SMART UTX: Smart Modular System for Ultrasonic Inspection in Extreme Conditions

Project Summary

Smart UTX (ultrasonic testing extreme) is an IRI project of Institute for Nuclear Technology (INETEC) in collaboration with Faculty of Electrical Engineering and Computing.

The project goal is to develop a new device for ultrasonic testing in extreme conditions of high and low temperatures, high pressures and moisture and high doses of ionizing radiation.

Non-destructive ultrasonic testing is a very useful method for detecting flaws in materials. It can be used for monitoring critical parts in:

  • Power plants
  • Oil and gas industry
  • Automotive industry
  • Space industry etc.

Testing in extreme conditions is currently limited due to a few factors. The most important one is the poor properties of piezoelectric elements used in the construction of ultrasonic probes.

Furthermore, most of the scans do not contain defects so manual inspection would be significantly faster if it could be automatized.

Objectives

Project objectives are:

  • Constructing a state-of-the-art instrument that will be able to provide a similar signal regardless of the condition in which the testing was done
  • Development of advanced algorithms that will enable automatic flaw classification

The first part is conducted by INETEC, and the second by the Image processing group at the Faculty of Electrical Engineering and Computing.

Project Team - Faculty of Electrical Engineering and Computing

  • Prof. Sven Lončarić, principal investigator
  • Assoc. Prof. Marko Subašić, researcher
  • Asst. Prof. Tomislav Petković, researcher
  • Duje Medak, researcher
  • Luka Posilović, researcher
  • Fran Milković, researcher
  • Branimir Filipović, researcher

Project Duration

The project duration is 48 months, starting from October 2018.

Publications

 

Journal articles

 

D. Medak, L. Posilović, M. Subašić, M. Budimir and S. Lončarić, "Automated Defect Detection From Ultrasonic Images Using Deep Learning", in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, vol. 68, no. 10, pp. 3126-3134,             Oct. 2021, doi: 10.1109/TUFFC.2021.3081750.

 

 

L. Posilović, D. Medak, M. Subašić, M. Budimir, S. Lončarić, "Generative adversarial network with object detector discriminator for enhanced defect detection on ultrasonic b-scans", in Neurocomputing, vol. 459, pp. 361-369, Oct 2021, doi: https://doi.org/10.1016/j.neucom.2021.06.094.

 

 

Conference articles

 

 B. Filipović, F. Milković, M. Subašić, S. Lončarić, T. Petković and M. Budimir, "Automated Ultrasonic Testing of Materials based on C-scan Flaw Classification", 12th International Symposium on Image and Signal Processing and Analysis (ISPA), 2021, pp. 230-234, doi: 10.1109/ISPA52656.2021.9552056.

 

 

L. Posilović, D. Medak, M. Subašić, T. Petković, M. Budimir and S. Lončarić, "Synthetic 3D Ultrasonic Scan Generation Using Optical Flow and Generative Adversarial Networks", 12th International Symposium on Image and Signal Processing and Analysis (ISPA), 2021, pp. 213-218, doi: 10.1109/ISPA52656.2021.9552069.

 

 

D. Medak, L. Posilović, M. Subasić, T. Petković, M. Budimir and S. Lončarić, "Rapid Defect Detection by Merging Ultrasound B-scans from Different Scanning Angles", 12th International Symposium on Image and Signal Processing and Analysis (ISPA), 2021, pp. 219-224, doi: 10.1109/ISPA52656.2021.9552050.

 

 

F. Milković, B. Filipović, M. Subašić, T. Petković, S. Lončarić and M. Budimir, "Ultrasound Anomaly Detection Based on Variational Autoencoders", 12th International Symposium on Image and Signal Processing and Analysis (ISPA), 2021, pp. 225-229, doi: 10.1109/ISPA52656.2021.9552041.

 

 

L. Posilović, D. Medak, M. Subašić, T. Petković, M. Budimir and S. Lončarić, "Flaw Detection from Ultrasonic Images using YOLO and SSD",  11th International Symposium on Image and Signal Processing and Analysis (ISPA), 2019, pp. 163-168, doi: 10.1109/ISPA.2019.8868929.

 

Funding

This research has been funded by the European Union through the European Regional Development Fund, under the grant KK.01.2.1.01.0151 (Smart UTX). The project is co-funded by the company INETEC d.o.o.