Local Quality Measures for Volumetric Image Surface Definition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for surface definition in volume data sets from non-destructive examination techniques like computed tomography lose valuable quality information, leading to uncertain measurement accuracy and inability to assess the reliability of surface data due to imaging artifacts.
Innovation Solution
A computer-implemented method that calculates and displays local quality measures for each surface point by analyzing the gray value structure, incorporating signal-to-noise ratio, edge detection repeatability, and other parameters to provide a quality value that weights surface points differently for further analysis, allowing for the assessment of measurement uncertainty and visualization of quality information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If surface definition is performed using threshold value processes in CT data, then surface profiles can be identified, but the quality of surface definition varies locally due to artifacts and no quality information is provided
Solution Approach 1:
The patent applies local quality assessment by calculating quality measures for different regions of the surface definition. The method determines local signal-to-noise ratios and edge detection repeatability at specific locations, allowing quality to vary spatially rather than applying a uniform assessment. This resolves the contradiction by providing localized quality information where needed while maintaining the threshold value surface definition process.
Solution Approach 2:
The patent implements feedback by using detected quality measures to influence subsequent processing steps. The quality information feeds back into the measurement process, allowing for uncertainty analysis and optimization of measurement parameters. This creates a closed-loop system where quality assessment informs further action, resolving the information loss by making quality data actionable.
2Measurement precision
If complex 3D measurement technology is used to obtain surface data, then measurement data can be acquired, but uncertainty analysis cannot be performed due to lack of quality information
Solution Approach 1:
The patent performs preliminary quality assessment during the surface definition process itself, before final measurement results are generated. By calculating signal-to-noise ratios and edge detection repeatability as part of the initial processing, the system establishes a foundation for subsequent uncertainty analysis. This preliminary action ensures quality information is available when needed for reliability assessment.
Solution Approach 2:
The patent changes the parameters being measured by not only determining surface position but also calculating quality parameters such as local signal-to-noise ratio and edge detection repeatability. These additional parameters provide the necessary data for uncertainty analysis, transforming the measurement process from purely geometric to include quality metrics that enable reliability assessment.
3Ease of operation
If imaging artifacts are present in CT data, then surface definition becomes difficult and quality varies locally, but no method provides quality assessment for these variations
Solution Approach 1:
The patent uses visual representation of quality measures, effectively applying the color changes principle by encoding quality information in visual form. Different quality levels are represented differently, allowing operators to quickly identify regions affected by artifacts. This makes the surface definition process easier by providing intuitive visual feedback about quality variations without adding complex manual assessment steps.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The invention relates to a method and a device for processing a volumetric image record. The method comprises the following steps: carrying out a non-optical image scanning method on an object to be analysed and generating a volumetric image record and extracting the object contour from the volumetric image record in order to determine the position of the object surface; defining an object surface point and a surrounding area for said object surface point and analysing the grey tones within the surrounding area; calculating a quality value, which reflects the localised quality of the surface, for the object surface point on the basis of the grey-tone analysis. The device comprises equipment for carrying out a non-optical image scanning method on an object to be analysed and for generating a volumetric image record and comprises a processing device which is programmed to extract an object contour from the volumetric image record in order to determine the position of the object surface, said processing unit, in addition, being programmed to define at least one trajectory or a surrounding area for an object surface point, to analyse the grey-tone progression along the at least one trajectory or to analyse the grey tones within the surrounding area, and to calculate a quality value for the object surface point in order to represent a localised quality of the surface.