Defect Sizing Using Convolution Neural Network Applied to Guided Wave Imaging

Kulakovskyi, Andrii; Mesnil, Olivier; Chapuis, Bastien; D’Almeida, Oscar

doi:10.12783/shm2019/32195

Cited by 2 publications

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since the seminal 2012 paper on image classification [50], deep convolutional neural networks (CNN) have shown that they are well adapted to all sorts of image-related tasks. Building upon the previous work [51], we propose to use a CNN network architecture specifically tailored on the scattering matrix data targeting inversion (regression) tasks. In the following, we refer to this architecture as "SMInvNet".…”

Section: ) Convolutional Neural Network For Regressionmentioning

confidence: 99%

Ultrasonic Defect Characterization Using the Scattering Matrix: A Performance Comparison Study of Bayesian Inversion and Machine Learning Schemas

Bai

Bourdais

Miorelli

et al. 2021

IEEE Trans. Ultrason., Ferroelect., Freq. Contr.

View full text Add to dashboard Cite

Accurate defect characterisation is desirable in ultrasonic non-destructive evaluation as it can provide quantitative information about the defect type and geometry. For defect characterisation using ultrasonic arrays, high resolution images can provide the size and type information if a defect is relatively large. However, the performance of image-based characterisation becomes poor for small defects that are comparable to the wavelength. An alternative approach is to extract the far-field scattering coefficient matrix from the array data and use it for characterisation. Defect characterisation can be performed based on a scattering matrix database that consists of the scattering matrices of idealised defects with varying parameters. In this paper, the problem of characterising small surface-breaking notches is studied using two different approaches. The first approach is based on the introduction of a general coherent noise model, and it performs characterisation within the Bayesian framework. The second approach relies on a supervised machine learning (ML) schema based on a scattering matrix database, which is used as the training set to fit the ML model exploited for the characterisation task. It is shown that convolutional neural networks (CNNs) can achieve the best characterisation accuracy among the considered machine learning approaches and they give similar characterisation uncertainty to that of the Bayesian approach if a notch is favourably oriented. The performance of both approaches varied for unfavourably oriented notches, and the machine learning approach tends to give results with higher variance and lower biases.

show abstract

Section: ) Convolutional Neural Network For Regressionmentioning

confidence: 99%