3D Point Cloud Clutter Scoring for Vision Alignment

Solution Overview

Problem

Current machine vision systems face challenges in accurately and efficiently aligning 3D target images with 3D model images, particularly due to the loss of information and accuracy when converting between 3D point cloud and range image representations, and the lack of practical techniques for handling clutter in 3D images.

Innovation Solution

A system and method that uses a clutter score to align 3D target images with 3D model images by identifying and calculating clutter in the target image, which is then used to determine pose candidates, rejecting poses with excessive clutter and generating alignment results based on a predetermined threshold.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If 3D point cloud images are converted to range image representations for processing, then the data can be processed with conventional 2D image processing techniques, but information and accuracy are lost during conversion

Engineering Contradiction:

Improveease of processingVSAvoidinformation loss

Core Design Contradiction:

Ease of manufactureVSLoss of information

Solution Approach 1:

The patent introduces a clutter score calculation mechanism as an intermediary step between 3D point cloud acquisition and alignment determination. This mediator evaluates the quality of the target image by quantifying clutter levels, allowing the system to identify and reject poor-quality images before they undergo conversion and processing, thereby preventing information loss from processing degraded data

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent performs preliminary evaluation of the target image by calculating a clutter score before converting to range image representation or performing alignment operations. This preliminary action identifies images with excessive clutter that would be unsuitable for accurate processing, allowing the system to reject these images in advance and avoid the information loss that would occur during conversion and processing of degraded data

Inventive Principle:

Principle #10Preliminary action

2Device complexity

If conventional 2D image processing techniques are applied to 3D range images, then processing can be simplified, but the back, sides, top and bottom of objects cannot be represented

Engineering Contradiction:

Improveprocessing complexityVSAvoidobject representation capability

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent operates directly in 3D space using point cloud coordinates (Xi, Yi, Zi) rather than converting to 2D range images. By maintaining and processing the third dimension throughout the alignment process, the system can represent and match all surfaces of objects including back, sides, top and bottom, while still using efficient computational approaches suitable for 3D data structures

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Device complexity

If 3D point cloud matching is performed without clutter scoring, then the alignment process is simpler, but accuracy is reduced due to extraneous features in the target image

Engineering Contradiction:

Improvealignment process complexityVSAvoidalignment accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent introduces a clutter score calculation mechanism as an intermediary step between 3D point cloud acquisition and alignment determination. This mediator evaluates the quality of the target image by quantifying clutter levels, allowing the system to identify and reject poor-quality images before they undergo conversion and processing, thereby preventing information loss from processing degraded data

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent performs preliminary evaluation of the target image by calculating a clutter score before converting to range image representation or performing alignment operations. This preliminary action identifies images with excessive clutter that would be unsuitable for accurate processing, allowing the system to reject these images in advance and avoid the information loss that would occur during conversion and processing of degraded data

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10452949B2System and method for scoring clutter for use in 3D point cloud matching in a vision system

Publication Date: 2019.10.22 COGNEX CORP

AI summary

This invention provides a system and method for aligning first three-dimensional (3D) point cloud image representing a model with a second 3D point cloud image representing a target, using a vision system processor. A passing overall score is established for possible alignments of the first 3D point cloud image with the second 3D point cloud image. A coverage score for at least one alignment of the first 3D point cloud image with the second 3D point cloud image is estimated so that the coverage score describes an amount of desired features in the first 3D point cloud image present in the second 3D point cloud image. A clutter score is estimated so that the clutter score describes extraneous features in the second 3D point cloud image. An overall score is computed as a difference between the coverage score and the clutter score.