Automated Weld Defect Detection in Industrial Ultrasonic B-Scan Images Using Deep Learning

Visualized characterization of diversified defects in thick aerospace composites using ultrasonic B-scan

In this study, an automated ultrasonic testing system and post signal and image processing techniques are developed in order to construct ultrasonic flaw images in weldments. Image processing algorithms are built into the flaw image processing system for the automated ultrasonic testing system. The developed signal and image analysis algorithms addressed in this study include an A-Scan data compression algorithm, ultrasonic image amplification algorithm and B-scan flaw image correction algorithm(SAFT). This flaw image processing system for the automated ultrasonic testing system can be applied to various inspection fields.

Ultrasonic B-scan imaging has proven to be a useful tool in the NDT technician's toolbox, but the interpretation of the images produced can be a time-consuming process. The success of this process is highly dependant on the skill and alertness of the operator. To improve the interpretation process, the background could be removed from the image leaving only the signals which are not constant within the image, such as reflections from anomalies within the test specimen. This paper proposes a new method for automatic background noise removal and defect highlighting to aid the operator in the location and sizing of the anomalies found.

The study considers an ultrasonic characterization on the thermal distribution in vicinity for a cylindrical thermal lesion formed in a biological tissue. The cylindrical heat source is made of a standard nichrome wire with the diameter of 1 mm. The wire was inserted inside a pork muscle housed in a cuboidal container made of perspex. The heat is conducted radially outwards from the wire to the surrounding tissue. Thermal distribution near the heated wire was predicted by numerically solving a bioheat transfer function using FemLab (Comsol, Inc.). As the wire temperature was raised from the environmental temperature 20 °C to more than 80 °C in steps of 5 °C, ultrasonic B-scan images were acquired at each temperature. We assessed the feasibility of detecting the lesion boundary using changes in echogenicity, changes in centroid frequency due to attenuation, tissue moving characteristics resulting from changes in the speed of sound, and elastograms. These observations will be of use in improving ultrasonic monitoring and guiding in HIFU surgery and thermo-therapeutic process in general.

The preoperative differentiation between benign parotid gland tumors (BPGTs) and malignant parotid gland tumors (MPGTs) is of great significance for therapeutic decision-making. Deep learning (DL), an artificial intelligence algorithm based on neural networks, can help overcome inconsistencies in conventional ultrasonic (CUS) examination outcomes. Therefore, as an auxiliary diagnostic tool, DL can support accurate diagnosis using massive ultrasonic (US) images. This current study developed and validated a DL-based US diagnosis for the preoperative differentiation of BPGT from MPGT. A total of 266 patients, including 178 patients with BPGT and 88 patients with MPGT, were consecutively identified from a pathology database and enrolled in this study. Ultimately, considering the limitations of the DL model, 173 patients were selected from the 266 patients and divided into 2 groups: a training set, and a testing set. US images of the 173 patients were used to construct the training set (including 66 benign and 66 malignant PGTs) and testing set (consisting of 21 benign and 20 malignant PGTs). These were then preprocessed by normalizing the grayscale of each image and reducing noise. Processed images were imported into the DL model, which was then trained to predict the images from the testing set and evaluated for performance. Based on the training and validation datasets, the diagnostic performance of the 3 models was assessed and verified using receiver operating characteristic (ROC) curves. Ultimately, before and after combining the clinical data, we compared the area under the curve (AUC) and diagnostic accuracy of the DL model with the opinions of trained radiologists to evaluate the application value of the DL model in US diagnosis. The DL model showed a significantly higher AUC value compared to doctor 1 + clinical data, doctor 2 + clinical data, and doctor 3 + clinical data (AUC =0.9583 vs. 0.6250, 0.7250, and 0.8025 respectively; all P<0.05). In addition, the sensitivity of the DL model was higher than the sensitivities of the doctors combined with clinical data (97.2% vs. 65%, 80%, and 90% for doctor 1 + clinical data, doctor 2 + clinical data, and doctor 3 + clinical data, respectively; all P<0.05). The DL-based US imaging diagnostic model has excellent performance in differentiating BPGT from MPGT, supporting its value as a diagnostic tool for the clinical decision-making process.

Regional pulse wave velocity (PWV) is the preferred index for quantitative evaluation of arterial stiffness locally, and has shown to be high clinical importance for effective prevention and early diagnosis of cardiovascular events. However, there have been far less systematic studies on the accurate performance of the regional PWV measurement along a segment of carotid artery and relevant influence factors. In this paper, an ultrasound simulation model for pulse wave propagation is proposed to investigate the accurate performance of PWV measurement and the influence factors such as beam density and frame rate of ultrasound scanning systemically. 30 sets of B-scan ultrasonic sequence images of carotid artery at different frame rates are simulated for obtaining the corresponding RF ultrasound signals, and then the delay time of pulse wave transit at different locations along the artery segment is calculated by two cross-correlation-based data processes. Finally, the average PWV is calculated from a linear fitting algorithm applied to time-distance relationship of pulse wave. The results show that the estimated PWV of carotid artery based on ultrasound RF signals suffers from errors ranged from 7 % to 11 % The frame rates of ultrasound imaging decreased from 1127 Hz to 376 Hz do not significantly affect the accuracy of PWV measurements, but decreased to 226 Hz would result in much larger measurement errors. Moreover, the accuracy is higher by the linear fitting algorithm based on the pulse wave transiting time from beam positions apart far away.

Welding is an extensively used technique in manufacturing, and as for every other process, there is the potential for defects in the weld joint that could be catastrophic to the manufactured products. Different welding processes use different parameter settings, which greatly impact the quality of the final welded products. The focus of research in weld defect detection is to develop a non-destructive testing method for weld quality assessment based on observing the weld with an RGB camera. Deep learning techniques have been widely used in the domain of weld defect detection in recent times, but the majority of them use, for example, X-ray images. An RGB image-based solution is attractive, as RGB cameras are comparatively inexpensive compared to X-ray image solutions. However, the number of publicly available RGB image datasets for weld defect detection is comparatively lower than that of X-ray image datasets. This work achieves a complete weld quality assessment involving lap shear strength prediction and visual weld defect detection from an extremely limited dataset. First, a multimodal dataset is generated by the fusion of image data features extracted using a convolutional autoencoder (CAE) designed in this experiment and input parameter settings data. The fusion of the dataset reduced lap shear strength (LSS) prediction errors by 34% compared to prediction errors using only input parameter settings data. This is a promising result, considering the extremely small dataset size. This work also achieves visual weld defect detection on the same limited dataset with the help of an ultrasonic weld defect dataset generated using offline and online data augmentation. The weld defect detection achieves an accuracy of 74%, again a promising result that meets standard requirements. The combination of lap shear strength prediction and visual defect detection leads to a complete inspection to avoid premature failure of the ultrasonic weld joints. The weld defect detection was compared against the publicly available image dataset for surface defect detection.

Kidney volume is associated with renal function and the severity of renal diseases, thus accurate assessment of the kidney is important. Although the voxel count method is reported to be more accurate than several methods, its laborious and time-consuming process is considered as a main limitation. In need of a new technology that is fast and as accurate as the manual voxel count method, the aim of this study was to develop the first deep learning model for automatic kidney detection and volume estimation from computed tomography (CT) images of dogs. A total of 182,974 image slices from 386 CT scans of 211 dogs were used to develop this deep learning model. Owing to the variance of kidney size and location in dogs compared to humans, several processing methods and an architecture based on UNEt Transformers which is known to show promising results for various medical image segmentation tasks including this study. Combined loss function and data augmentation were applied to elevate the performance of the model. The Dice similarity coefficient (DSC) which shows the similarity between manual segmentation and automated segmentation by deep-learning model was 0.915 ± 0.054 (mean ± SD) with post-processing. Kidney volume agreement analysis assessing the similarity between the kidney volume estimated by manual voxel count method and the deep-learning model was r = 0.960 (p < 0.001), 0.95 from Lin's concordance correlation coefficient (CCC), and 0.975 from the intraclass correlation coefficient (ICC). Kidney volume was positively correlated with body weight (BW), and insignificantly correlated with body conditions score (BCS), age, and sex. The correlations between BW, BCS, and kidney volume were as follows: kidney volume = 3.701 × BW + 11.962 (R2 = 0.74, p < 0.001) and kidney volume = 19.823 × BW/BCS index + 10.705 (R2 = 0.72, p < 0.001). The deep learning model developed in this study is useful for the automatic estimation of kidney volume. Furthermore, a reference range established in this study for CT-based normal kidney volume considering BW and BCS can be helpful in assessment of kidney in dogs.

An Uncertainty-Aware Deep Learning Framework for Defect Detection in Casting Products

An adaptive filter for smoothing images corrupted by signal dependent noise is presented. The filter is mainly developed for speckle suppression in medical B-scan ultrasonic imaging. The filter is based on mean filtering of the image using appropriately shaped and sized local kernels. Each filtering kernel, fitting to the local homogeneous region, is obtained through local statistics based region growing. Performance of the proposed scheme have been tested on a B-scan image of a standard tissue-mimicking ultrasound resolution phantom. The results indicate that the filter effectively reduces the speckle while preserving the resolvable details. The performance figures obtained through computer simulations on the phantom image are presented in a comparative way with some existing speckle suppression schemes.

As a key component in the hydraulic brake system of automobiles, the brake joint directly affects the braking performance and driving safety of the vehicle. Therefore, improving the quality of brake joints is crucial. During the processing, due to the complexity of the welding material and welding process, the weld seam is prone to various defects such as cracks, pores, undercutting, and incomplete fusion, which can weaken the joint and even lead to product failure. Traditional weld seam detection methods include destructive testing and non-destructive testing; however, destructive testing has high costs and long cycles, and non-destructive testing, such as radiographic testing and ultrasonic testing, also have problems such as high consumable costs, slow detection speed, or high requirements for operator experience. In response to these challenges, this article proposes a defect detection and classification method for laser welding seams of automotive brake joints based on machine vision inspection technology. Laser-welded automotive brake joints are subjected to weld defect detection and classification, and image processing algorithms are optimized to improve the accuracy of detection and failure analysis by utilizing the high efficiency, low cost, flexibility, and automation advantages of machine vision technology. This article first analyzes the common types of weld defects in laser welding of automotive brake joints, including craters, holes, and nibbling, and explores the causes and characteristics of these defects. Then, an image processing algorithm suitable for laser welding of automotive brake joints was studied, including pre-processing steps such as image smoothing, image enhancement, threshold segmentation, and morphological processing, to extract feature parameters of weld defects. On this basis, a welding seam defect detection and classification system based on the cascade classifier and AdaBoost algorithm was designed, and efficient recognition and classification of welding seam defects were achieved by training the cascade classifier. The results show that the system can accurately identify and distinguish pits, holes, and undercutting defects in welds, with an average classification accuracy of over 90%. The detection and recognition rate of pit defects reaches 100%, and the detection accuracy of undercutting defects is 92.6%. And the overall missed detection rate is less than 3%, with both the missed detection rate and false detection rate for pit defects being 0%. The average detection time for each image is 0.24 s, meeting the real-time requirements of industrial automation. Compared with infrared and ultrasonic detection methods, the proposed machine-vision-based detection system has significant advantages in detection speed, surface defect recognition accuracy, and industrial adaptability. This provides an efficient and accurate solution for laser welding defect detection of automotive brake joints.

Nowadays, the welding process is essential in various manufacturing industrial fields, such as aerospace, vehicle production, and shipbuilding. The welding defects caused in the process need to be monitored as they can cause serious accidents and losses. Traditional computer vision methods in an industrial application are inefficient when the detection targets have variations in shape, scale, and color because the detection performance depends on the hand-crafted features. To overcome this limitation, deep learning models, such as the convolutional neural network (CNN), are applied to industrial defect detection. These CNN-based models trained on static images, however, have a low performance that cannot meet the industrial requirements. To deal with the challenge, bidirectional Convolutional Recurrent Reconstructive Network (bi-CRRN) is proposed for welding defect detection and localization based on welding video. Spatio-temporal data, specifically the forward and backward sequences, are considered in our bi-CRRN to get high detection performance. Moreover, an automatic defect detection equipment is developed to weld a material and monitor the welding bead simultaneously. We demonstrate that the proposed bi-CRRN outperforms the other segmentation network models in welding defect detection.

On-line weld pool monitoring and defect detection using ultrasonic

A large diameter high strength titanium free-hanging catenary riser was evaluated by the Demo 2000 Ti-Rise project, from initiative of the Kristin Field development license. In order to reduce the uncertainties related to the schedule, cost, and special technical issues identified in the work related to a similar riser for future installation on the A˚sgard B semi-submersible platform, a fabrication qualification of a full scale riser in titanium was run. Several full-scale production girth welds were made in an in-situ fabrication environment. The welding was performed on extruded titanium grade 23 (ASTM) pipes with an ID of 25.5″) and wall thickness of 30 mm. The main challenge was to develop a highly productive TIG orbital welding procedure, which produced welds with as low pore content as possible. It is well known that sub-surface pores often are initiation sits for fatigue cracks in high strength titanium welds. This paper describes how a greatly improved productivity was obtained in combination with a high weld quality. NDT procedures were developed whit the main on the reliability to detect and locate possible sub-surface weld defects, volumetric defects such as pores and tungsten particles and planar defects such as lack of fusion. The results from the actual Non Destructive Testing (NDT), the mechanical testing, and the fatigue testing of the subjected welds are presented. The response of the catenary is optimised by varied distribution of weight coating along the riser’s length. A satisfactory weight coating with sufficient strength, bond strength, and wear properties was developed and qualified. The riser is planned to be fabricated from extruded titanium pipes, welded together onshore to one continuous piece. The field coating is added and the riser is loaded into the sea and towed offshore and installed.

This study presents a laser ultrasonic imaging and damage detection technique that creates images of ultrasonic waves propagating on a rotating structure and identifies damage. Laser ultrasonics is attractive for nondestructive testing mainly because of two reasons: (1) ultrasonic waves can be generated and/or measured in a noncontact manner and (2) even a small defect can be detected when laser ultrasonic scanning produces ultrasonic images with high spatial resolution. However, when it comes to a moving target, it becomes challenging to create reliable ultrasonic images. In this study, ultrasonic wave propagation images are obtained from a rotating blade using a pulse laser beam for ultrasonic generation, a galvanometer for laser scanning, and an embedded piezoelectric sensor for ultrasonic measurement. To properly estimate the laser excitation points during the scanning process rather than to precisely control the excitation points, a simple but rather effective localization technique is developed so that ultrasonic images can be constructed even from a moving target. Once the ultrasonic wave propagation images are created, damage on the target structure is visualized using a specially designed standing wave filter.