determining, based on the face recognition, user state information that includes whether a user exists, a direction of the user's face, a distance from the mobile terminal, and/or a position of the user's face; and performing a predetermined function of the mobile terminal according to the user state information. According to the method, functions of the mobile terminal may be controlled without direct inputs from the user.