Point-Cloud Clip Filter for Occlusion-Free Region Selection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Point clouds generated by laser scanners from multiple viewpoints can be difficult for users to interpret due to inclusion of occluded elements and lack of clear surface definitions, making it challenging to select and isolate specific regions of interest for modeling and operations.

Innovation Solution

A method is introduced that allows users to define a clip region in a first viewer, which includes selecting a specific point and specifying a clip volume, and then applies this region to a second viewer to render only the data points within this volume, effectively filtering out unwanted data and providing a clearer, more interpretable image.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If point cloud data from multiple scan positions is combined in a single coordinate system, then the completeness of the scanned scene is improved, but the interpretability and clarity of the rendered image deteriorates due to occluded elements from different viewpoints

Engineering Contradiction:
Improvecompleteness of scanned sceneVSAvoidinterpretability of rendered image
Core Design Contradiction:
Quantity of substanceVSEase of operation

Solution Approach 1:

The patent divides the point cloud data into multiple view-specific data sets, where each data set contains only points visible from a particular scan position. This segmentation allows the system to maintain completeness across multiple views while ensuring that each individual rendered image contains only non-occluded elements, thereby resolving the contradiction between scene completeness and image interpretability.

Inventive Principle:
Principle #1Segmentation

2Quantity of substance

If all data points from multiple viewpoints are included in the rendered image, then the completeness of the scene representation is improved, but the accuracy of surface definition deteriorates due to inclusion of occluded elements

Engineering Contradiction:
Improvecompleteness of scene representationVSAvoidaccuracy of surface definition
Core Design Contradiction:
Quantity of substanceVSManufacturing precision

Solution Approach 1:

The patent performs preliminary processing to determine visibility of each data point from the scan position before rendering. By pre-calculating which points are visible and which are occluded, the system can accurately define surfaces in the rendered image while maintaining completeness of the overall scene representation across multiple views.

Inventive Principle:
Principle #10Preliminary action

3Area of stationary object

If point cloud data from multiple locations is combined, then the coverage of the scanned scene is improved, but the difficulty of selecting and isolating regions of interest increases

Engineering Contradiction:
Improvecoverage of scanned sceneVSAvoiddifficulty of selecting regions of interest
Core Design Contradiction:
Area of stationary objectVSDifficulty of detecting and measuring

Solution Approach 1:

The patent segments the combined point cloud data into view-specific data sets, where each set corresponds to a particular scan position. This segmentation simplifies region selection by allowing users to work with smaller, more manageable data subsets that are already organized by viewpoint, thereby reducing the difficulty of detecting and measuring regions of interest while maintaining comprehensive scene coverage.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP2159756B1Point-cloud clip filter
Publication Date: 2013.10.02 LEICA GEOSYSTEMS AG
  • EP2159756B1 patent drawingFigure 1A
  • EP2159756B1 patent drawingFigure 1B
  • EP2159756B1 patent drawingFigure 2

AI summary

A viewer renders a first image from a point cloud of data of a scanned scene accessed from a data store. A region of interest may then be defined. In response, the image is updated to only include data included within the region of interest. Alternatively, the region of interest may be provided to a second viewer, where a second image is rendered or updated to only include data included within the region of interest.