A deep learning based methodology for artefact identification and suppression with application to ultrasonic images

This paper proposes a deep learning framework for artefact identification and suppression in the context of nondestructive evaluation. The model, based on the concept of autoencoders, is developed for enhancing ultrasound inspection and defect identification through images obtained from full matrix capture data and the total focusing method. An experimental case study is used to prove the effectiveness of the method while exploring its practical limitations. A comparison with a state-of-the-art methodology based on image analysis is addressed for the identification and suppression of artefacts. In general, the proposed method efficiently provides accurate suppression of artefacts in complex scenarios, even when the defect is located below the footprint of the ultrasonic probe, and also yields the physical parameters needed for imaging as a by-product.

Automated summary

This paper presents a deep learning framework for artefact identification and suppression in nondestructive evaluation, which is based on the concept of autoencoders. Through experimental case study, it proves to be effective in accurately suppressing artefacts in complex scenarios and provides the physical parameters needed for imaging as well. Comparisons with state-of-the-art methodology based on image analysis are also made for artefact identification and suppression.