This technology provides an automated approach for evaluating manufacturing quality using X-ray inspection and advanced data-driven analysis. It addresses the challenge of reliably detecting hidden internal defects and material loss in complex parts where conventional inspection methods are limited. By combining multiple forms of X-ray imaging with a trained learning framework, the system delivers scalable, high-resolution quality indicators directly during inspection. The approach supports faster, more consistent decision-making and enables manufacturers to better understand defect formation and its impact on part performance across a wide range of production processes.
Description
The invention integrates two complementary X-ray inspection modalities within a unified learning framework to improve automated defect detection and quality assessment. High-fidelity volumetric inspection data are used to create reference information that represents internal defect regions and material loss. This reference information is aligned with conventional two-dimensional X-ray images to generate training data for supervised learning models.
Once trained, the system enables quality prediction directly from standard X-ray radiographs without requiring volumetric scans during routine inspection. The framework produces multiple inspection outputs, including indicators of internal defects and localized material condition, which can be further interpreted to support quantitative quality evaluation.
The approach was demonstrated for welded manufacturing joints and is designed to be transferable to other manufacturing processes and part geometries. By enabling automated detection, classification, and quantification of internal anomalies, the technology supports high-throughput inspection while maintaining high spatial resolution and consistency.
Benefits
- Automated detection and classification of internal manufacturing defects
- Scalable inspection suitable for high-throughput production environments
- Supports quantitative quality evaluation and performance correlation
Applications and Industries
- Advanced manufacturing and industrial fabrication
- Automotive and transportation components
- Electronics and precision joining processes
Contact
To learn more about this technology, email [email protected] or call 865-574-1051.
Researchers
Contact
To learn more about this technology, email [email protected] or call 865-574-1051.