VCT Defect Detection for Composite Parts With Artifact Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting defects in composite parts using volumetric computed tomography (VCT) are time-consuming, laborious, and prone to errors due to the need for manual slice-by-slice inspection, and existing automation techniques are limited by beam hardening and scattering artifacts, especially in complex geometries.
Innovation Solution
A method involving automated defect detection using VCT data, including segmentation, normalization, denoising, region growing, and classification to identify potential defects, with self-normalization to reduce artifacts and minimize false positives, and zoning rules for operator review.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual slice-by-slice inspection is performed, then defect detection accuracy is maintained, but inspection time and labor intensity increase significantly
Solution Approach 1:
The inspection process is segmented into automated preprocessing steps (normalization, denoising, region growing) that prepare the data, followed by operator review of only the flagged potential defects. This segmentation allows the system to handle the bulk of time-consuming processing automatically while preserving human judgment for final defect confirmation.
Solution Approach 2:
The manual mechanical process of reviewing each slice is replaced with an automated computational system that performs normalization, denoising, and region growing algorithms to identify and flag potential defects, substituting human labor with automated image processing techniques.
2Reliability
If normalization to a standard (e.g., aluminum rod) is performed to reduce artifacts, then beam hardening and scattering artifacts are reduced, but the approach is limited to linear CT scans and requires geometric similarity
Solution Approach 1:
The system performs self-normalization using the part's own geometry and material properties rather than requiring an external standard object. The normalization process adapts to each part's specific characteristics, enabling artifact reduction for complex geometries without needing a reference aluminum rod or geometric similarity assumptions.
Solution Approach 2:
The normalization approach changes from using a fixed external standard to dynamically adjusting normalization parameters based on the actual part geometry and scanned volume. This allows the system to adapt to different part shapes and sizes, making it versatile for complex geometries while maintaining artifact reduction effectiveness.
3Productivity
If automated defect detection is implemented, then inspection speed and consistency are improved, but beam hardening and scattering artifacts degrade image quality
Solution Approach 1:
The system performs preliminary normalization and denoising operations on the scanned volume before defect detection begins. This preliminary processing removes beam hardening and scattering artifacts in advance, ensuring that the automated detection algorithms work with clean, high-quality images that maintain both speed and accuracy.
Solution Approach 2:
The system converts the harmful effects of beam hardening and scattering artifacts into beneficial processing opportunities by using these artifacts as input for the normalization algorithm. The normalization process specifically targets and corrects these artifact patterns, transforming image degradation into improved image quality that enhances automated detection performance.
4Reliability
If operator review of all slices is required, then consistent defect detection is maintained, but operator fatigue and variability between operators increase errors
Solution Approach 1:
Instead of requiring operators to review all slices, the system performs partial automated review using region growing and classification algorithms to flag only the most suspicious areas. This partial automation reduces operator workload and fatigue while maintaining high detection consistency, as operators focus their attention only on pre-identified potential defects rather than every slice.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances defect detection efficiency by reducing human error and inconsistency across operators, allowing for faster and more accurate identification of defects in composite parts.
Implementation Method 1
volumetric computed tomography (VCT) scan
Implementation Method 2
to reduce beam hardening and scattering artifacts
Implementation Method 3
beam hardening and scattering artifacts
Implementation Method 4
detector oriented at a plurality of projection angles relative to the object
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods, apparatus and computer-readable media for detecting potential defects in a part are disclosed. A potential defect may be automatically detected in a part, and may be reported to an operator in various ways so that the operator may review the defect and take appropriate action.