Object Identification Device Automatic Clustering

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing object identification devices require manual determination of posture-pattern categories, leading to arbitrary and potentially improper categorization, which can be challenging for designers, especially when dealing with numerous posture patterns.

Innovation Solution

An object identification device that automatically classifies score vectors into clusters using known clustering methods, eliminating the need for pre-determined categories by generating score vectors and using previously determined identification conditions for each cluster, allowing for accurate object identification without manual categorization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If manual determination of posture-pattern categories is used, then the identification process can be structured, but the categorization becomes arbitrary and may be improper for object identification

Engineering Contradiction:

Improveease of creating identification structureVSAvoidreliability of posture-pattern categorization

Core Design Contradiction:

Ease of manufactureVSReliability

Solution Approach 1:

The system performs automatic clustering of score vectors without requiring manual determination of posture-pattern categories. The computer automatically classifies score vectors into clusters based on their characteristics, eliminating the need for designer intervention and arbitrary categorization while improving reliability through data-driven clustering.

Inventive Principle:

Principle #25Self-service

2Adaptability or versatility

If manual categorization of posture patterns is performed, then categories can be established, but it becomes difficult for designers when dealing with numerous posture patterns

Engineering Contradiction:

Improveability to handle multiple posture patternsVSAvoidease of determining posture-pattern categories

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

The manual mechanical process of categorizing posture patterns by designers is replaced with an automated computational clustering system. The computer automatically clusters score vectors corresponding to numerous posture patterns, making the system adaptable to any number of posture patterns without increasing designer workload.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If pre-determined posture-pattern categories are used, then the identification process can proceed, but improper clusters may occur reducing identification accuracy

Engineering Contradiction:

Improvespeed of object identificationVSAvoidaccuracy of object identification

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system performs preliminary automatic clustering of score vectors into meaningful groups before the actual object identification process. This preliminary action creates accurate, data-driven clusters that improve both the speed and accuracy of subsequent identification, eliminating improper categorization while maintaining efficiency.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9367735B2Object identification device

Publication Date: 2016.06.14 DENSO CORP

AI summary

In an object identification device, each score calculator extracts a feature quantity from the image, and calculates a score using the extracted feature quantity and a model of the specified object. The score represents a reliability that the specified object is displayed in the image. A score-vector generator generates a score vector having the scores as elements thereof. A cluster determiner determines, based on previously determined clusters in which the score vector is classifiable, one of the clusters to which the score vector belongs as a target cluster. An object identifier identifies whether the specified object is displayed in the image based on one of the identification conditions. The one of the identification conditions is previously determined for the target cluster determined by the cluster determiner.