Depth Evaluation for Metal Surface Defects by Eddy Current Testing Using Deep Residual Convolutional Neural Networks

Abstract

Eddy current testing (ECT) is an effective technique for evaluating depth of metal surface defects. However, in practice, evaluation primarily relies on the experience of an operator and is often carried out by manual inspection. In this paper, we address the challenges of automatic depth evaluation of metal surface defects by virtual of state-of-the-art deep learning (DL) techniques. The main contributions are three-fold. Firstly, a highly-integrated portable ECT device is developed, taking the advantage of an advanced field programmable gate array (Zynq-7020 system on chip) and provides fast data acquisition and in-phase/quadrature demodulation. Secondly, a dataset, termed as MDDECT, is constructed using the ECT device by human operators and made openly available. It contains 48,000 scans from 18 defects of different depths and lift-offs. Thirdly, the depth evaluation problem is formulated as a time series classification problem, and various state-of-the-art 1-d residual convolutional neural networks are trained and evaluated on the MDDECT dataset. A 38-layer 1-d ResNeXt achieves an accuracy of 93.58% in discriminating the surface defects in a stainless steel sheet with depths from 0.3 mm to 2.0 mm in the resolution of 0.1 mm. In addition, results show that the trained ResNeXt1D-38 model is immune to lift-off signals.

Eddy Current Measurement for Planar Structures

“…Although the emerging deep learning (DL) models have demonstrated advantages regarding defect reconstruction and lift-off tolerance [160], the robustness of DL models is still one of the primary challenges. For instance, when measuring anisotropic ferromagnetic plates and unknown materials, the prediction errors tend to increase, due to large discrepancies between the available training samples prepared beforehand and test samples [161].…”

Deep learning techniques for non-destructive testing and evaluation

“…The results obtained, based on numerical datasets only were quite promising in view of an extension to more challenging problems, e.g., involving experimental signals. In [69], a set of deep residual convolutional neural networks has been tested for crack depth classification based on massive set of acquisitions performed on steel plate containing 20 machined slot defects. The study showed that the architectures considered were capable to distinguish the different defects classes with good accuracy.…”

Inception meets Swin Transformer: A Novel Approach for Metal Defect Recognition

“…However, CNN methods such as VGG16 [7], GoogLeNet [8] and ResNet34 [9] have significant drawbacks, including large and complex parameters, high computational complexity, high training cost, and inability to capture global feature information. For instance, Balcioglu et al [10] utilized a deep convolutional neural network (DCNN) to detect and recognize surface defects in metal gears, Meng Tian et al [11] combined ResNet with an eddy current testing technique to evaluate the depth of metal surface defects effectively, and He et al [12] employed a multi-scale convolutional neural network to classify hot rolled steel defects. These studies chose deeper convolutional neural networks to accomplish the defect recognition task.…”