Face-Image Registration Device Automatic Dictionary Extraction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional face recognition systems face challenges in achieving high recognition accuracy due to issues like background brightness changes and require users to perform cumbersome procedures for registering face images under various conditions.

Innovation Solution

A face-image registration device and method that extracts representative and registration face images from moving images, registering them in a dictionary without requiring users to manually select images, thereby simplifying the process and increasing the number of registered face images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple face images are registered under various conditions to improve recognition accuracy, then recognition accuracy is improved, but user effort and complexity of registration process increases

Engineering Contradiction:

Improverecognition accuracyVSAvoiduser effort in registration

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The system automatically extracts representative face images and registers them in the dictionary without requiring user intervention. The face-image registration means autonomously performs extraction and registration of face images showing different conditions (brightness, background, expression) based on predetermined conditions, eliminating the need for users to manually select and register multiple images.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system pre-extracts multiple face images showing different conditions (different brightness, different background, different expression) before recognition is needed. By preparing these representative face images in advance and registering them in the dictionary, the system ensures high recognition accuracy is already in place when recognition operations occur.

Inventive Principle:

Principle #10Preliminary action

2Manufacturing precision

If manual selection of face images for registration is performed to ensure quality, then registration quality is improved, but time consumption and productivity decreases

Engineering Contradiction:

Improveregistration qualityVSAvoidregistration speed

Core Design Contradiction:

Manufacturing precisionVSProductivity

Solution Approach 1:

The system replaces the manual mechanical process of selecting and registering face images with an automated image processing system. The representative-face-image extracting means and face-image registration means automatically identify, extract, and register appropriate face images based on predetermined conditions, eliminating the need for manual intervention while maintaining registration quality.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system uses predetermined conditions as feedback criteria to automatically evaluate and select suitable face images for registration. By comparing extracted face images against these conditions (brightness levels, background variations, expressions), the system autonomously determines which images meet quality standards for registration without manual review.

Inventive Principle:

Principle #23Feedback

3Device complexity

If only a limited number of face images are registered to simplify the dictionary, then device complexity is reduced, but recognition accuracy under various conditions deteriorates

Engineering Contradiction:

Improvedictionary sizeVSAvoidrecognition accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The system registers multiple face images showing different local conditions (different brightness levels, different backgrounds, different expressions) in the dictionary. Each face image is selected to represent a specific condition category, ensuring comprehensive coverage of various recognition scenarios while maintaining organized structure in the dictionary.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS8848985B2Face-image registration device, face-image registration method, face-image registration program, and storage medium

Publication Date: 2014.09.30 OMRON CORP

AI summary

A face-image registration device extracts, from a moving image which is inputted thereto, a face image showing a face of a person and registers the face image in a dictionary. The face-image registration device includes representative-face-image extracting means for extracting, from the moving image, at least one face image which satisfies a predetermined representative condition, so as to obtain a representative face image, and registration-face-image extracting means for extracting, from the moving image, at least one face image which shows the person shown in the representative face image but is not the representative face image and which satisfies a predetermined registration condition, so as to obtain a registration face image. The face-image registration device also includes face-image registration means for registering, in the dictionary, the registration face image in association with the representative face image.