3D Navigation Watershed Segmentation for Processing Bottlenecks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Processing three-dimensional data from vision-based navigation systems is computationally challenging due to the large amounts of data generated, making it difficult to determine movement and position within a reasonable time frame.

Innovation Solution

The system employs a watershed method for segmenting three-dimensional frames by computing residual maps based on orthogonal distances from data points to fitted polynomial surfaces, followed by morphological processing and watershed algorithm application to generate stable environmental features, allowing for efficient navigation solution calculation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If three-dimensional scan data is processed using traditional methods, then navigation accuracy can be achieved, but the processing time becomes unreasonably long due to large data amounts

Engineering Contradiction:
Improvenavigation accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies segmentation by dividing the three-dimensional scan data into multiple frames and further segmenting each frame into distinct surfaces using a watershed algorithm. This breaks down the large computational task of processing entire point clouds into smaller, manageable surface segments, enabling faster processing while maintaining navigation accuracy through precise surface identification and matching between frames.

Inventive Principle:
Principle #1Segmentation

2Loss of information

If traditional three-dimensional data processing methods are used, then complete environmental description is achieved, but computational complexity increases making real-time navigation difficult

Engineering Contradiction:
Improveenvironmental description completenessVSAvoidcomputational complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent extracts only the essential surface information from complete three-dimensional scan data by fitting polynomial surfaces to point cloud data and identifying dominant surfaces. This extraction process removes redundant information while preserving the critical geometric features needed for navigation, significantly reducing computational complexity while maintaining sufficient environmental description for accurate navigation.

Inventive Principle:
Principle #2Taking out (Extraction)

3Manufacturing precision

If polynomial surface fitting is applied to all data points, then surface accuracy is improved, but processing speed decreases due to increased calculations

Engineering Contradiction:
Improvesurface fitting accuracyVSAvoidprocessing speed
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent applies polynomial surface fitting selectively rather than to all data points uniformly. The watershed algorithm identifies and segments only the dominant surfaces that are most relevant for navigation, applying accurate polynomial fitting only to these critical surfaces. This partial application of surface fitting maintains sufficient accuracy for navigation while avoiding the computational overhead of processing every data point, thus improving processing speed.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP2757526B1Systems and methods for 3D data based navigation using a watershed method
Publication Date: 2019.08.21 HONEYWELL INTERNATIONAL INC
  • EP2757526B1 patent drawingFigure 1
  • EP2757526B1 patent drawingFigure 2
  • EP2757526B1 patent drawingFigure 3~4

AI summary

Systems and methods for 3D data based navigation using a watershed method are provided. In at least one embodiment, a method for segmenting three-dimensional frames of data comprises acquiring at least one frame from at least one sensor, wherein the at least one frame provides a three-dimensional description of an environment containing the at least one sensor; and identifying a surface in the at least one frame. The method further comprises computing at least one residual map for the at least one frame based on the orthogonal distance from data points on the surface to at least one polynomial surface fitted to the surface; and segmenting the at least one residual map by performing a watershed algorithm on the residual map.