3D Flow Visualization via Curvature and Torsion Classification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for visualizing 3-dimensional flow or vector field patterns fail to provide intuitive representations, often confusing users by mixing multiple flow patterns in a single image, especially for those unfamiliar with the underlying mathematics.

Innovation Solution

A system that applies transfer functions to classify flow patterns based on curvature and torsion values, allowing for the visualization of specific flow patterns by assigning renderable properties to locations within a 3D volume, and using mixing units to compute 2D images through volume rendering techniques, enabling users to define and visualize flow patterns of interest.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If conventional flow visualization methods are used to display all flow data, then complete flow information is provided, but user confusion increases due to mixing multiple flow patterns

Engineering Contradiction:
Improveflow information completenessVSAvoiduser understanding
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The patent segments the continuous flow data into discrete flow pattern classes based on curvature and torsion thresholds. Each location in the volume is classified into a specific flow pattern category (e.g., linear, curved, helical, planar, or unknown), allowing the visualization system to separate and display different flow patterns distinctly rather than showing all patterns mixed together, thus reducing user confusion while preserving complete flow information.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different visual properties (colors, opacities, rendering styles) to different flow pattern classes based on their local characteristics. Each flow pattern class is assigned specific renderable properties that make it visually distinct and recognizable. This allows users to intuitively understand different flow patterns through their visual appearance without needing to interpret complex mathematical concepts.

Inventive Principle:
Principle #3Local quality

2Ease of operation

If flow patterns are classified and visualized separately, then user understanding improves, but visualization complexity increases

Engineering Contradiction:
Improveuser understandingVSAvoidvisualization system complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent uses curvature and torsion values as key parameters to automatically classify flow patterns. By establishing threshold values for these parameters, the system can automatically determine which flow pattern class applies to each location. This parameter-based classification approach provides a systematic and manageable way to handle flow pattern visualization complexity, as users only need to understand and adjust a few key threshold parameters rather than dealing with complex visualization algorithms.

Inventive Principle:
Principle #35Parameter changes

3Ease of operation

If multiple flow pattern classes are defined and rendered, then specific flow patterns can be highlighted, but rendering time increases

Engineering Contradiction:
Improveflow pattern identificationVSAvoidrendering speed
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent performs flow pattern classification and assigns renderable properties to each location in the volume before the actual visualization rendering takes place. This pre-computation step categorizes all locations into flow pattern classes and determines their visual properties in advance. When rendering is needed, the system can quickly display the pre-classified and pre-configured flow patterns without performing complex real-time calculations, thus maintaining high rendering speed while still providing detailed flow pattern identification.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2656316B1Visualization of flow patterns
Publication Date: 2020.04.29 KONINKLIJKE PHILIPS NV
  • EP2656316B1 patent drawingFigure 1
  • EP2656316B1 patent drawingFigure 2A~2H
  • EP2656316B1 patent drawingFigure 3~4

AI summary

The invention relates to a system (SYS) for visualizing a flow within a volume of a 3-dimensional (3-D) image, the system comprising a first transfer unit (U30) for applying a first transfer function, which assigns a renderable property to each location of a first plurality of locations within the volume, on the basis of a flow pattern assigned to said location; a second transfer unit (U40) for applying a second transfer function, which assigns a renderable property to each location of a second plurality of locations within the volume, on the basis of a value of the 3-D image assigned to said location; and a mixing unit (U50) for computing a 2-D image based on the renderable property assigned to each location of the first plurality of locations and on the renderable property assigned to each location of the second plurality of locations, wherein the 2-D image visualizes the flow pattern and the 3-D image. The mixing unit can be adapted for computing a renderable property at each location of a third plurality of locations within the volume on the basis of the renderable property assigned to each location of the first and second plurality of locations, and for computing the 2-D image based on the renderable property computed at each location of the third plurality of locations using any suitable volume rendering technique known in the art, for example, the direct volume rendering technique. Alternatively, the system may be adapted for computing a first 2-D image based on the renderable property assigned to each location of the first plurality of locations and a second 2-D image based on the renderable property assigned to each location of the second plurality of locations, and for alternatingly displaying the first and second 2-D image or for merging the first and second 2-D image.