Multi-Angle Object Identification Using Pre-Registered Feature Databases

Solution Overview

Problem

Existing object identification systems face inaccuracies and complexity when identifying the same object from different imaging points, leading to potential misidentification or repeated identification processes.

Innovation Solution

A system and method that register object information in a database and compare it with image data from multiple angles, using a first imaging unit and a second imaging unit to capture objects from predetermined and different directions, and an identification unit to accurately identify objects by comparing feature amounts extracted from images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If object identification is performed based on distance from imaging point to object, then identification can be conducted, but the distance may change even for the same object causing wrong identification or requiring multiple identification processes

Engineering Contradiction:

Improveidentification accuracyVSAvoididentification process complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system performs preliminary registration of object information including images taken from multiple directions and distances before actual identification occurs. This pre-established database allows the identification unit to compare new images against registered images without needing to recalculate distance-based parameters, thereby ensuring consistent identification results regardless of imaging conditions.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces an image database as an intermediary between the imaging units and the identification process. Instead of directly comparing distance parameters, the system uses registered images from multiple angles as reference mediators. The identification unit compares the feature amounts of new images against these intermediary reference images, eliminating the reliability issues caused by distance variations.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If multiple identification processes are performed to ensure accurate object identification, then identification accuracy improves, but the process becomes more complex and time-consuming

Engineering Contradiction:

Improveidentification accuracyVSAvoididentification time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary registration of object information including images taken from multiple directions and distances before actual identification occurs. This pre-established database allows the identification unit to compare new images against registered images without needing to recalculate distance-based parameters, thereby ensuring consistent identification results regardless of imaging conditions.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The registered database serves multiple functions simultaneously: it stores reference images from various angles, captures feature amount information, and provides comparison data for identification. This multi-functional database eliminates the need for separate identification processes for different imaging conditions, reducing both complexity and time while maintaining high accuracy.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Adaptability or versatility

If object identification relies on imaging point and imaging direction information, then identification can be performed, but the same object may appear different from different directions leading to identification errors

Engineering Contradiction:

Improveidentification from multiple directionsVSAvoididentification accuracy

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system performs preliminary registration of object information including images taken from multiple directions and distances before actual identification occurs. This pre-established database allows the identification unit to compare new images against registered images without needing to recalculate distance-based parameters, thereby ensuring consistent identification results regardless of imaging conditions.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent transitions from relying on single-dimension parameters (distance, imaging angle) to multi-dimensional feature amount comparisons. By extracting and comparing feature amounts from images taken from multiple directions, the system creates a comprehensive multi-dimensional reference database that accurately represents the object from all angles, thereby maintaining high identification accuracy across varying imaging conditions.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS9959483B2System and method for information identification

Publication Date: 2018.05.01 OPTIM

AI summary

The present invention is to provide a system and a method for information identification to more accurately identify an object and make the process more easily. The system for information identification includes a database that registers information on an object; images the object from a predetermined direction by using the first imaging unit; images the object from a different direction by the second imaging unit; registers information on the object imaged by the first imaging unit in the database; and compares information on the object imaged by the second imaging unit with information registered in the database to identify the object.