Deep Learning-Based Defect Detection From Sequences of Ultrasonic B-Scans

Medak, Duje; Posilović, Luka; Subašić, Marko; Budimir, Marko; Lončarić, Sven

doi:10.1109/jsen.2021.3134452

Cited by 13 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many NDT&E technologies have applied intelligent algorithms for the automatic detection and identification of defects through artificial neural networks. For instance, convolutional neural networks have been used to automatically extract features from raw data, classify ultrasonic signals (Shi et al, 2022), and train deep learning object detectors from the sequences of ultrasonic B-scans (Medak et al, 2021). In general, the algorithm enhancements can yield better detection accuracy or faster detection speed than previous analysis approaches.…”

Section: Conclusion and Future Trendsmentioning

confidence: 99%

Nondestructive testing and evaluation techniques of defects in fiber-reinforced polymer composites: A review

Chen

Jin²

2022

Front. Mater.

View full text Add to dashboard Cite

Fiber-reinforced polymer composites have excellent mechanical properties and outstanding development potential and are cost-effective. They have increasingly been used in numerous advanced and engineering applications as materials for wind turbine blades, helicopter rotors, high-pressure pipelines, and medical equipment. Understanding and assessing structural failure promptly in the whole lifecycle of a composite is essential to mitigating safety concerns and reducing maintenance costs. Various nondestructive testing and evaluation (NDT&E) technologies based on different evaluation principles have been established to inspect defects under different conditions. This paper reviews the established types of NDT&E techniques: acoustic emission, ultrasonic testing, eddy current testing, infrared thermography, terahertz testing, digital image correlation, shearography, and X-ray computed tomography, which is divided into three categories based on the operation frequency and data processing means of the output signal that is directly under analysis. We listed four types of defects/damage that are currently of great interest, namely, voids and porosity, fiber waviness and wrinkling, delamination and debonding, as well as impact damage. To identify a suitable method for different defects/damage, we performed characterization and evaluation by using these NDT&E techniques for typical defects/damage. Then, the cost, inspection speed, benefits and limitations, etc. were compared and discussed. Finally, a brief overview of the development of the technologies and their applications in the field of composite fabrication was discussed.

show abstract

Section: Conclusion and Future Trendsmentioning

confidence: 99%

Nondestructive testing and evaluation techniques of defects in fiber-reinforced polymer composites: A review

Chen

Jin²

2022

Front. Mater.

View full text Add to dashboard Cite

show abstract

“…For instance, Medak et al trained an EfficientDet-based model to accomplish the location of multi-defects and obtained good performance on ultrasonic images [17]. They further optimized this model using high-dimensional feature map fusion [18], and the detection accuracy has substantially improved. Posilovic et al [19] introduced two new methods: YOLO and SSD algorithms for the detection of internal defects based on Bscan images, and achieve the average accuracy of 89.7% and 84.5% respectively.…”

Section: Introductionmentioning

confidence: 99%

SM-GMVAE: An intelligent evaluation model for defect depth based on few ultrasonic signals

Li,

Liu,

et al. 2023

Preprint

View full text Add to dashboard Cite

Ultrasonic non-destructive detection is widely used for recognition and estimation of structural defects. Deep learning, especially deep neural network (DNN) has become a research hotspot for defect automated evaluation. Nonetheless, most current models are based on supervised learning approaches. To improve the performance of model, more data is needed to train model. Unfortunately, the collection of data in industrial scenarios is often limited and data labeling is also a time-consuming and labor-intensive task. In order to overcome this problem, This paper proposed a novel Similarity Metric Gaussian Mixture Variational Auto-Encoder model (SM-GMVAE) that combines few-shot learning and non-destructive testing techniques to evaluate defect depth with limited data. This model is designed into two modules: feature extraction (FE) module and similarity metric (SM) module. The FE module is designed to extract the feature of defect signal via the Variational Auto-Encoder (VAE). The SM module is used to measure the similarity of two defect signal based on the Gaussian Mixture Model (GMM). Moreover, sparse filtering techniques are used to enhance the fused features in the SM module. To validate proposed model, several specimens containing defects of different depths were produced. We construct the defect dataset based on defective ultrasound detection signals and several case studies on this datasets. The results demonstrate that the GMM and sparse filtering techniques used in our model can improve the model evaluation accuracy, and the proposed model outperforms other models.

show abstract

“…Furthermore, they developed a large dataset of phased array ultrasonic images, which they were able to use to conduct a robust validation of their results using a 5-fold cross-validation. In a more recent work [15], Medak et al developed two approaches to incorporate the A-scan signal surrounding a defect area into the detection architecture. One approach was based on model expansion, the second method extracts the features separately, and they are then merged in the detection head.…”

Section: Introductionmentioning

confidence: 99%

“…The dataset consists of several C-scans with different cases of void structures. The idea was to train a CNN on 2D cases, evaluated on a 2D test set, and use the trained network to infer ultrasonic 3D data as a sequence of C-scans, long the lines of the approach using B-scans reported in [15]. Experiments were performed on two 150 × 40 × 5 mm 3 CFRP coupons.…”

Section: Introductionmentioning

confidence: 99%

A Methodology to Automatically Segment 3D Ultrasonic Data Using X-ray Computed Tomography and a Convolutional Neural Network

et al. 2023

View full text Add to dashboard Cite

Ultrasonic non-destructive testing (UT) is a proficient method for detecting damage in composite materials; however, conventional manual testing procedures are time-consuming and labor-intensive. We propose a semi-automated defect segmentation methodology employing a convolutional neural network (CNN) on 3D ultrasonic data, facilitated by the fusion of X-ray computed tomography (XCT) and Phased-Array Ultrasonic Testing (PAUT) data. This approach offers the ability to develop supervised datasets for cases where UT techniques inadequately assess defects and enables the creation of models with genuine defects rather than artificially introduced ones. During the training process, we recommend processing the 3D volumes as a sequence of 2D slices derived from each technique. Our methodology was applied to segment porosity, a common defect in composite materials, for which characteristics such as void size and shape remain immeasurable via UT. Precision, recall, F1 score, and Intersection over Union (IoU) metrics were used in the evaluation. The results of the evaluation show that the following challenges have to be faced for improvement: (i) achieving accurate 3D registration, (ii) discovering suitable similar keypoints for XCT and UT data registration, (iii) differentiating ultrasonic echoes originating from porosity versus those related to noise or microstructural features (interfaces, resin pockets, fibers, etc.), and, (iv) single out defect echoes located near the edges of the component. In fact, an average F1 score of 0.66 and IoU of 0.5 were obtained.

show abstract

Deep Learning-Based Defect Detection From Sequences of Ultrasonic B-Scans

Cited by 13 publications

References 28 publications

Nondestructive testing and evaluation techniques of defects in fiber-reinforced polymer composites: A review

Nondestructive testing and evaluation techniques of defects in fiber-reinforced polymer composites: A review

SM-GMVAE: An intelligent evaluation model for defect depth based on few ultrasonic signals

A Methodology to Automatically Segment 3D Ultrasonic Data Using X-ray Computed Tomography and a Convolutional Neural Network

Contact Info

Product

Resources

About