Endoscope Measurement Device Using Structure From Motion

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional endoscope devices face challenges in accurately measuring the size of defects and damages in complex or deep portions of objects, particularly when the reference object and measurement object are not on a common plane perpendicular to the optical axis, leading to low measurement accuracy in three-dimensional measurements.

Innovation Solution

A method for operating a measurement device that includes image acquisition, display, and measurement steps, where reference positions are set and measured using stored reference information, allowing for the association of reference values with images and enabling three-dimensional shape restoration and measurement based on structure from motion technology.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If scaler measurement or reference length measurement is used, then measurement accuracy is high when reference object and measurement object are on a common plane perpendicular to optical axis, but measurement accuracy becomes low when image of object having deep portion is used

Engineering Contradiction:

Improvemeasurement accuracyVSAvoidapplicability to deep portions

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent transitions from two-dimensional scaler measurement to three-dimensional measurement by introducing depth information through structure from motion technology. Multiple images captured from different positions are processed to reconstruct the three-dimensional shape of the measurement object, enabling accurate measurement of deep portions that cannot be measured accurately in two dimensions.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent introduces a reference object as an intermediary element to establish spatial relationships. By detecting feature points on both the reference object and measurement object, and utilizing the known three-dimensional shape of the reference object, the system can accurately determine positions in three-dimensional space even for deep portions of the measurement object.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Adaptability or versatility

If three-dimensional measurement using structure from motion is used, then measurement of deep portions becomes possible, but measurement accuracy decreases compared to scaler measurement on common plane

Engineering Contradiction:

Improveability to measure deep portionsVSAvoidmeasurement accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent performs preliminary actions by capturing multiple images from different positions before the actual measurement. These images are stored and processed to reconstruct the three-dimensional shape of the measurement object. This preliminary image acquisition and three-dimensional reconstruction enables subsequent accurate measurement of deep portions with improved precision.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses feature point detection and matching between multiple images as a feedback mechanism. By detecting feature points on the measurement object in multiple images and matching them with the reference object's feature points, the system iteratively refines the three-dimensional reconstruction, thereby improving measurement accuracy for deep portions.

Inventive Principle:

Principle #23Feedback

3Ease of operation

If manual lookup of reference values is required, then operation complexity increases, but measurement flexibility is maintained

Engineering Contradiction:

Improveease of setting reference positionsVSAvoidtime for manual lookup

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The patent implements self-service by enabling the measurement device to automatically detect feature points and determine reference positions without requiring manual intervention. The system automatically processes multiple images, reconstructs the three-dimensional shape, and identifies reference positions based on the reference object's known geometry, eliminating the need for manual lookup of reference values and significantly improving operational efficiency.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical operations (manual lookup and setting of reference positions) with automated image processing and computational algorithms. Structure from motion technology and feature point detection algorithms automatically determine reference positions and measurement values, substituting manual processes with automated computational methods that are both faster and more accurate.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10964044B2Method of operating measurement device, measurement device, and recording medium

Publication Date: 2021.03.30 EVIDENT CORP

AI summary

In a method of operating a measurement device, an image acquisition unit acquires a first image of a subject. A display control unit displays the first image or a second image regarding the first image on a display. A reading unit reads one or more pieces of reference information from a storage medium. The reference information represents two reference positions on the subject. The display control unit displays the one or more pieces of reference information on the display. A setting unit sets two reference positions on the first image or the second image after the first image or the second image is displayed on the display and the reference information is displayed on the display.