Radiographic Pipe Thickness Measurement With Neural Network Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for measuring pipe thickness using computed radiography are subjective and require experienced inspectors, making it difficult to establish a standardized measurement and perform accurate historical tracking due to inconsistent image capture.
Innovation Solution
A method and apparatus using neural networks to analyze radiographic images, distinguishing pipes from non-pipe objects, and measuring thickness by identifying pixels on the inner and outer circumferential surfaces, with optional averaging or interpolation of multiple measurement methods for accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual inspection by experienced inspectors is used, then measurement accuracy can be achieved, but subjectivity and inconsistency prevent standardized measurement
Solution Approach 1:
The patent replaces manual mechanical inspection with an automated image processing system that uses edge detection algorithms and coordinate transformation to automatically measure pipe thickness from radiographic images, eliminating human subjectivity while maintaining measurement precision through standardized computational methods
Solution Approach 2:
The system enables self-service measurement by automatically processing radiographic images through a standardized workflow that includes edge detection, coordinate system transformation, and thickness calculation, allowing consistent measurements without requiring experienced inspectors for each measurement task
2Measurement precision
If visual inspection after converting image plate data is used, then pipe thickness can be assessed, but difficulty in capturing consistent images prevents accurate historical tracking
Solution Approach 1:
The patent applies preliminary action by establishing a standardized image processing pipeline that automatically performs edge detection, coordinate transformation, and thickness measurement on radiographic images, ensuring consistent processing of all images regardless of capture conditions and enabling reliable historical tracking
Solution Approach 2:
The system changes parameters by transforming coordinates from the image plate coordinate system to a standardized measurement coordinate system, and by automatically adjusting measurement parameters based on detected pipe geometry, ensuring consistent measurements across different imaging conditions and time points
3Reliability
If automated image analysis is implemented, then measurement standardization is achieved, but complexity of image processing increases
Solution Approach 1:
The patent segments the image processing task into distinct modular steps: edge detection to identify pipe boundaries, coordinate transformation to standardize the coordinate system, and thickness calculation to compute final measurements. This segmentation reduces overall complexity by making each step independent and manageable
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
Provides consistent and normalized pipe thickness measurements through image analysis, reducing subjectivity and enabling standardized historical tracking.
Implementation Method 1
measuring, by a measuring part, a pipe thickness which is a distance between a first pixel on a first outer circumferential surface of the pipe and a second pixel on an inner circumferential surface... by analyzing the pipe image
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Proposed is a method for measuring a thickness. The method includes training neural networks generating a pipe image that distinguishes a pipe from non-pipe object in a radiographic image, generating, by a recognition part, the pipe image using the neural networks; and measuring a pipe thickness as a distance between a first pixel of one outer circumferential surface of the pipe and a second pixel of an inner circumferential surface closest to the first pixel, and a total pipe thickness as a distance between the first pixel and a third pixel of a second outer circumferential surface of the pipe closest to the first pixel, by analyzing the pipe image.