A multi-view ultrasonic imaging solution for buried sewer pipe inspection

Area of Science:

• Robotics
• Non-destructive Testing
• Ultrasonic Imaging

Background:

• Air-coupled ultrasonic arrays show promise for imaging complex structures.
• Limitations include high element counts, large data, and slow processing, hindering autonomous robotic applications.

Purpose of the Study:

• To enhance the efficiency of air-coupled ultrasonic arrays for robotic sensing in buried pipe systems.
• To reduce data acquisition and processing times for improved inspection performance.

Main Methods:

• Implementation of a multi-view ultrasonic imaging algorithm to decrease element requirements.
• Application of a modified Radon transform-based method for feature extraction from 3D ultrasonic pipe images.

Main Results:

• A 24-element array using the multi-view technique achieved an average dimensional characterization error of 2.0 cm.
• This performance is comparable to a 64-element array using Tippet's fusion method (2.8 cm error).

Conclusions:

• The multi-view imaging technique significantly improves the efficiency of air-coupled ultrasonic arrays.
• Smaller arrays can achieve high-quality imaging results, making robotic inspection of buried pipes more feasible.