Radiographic Defect Detection via Residual Image Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual radiographic inspection in industrial settings is prone to operator fatigue, leading to low reliability and inefficiency in defect detection, especially in complex and high-volume production environments, where existing automated techniques require large training sets and are affected by variations in object structure and flaw morphology.
Innovation Solution
A method and system that acquire radiographic image data, identify regions of interest, align inspection test images with reference images to compute residual images, and calculate defect probability values for pixel-based defect identification, enhancing defect detection accuracy and efficiency by utilizing prior knowledge and statistical modeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual inspection is used, then operator flexibility is maintained, but inspection reliability decreases due to operator fatigue
Solution Approach 1:
The system performs self-inspection by automatically comparing test images against reference images and defect models, eliminating the need for human operators while maintaining high reliability through automated defect probability computation and statistical analysis
Solution Approach 2:
The patent replaces manual visual inspection with an automated computer-based system that uses image processing algorithms, statistical modeling, and probability calculations to detect defects, substituting human cognitive processes with mechanical computation
2Productivity
If automated defect recognition techniques are used, then inspection efficiency increases, but system complexity increases due to large training sets and supervised learning schemes
Solution Approach 1:
The system performs preliminary actions by pre-computing defect probability values and creating statistical models during system setup, which are then reused during inspection without requiring large training sets during operation, reducing ongoing system complexity
Solution Approach 2:
The patent changes the approach from using large training sets with labeled flaws to using statistical parameters and probability distributions that can be computed from fewer data points, simplifying the system while maintaining automated inspection capabilities
3Measurement precision
If feature-based classification with supervised learning is used, then defect detectability improves, but training time and operation setup time increase
Solution Approach 1:
The system extracts only the essential statistical features and probability parameters needed for defect detection, rather than using comprehensive feature sets requiring extensive training, thereby reducing training time while maintaining detectability
Solution Approach 2:
The patent uses a partial approach by focusing on key statistical parameters and probability computations rather than exhaustive feature analysis, achieving sufficient defect detectability with reduced training requirements
Data Source
AI summary
A method for identifying defects in radiographic image data corresponding to a scanned object is provided. The method includes acquiring radiographic image data corresponding to a scanned object. In one embodiment, the radiographic image data includes an inspection test image and a reference image corresponding to the scanned object. The method includes identifying one or more regions of interest in the reference image and aligning the inspection test image with the regions of interest identified in the reference image, to obtain a residual image. The method further includes identifying one or more defects in the inspection test image based upon the residual image and one or more defect probability values computed for one or more pixels in the residual image.