Enhanced Ultrasonic Flaw Detection using an Ultra-high Gain and Time-dependent Threshold

Area of Science:

• Materials Science
• Non-Destructive Testing
• Ultrasonic Testing

Background:

• Conventional C-scan systems face limitations in flaw detection sensitivity.
• Accurate identification of material anomalies is crucial for structural integrity.

Purpose of the Study:

• To enhance ultrasonic testing capability through an ultra-high gain flaw detection method.
• To develop a time-dependent threshold for improved image segmentation and anomaly identification.

Main Methods:

• Implemented an ultra-high gain approach for ultrasonic testing.
• Utilized a time-dependent threshold for automatic material anomaly segmentation.
• Employed a singly-scattered response (SSR) model with extreme value statistics to model grain noise.

Main Results:

• The developed method demonstrated superior performance over fixed threshold approaches, reducing false positives and missed flaws.
• Ultrasonic imaging experiments confirmed the effectiveness of the time-dependent threshold.
• Low gain was found to be detrimental to detecting subwavelength micro-flaws.

Conclusions:

• The novel time-dependent threshold method significantly improves flaw detection in ultrasonic testing.
• The approach offers a more reliable segmentation of material anomalies compared to traditional methods.
• The forward model shows promise for future probability of detection (POD) assessments and coarse-grained material inspection.