B-Scan Normalization Using Position Data for Uniform NDT Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-destructive testing (NDT) techniques produce distorted B-Scan images due to variations in scanner motion, leading to uneven A-Scan density and size, which hinder accurate feature detection and processing by automated tools.
Innovation Solution
Normalize B-Scan images using distance measurements from a position sensor to redistribute data uniformly, adjusting A-Scan density through averaging or interpolation, and slice the normalized B-Scans into fixed-size sub-images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the NDT scanner moves at variable speed during scanning, then the scanning coverage area increases, but the A-Scan density becomes uneven causing B-Scan distortion
Solution Approach 1:
The patent applies parameter changes by transforming the B-Scan image from a time-based representation to a distance-based representation. By using position sensor data to map A-Scans to their actual spatial locations and then resampling them onto a uniform distance grid, the system changes the parameter basis from temporal to spatial, eliminating distortion caused by variable scanning speed while preserving the full scanning coverage area
Solution Approach 2:
The patent introduces an intermediary processing step between data acquisition and final image generation. A position sensor acts as a mediator to capture actual scanner location data, which then serves as an intermediate reference to correctly map and redistribute A-Scan data points onto the B-Scan image, decoupling the image uniformity from the scanning speed variations
2Manufacturing precision
If the raw B-Scan is normalized to correct distortion, then the B-Scan image uniformity improves, but the computational complexity increases
Solution Approach 1:
The patent segments the B-Scan normalization process into distinct computational stages: (1) receiving and storing position sensor data, (2) matching A-Scans with corresponding position data, (3) calculating A-Scan density in distance segments, (4) determining normalization factors, and (5) generating the normalized B-Scan. This segmentation allows each step to be processed independently and optimally, reducing overall computational complexity
Solution Approach 2:
The patent performs preliminary action by pre-calculating and storing position data from the position sensor before the normalization process. The system also pre-determines A-Scan density distributions and normalization factors based on distance measurements, so that when the actual normalization is performed, the computationally intensive calculations have already been completed or are readily available
Data Source
AI summary
A method can include receiving, from a non-destructive testing (NDT) scanner, a plurality of A-Scans of a length of a scanned region, generating a raw B-Scan of the scanned region using the plurality of A-Scans, receiving, from a position sensor coupled to the NDT scanner, a plurality of distance measurements of the NDT scanner along the length of the scanned region, matching each of the plurality of A-Scans with a distance measurement of the plurality of distance measurements to form a plurality of distance-assigned A-Scans, and normalizing the raw B-Scan, using the plurality of distance-assigned A-Scans, to form a normalized B-Scan, wherein each column of the normalized B-Scan represents the same distance along the length of the scanned region.