Ultrasound Defect Detection Using Echo Amplitude Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current ultrasonic non-destructive testing (NDT) for defect detection in materials is a manual and time-consuming process, prone to missed defects due to noise in raw data and inefficient visualization, despite the use of machine learning-based defect recognition methods which require high-quality image data.
Innovation Solution
A method that processes ultrasound scan data by removing echo amplitude values after a predetermined threshold time, generating image patches to enhance data quality, and applying these patches to an automated defect recognition process, improving accuracy and reducing operator time through the use of a computer program and apparatus with an amplitude filter and image generator.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual visual inspection is used to detect defects in ultrasound scan data, then the operator can identify potential defects using visual cues, but the process is time-consuming and may miss defects due to noise in raw data
Solution Approach 1:
The ultrasound scan data is divided into multiple image patches representing different spatial and temporal regions. Each patch is independently processed and analyzed by the machine learning model, allowing parallel processing and more thorough examination of defect patterns without requiring the operator to manually scan entire large datasets sequentially
Solution Approach 2:
The manual visual inspection process is replaced with an automated machine learning-based defect recognition system. The machine learning model automatically analyzes the ultrasound image patches to identify defects, eliminating the need for manual visual inspection while improving both speed and consistency of defect detection
2Extent of automation
If machine learning-based defect recognition is applied to raw ultrasound scan data, then automation is achieved, but the accuracy is reduced due to noise and poor data quality
Solution Approach 1:
The ultrasound scan data undergoes preprocessing to generate high-quality image patches before being fed into the machine learning model. This preliminary action of data preparation and transformation ensures that the input data for automated recognition is clean, properly formatted, and optimized for defect detection, thereby maintaining high accuracy while achieving full automation
Solution Approach 2:
The raw ultrasound scan data is transformed into image patches with optimized parameters for machine learning processing. This includes adjusting spatial and temporal parameters to create patches that highlight defect features while suppressing noise, thereby improving the quality of input data for automated recognition without requiring manual inspection
3Reliability
If the entire ultrasound scan data is processed for defect detection, then comprehensive coverage is achieved, but the processing time and computational resources increase significantly
Solution Approach 1:
The large ultrasound scan dataset is segmented into multiple smaller image patches that can be processed in parallel. This segmentation allows the system to maintain comprehensive defect detection coverage across the entire scan while significantly reducing the computational burden and processing time compared to analyzing the complete dataset as a single unit
Solution Approach 2:
Instead of processing the entire ultrasound scan data uniformly, the system applies targeted analysis to specific image patches that are most likely to contain defects based on their spatial and temporal characteristics. This partial action approach maintains high defect detection coverage while reducing overall processing time and computational resources required
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
This approach automates defect detection, improving the accuracy of quality control processes by filtering out noise and focusing on relevant structures, thus reducing human effort and time required for inspections while ensuring comprehensive defect coverage.
Implementation Method 1
one or more ultrasound probes emit sound waves that propagate inside the object and receive the echoes resulting from interactions with the internal structures
Implementation Method 2
receive the echoes resulting from interactions with the internal structures
Data Source
AI summary
A defect detection method and apparatus detecting a defect in an object. The method comprises: obtaining ultrasound scan data derived from an ultrasound scan of the object under consideration, the ultrasound scan data being in the form of a set of echo amplitude values representing the amplitude of echoes received from the object during ultrasound scanning at certain spatial and temporal points; processing the ultrasound scan data to remove echo amplitude values received after a predetermined threshold time; generating at least one image from the processed ultrasound scan data; subjecting each generated image to an automated defect recognition process to determine whether there is a defect in the portion of the object represented by the image; issuing a notification indicating whether or not a defect has been found; and, if a defect has been found, storing the result of the automated defect recognition process in a defect database.