Automated X-ray Image Quality Verification for Aeronautical Parts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing non-destructive aeronautical part control systems using X-ray technology rely on manual operator intervention for image quality measurement, making the process non-reproducible and unreliable.
Innovation Solution
A system that acquires radiographic images of aeronautical parts by recognizing predefined structures in a reference image using information extraction and combination with prior knowledge, employing techniques like BLOB detection, Canny detection, and Markov chain modeling to automate image quality measurement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual operator selection is used to define pixels for image quality measurement, then the process is flexible and adaptable, but the measurement is not reproducible and reliable
Solution Approach 1:
The system performs self-verification by automatically detecting and measuring image quality indicators without requiring external operator intervention. The automated detection algorithm independently identifies predefined structures in the reference part and calculates image quality metrics, eliminating human subjectivity and ensuring reproducible measurements across different operators and time periods.
Solution Approach 2:
The manual mechanical process of operator-based pixel selection is replaced with an automated digital image processing system. The system uses computer algorithms to detect edges, identify predefined structures, and calculate image quality measurements, substituting human cognitive and manual operations with automated computational processes that provide consistent and reproducible results.
2Reliability
If automated verification is implemented, then measurement reproducibility is improved, but system complexity increases
Solution Approach 1:
The reference part is pre-designed with known predefined structures (such as specific hole patterns or geometric features) before the measurement process. This preliminary preparation of the test object with standardized features allows the automated system to reliably identify and measure image quality indicators without requiring complex adaptive algorithms, thereby improving reproducibility while controlling system complexity.
Solution Approach 2:
The system changes the parameters of the reference part by incorporating predefined structures with specific geometric properties and known positions. These parameter-defined features serve as reliable targets for automated detection, enabling the system to achieve high measurement reproducibility through standardized parameter-based identification rather than complex pattern recognition.
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
The solution enables reproducible and reliable image quality measurements, eliminating the need for manual operator selection and improving the consistency of system verification.
Implementation Method 1
The invention relates to the non-destructive testing of aeronautical parts using a radiography system
Data Source
Figure 1~2
Figure 3~4
AI summary
A method for the non-destructive examination of aeronautical parts utilising a system for acquiring x-ray images of the aeronautical parts, the method comprising verifying the system for acquiring x-ray images of the aeronautical parts, the verification comprising the steps of: - acquiring (E1) an x-ray image (IR) of a known reference part (IQI) comprising predefined structures, - recognising (E2, E3, E4, E5) the predefined structures of the known reference part in the acquired x-ray image - measuring (E6) image quality on the basis of the result of the recognition step, characterised in that the recognition of the predefined structures of the known reference part in the acquired x-ray image comprises the steps of: - using information extracted from the acquired image (E2, E4) in order to produce first structure information, and - combining the first structure information with information representing a prior knowledge of the structures (E3, E5), by modelling using a Markov chain, in order to produce second structure information.