3D Modeling Feature Detection for Mechanical Structures

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for creating 3D models of mechanical structures, such as pipes and ducts, are inefficient as they require extensive data processing and large point clouds, leading to lengthy processing times and the need for post-processing verification, which can be cumbersome and time-consuming.

Innovation Solution

A system and method that utilizes feature detection to create 3D solid models in real-time using minimal measurements, integrating imaging and position measurement units with a processor to associate coordinate position data with imaging data, allowing for the creation of 3D models directly in the field with reduced data filtering needs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional 3D scanning methodologies are used to create 3D models, then comprehensive visual information is captured, but processing time is significantly increased and extensive data cleanup is required

Engineering Contradiction:
Improvecomprehensive visual informationVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts and utilizes only the essential geometric features (edges, corners, vertices) from the captured image data to create 3D models, rather than processing the entire point cloud. This selective extraction of critical features dramatically reduces processing time while maintaining model accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the object recognition process into distinct feature detection stages (edge detection, corner detection, vertex identification) rather than processing all data points uniformly. This segmentation allows the system to focus computational resources on determining geometric parameters efficiently.

Inventive Principle:
Principle #1Segmentation

2Loss of information

If traditional 3D scanning methodologies are used to create 3D models, then complete object data is captured, but data cleanup and filtering requirements increase

Engineering Contradiction:
Improvecomplete object dataVSAvoiddata cleanup effort
Core Design Contradiction:
Loss of informationVSEase of manufacture

Solution Approach 1:

The system extracts only the necessary geometric features (edges, corners, vertices) required for 3D model creation, automatically discarding redundant data points. This extraction approach eliminates the need for extensive data cleanup while preserving all essential object information.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The feature detection algorithm automatically identifies and processes critical geometric features without requiring manual data filtering or cleanup. The system self-corrects by inherently focusing on relevant features, eliminating the post-processing effort typically needed to remove extraneous data.

Inventive Principle:
Principle #25Self-service

3Loss of time

If feature detection with minimal measurements is used, then processing time is reduced, but measurement precision requirements increase

Engineering Contradiction:
Improveprocessing timeVSAvoidmeasurement precision
Core Design Contradiction:
Loss of timeVSMeasurement precision

Solution Approach 1:

The patent applies high-precision measurement techniques specifically at critical feature points (edges, corners, vertices) rather than uniformly across the entire object. This localized precision approach maintains measurement accuracy where it matters most while reducing overall data collection time.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent replaces extensive mechanical scanning measurements with optical image-based feature detection. By using computer vision algorithms to identify geometric features from images, the system achieves high measurement precision with minimal physical measurements and significantly reduced processing time.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP3055648B1Method and system for 3D modeling using feature detection
Publication Date: 2019.11.06 TRIMBLE INC
  • EP3055648B1 patent drawingFigure 1
  • EP3055648B1 patent drawingFigure 2
  • EP3055648B1 patent drawingFigure 3A~3B

AI summary

A method includes providing a processor, obtaining an image of a scene including one or more objects, and presenting, using the processor, the image of the scene to a user. The method also includes receiving a geometry type associated with one of the one or more objects, receiving a set of inputs from the user related to the one of the one or more objects, and determining, using the processor, a centerline of the one of the one or more objects. The method further includes measuring, using the processor and inputs from the user, two or more coordinate positions along the centerline, receiving a dimension associated with the one of the one or more objects, and creating, using the processor, a 3D solid model using the geometry type, the dimension, the set of inputs, and the two or more coordinate positions.