Weld Defect Detection Research Articles

Data enhanced YOLOv8s algorithm for X-ray weld defect detection

ABSTRACT X-ray weld images often suffer from low contrast, poor quality, and other limitations. Additionally, due to factors such as inspection equipment and cost constraints, the number of available samples for weld images is limited, which may not suffice for deep learning training requirements. To address these challenges and enable intelligent weld seam image detection, a comprehensive approach is proposed, comprising a data enhancement method and an enhanced YOLOv8s defect detection algorithm. The method unfolds as follows: firstly, 4k+ images are reorganised into weld defect images with consistent size utilising a weld defect image feature reorganisation method, which retains as much of the weld defect information. Subsequently, data enhancement of X-ray weld defect images is carried out through translation, luminance, noise, mirroring, and cropping degradation models and the improved SRGAN model. Finally, a novel data enhancement method is proposed based on the Multi-Head Self-Attention (MHSA), the detection head, and the Channel-wise Convolutional Spatial Gating (C2f_SGE) in the improved YOLOv8s model. This enhanced defect detection method, YOLOv8sDMS, compared with the YOLOv8s model, indicates improvements in mAP50 by 0.035, mAP50:95 by 0.021, Precision by 0.037, and Recall by 0.050. The proposed improved method achieves superior results in weld defect detection.

Ultrasonic Weld Quality Inspection Involving Strength Prediction and Defect Detection in Data-Constrained Training Environments.

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.

Segmentation-assisted classification model with convolutional neural network for weld defect detection

Detecting weld defects in battery trays is crucial for the safety of new energy vehicles. Existing methods for weld surface defect detection, relying on traditional computer vision algorithms and convolutional neural networks with substantial image-level labeled data, face challenges in accurately identifying small defects, especially with limited samples. To address these issues, we developed an innovative Segmentation-Assisted Classification with Convolutional Neural Networks (SACNN) model. SACNN integrates a common feature extraction subnet, a segmentation subnet enhanced by a multi-scale feature fusion module, and a classification subnet specifically designed for precise defect detection. A joint loss function co-trains the segmentation and classification subnets using both image-level and pixel-level labels, enhancing the model’s ability to accurately detect small defect regions. Our model demonstrates notable improvement, achieving accuracy gains ranging from 2% to 18% compared to existing state-of-the-art methods, with an overall accuracy of 94.09% on an industrial dataset of battery tray welds. To further evaluate the generalization capability of our model, we evaluated it on the publicly available Magnetic Tile dataset, achieving state-of-the-art results in this challenging context. Additionally, we conducted comprehensive ablation studies to validate the contribution of each component in our approach and utilized visualization techniques to enhance the interpretability of our model. These advancements represent a significant contribution to the state of the art in aluminum alloy weld defect detection.

Method based on a multi-image feature fusion model for detecting weld defects in time-of-flight diffraction images

Method based on a multi-image feature fusion model for detecting weld defects in time-of-flight diffraction images

Research on ultrasonic phased array total focusing multi-mode fusion method

Abstract Welded seams are one of the important structural forms in the construction of industrial equipment. But during use, hazardous defects may occur within the weld. Ultrasonic phased-array total focusing method (TFM) is a relatively new post-processing imaging method that can locate defects. In this study, a multi-mode ultrasonic phased-array total focusing fusion method was explored for weld defects to achieve effective quantification and localization of different types of defects. This study explores the application of different fusion methods such as maximum fusion, average fusion and effective region fusion in multi-mode phased array total focusing method. The experimental results show that the fusion methods can improve the accuracy of defect detection compared to single imaging methods, but the effective area fusion method has the best results in detecting weld defects. Taken together, the effective region fusion method can effectively improve the resolution and signal-to-noise ratio of multi-mode imaging methods.

Defect recognition and classification from ultrasonic phased array total focusing method imaging based on domain adaption

Abstract With the rapid development of China’s economy, the demand for oil and gas resources is continuously increasing, leading to the expansion of the scale of oil and gas long-distance pipelines. Consequently, the risk of safety incidents in oil and gas pipelines is on the rise. Welding defects are identified as one of the significant factors contributing to safety incidents in long-distance oil and gas pipelines. In comparison to traditional conventional ultrasonic detection techniques, which exhibit low efficiency and accuracy, ultrasonic phased array detection technology has the capability to accurately detect welding defects in oil and gas long-distance pipelines. This is particularly true with the development of total focusing method imaging technology, enabling quantitative defect analysis and localization in oil and gas long-distance pipelines.However, in the current scenario, analyzing the types of defects in massive data from long-distance pipelines requires human intervention, leading to low detection efficiency and subjective influence on result analysis. Addressing challenges such as low efficiency in human evaluation of defects in ultrasonic phased array detection, low accuracy in traditional image recognition methods, strong subjectivity in manually extracting features, and the absence of a standardized industrial dataset for ultrasonic phased array detection defects, especially in the case of limited data on ultrasonic phased array detection of welding defects, this study focuses on small sample defects in phased array detection images. The study proposes an improved ResNet50 algorithm through domain adaptation in a deep neural network, trained on a large-scale model data, and then applied to ultrasonic phased array detection data. This approach achieves accurate identification and classification of typical volumetric defects (such as pores) and typical area defects (such as cracks) in the phased array detection of welding defects, laying the foundation for intelligent recognition of defects in long-distance oil and gas pipelines.

Improving image-based weld defects detection under interference information via a novel semantic supervision strategy

ABSTRACT Weld defects of skin-skeleton structures are invisible and image-based visual inspections using deep learning neural networks are in demand. The main limitation of previous detection algorithms is interference information in images. The reflective light and uneven brightness distribution hinder algorithms to achieve higher detection reliability. In this work, a novel supervision strategy for defects detection algorithms was proposed to break the limitation. A new semantic gate convolution block was developed to help the neural networks to distinguish between targets and interference. The block utilised semantic segmentation labels and a gate function to control the information transmitted to the detection output. Additionally, a new category-pixel-accuracy loss function was adopted to improve the effectiveness of semantic features. The new function reduced the effects of negative pixels to avoid over emphasis on semantic completeness, thus the ineffective features of interference could be eliminated. The results indicated that the back-propagated gradients from the supervision part taught the backbone networks to focus on correct objects instead of interference. To testify the validity of the current method, the algorithms was verified in burns and collapse defects. The detection accuracy under the new semantic supervision reached 99.1%, which was superior to common CNN.

Machine learning-based in-process monitoring for laser deep penetration welding: A survey

In-process monitoring (IPM) of laser deep penetration welding (LDPW) has witnessed a rapid growth in approaches that embrace machine learning algorithms, utilizing raw sensor input to generate various weld quality evaluations, instead of concentrating on thermomechanical modeling that is hypotheses-driven and hence biased by it. Benefitting from the capability to unravel hidden interactions in the complex laser welding process, numerous data-driven IPM methods have been proposed to address different problems in this area. In this survey, we present a comprehensive analysis of both classical and recent studies, covering the unique physical mechanisms, sensing techniques, methodologies, strengths, and limitations of machine learning-based IPM-LDPW systems. We delve into several critical tasks, including mechanical performance prediction, weld penetration estimation, and weld defects detection. Meanwhile, we explore the latest developments in deep learning and how to incorporate these techniques into IPM systems for LDPW. To inspire future research, we outline unresolved challenges and explore potential opportunities and new perspectives for addressing these challenges.

Welding defect detection based on YOLOv8

Abstract In this paper, we propose an improvement scheme for the unclear boundary capture issues of the YOLOv8 algorithm in weld defect detection. We replace the Conv module with the CG black module in Backbone to enhance the algorithm’s ability to extract image features so as to capture weld defects more accurately. Meanwhile, the CBAM attention mechanism is introduced in the tail of the Backbone to improve the model’s capability to learn the correlation of different regions of the image and improve the generalization performance. The experimental findings substantiate that the enhanced algorithm exhibits superior accuracy in comparison to the original algorithm, recall rate, Map@0.5, and other indicators of the improved algorithm. The enhanced algorithm demonstrates a 3.6% increase in accuracy when contrasted with the original algorithm.

Artificial intelligence enabled smart design and manufacturing of advanced materials: The endless Frontier in AI+ era

AbstractFuture‐oriented Science & Technology (S&T) Strategies trigger the innovative developments of advanced materials, providing an envision to the significant progress of leading‐/cutting‐edge science, engineering, and technologies for the next few decades. Motivated by Made in China 2025 and New Material Power Strategy by 2035, several key viewpoints about automated research workflows for accelerated discovery and smart manufacturing of advanced materials in terms of AI for Science and main respective of big data, database, standards, and ecosystems are discussed. Referring to classical toolkits at various spatial and temporal scales, AI‐based toolkits and AI‐enabled computations for material design are compared, highlighting the dominant role of the AI agent paradigm. Our recent developed ProME platform together with its functions is introduced briefly. A case study of AI agent assistant welding is presented, which is consisted of the large language model, auto‐coding via AI agent, image processing, image mosaic, and machine learning for welding defect detection. Finally, more duties are called to educate the next generation workforce with creative minds and skills. It is believed that the transformation of knowledge‐enabled data‐driven integrated computational material engineering era to AI+ era promotes the transformation of smart design and manufacturing paradigm from “designing the materials” to “designing with materials.”

Ensemble-based deep learning model for welding defect detection and classification

This study introduces an ensemble-based deep learning approach for monitoring and detecting submerged arc weld defects in weld beads and adjacent zones during Non-Destructive Testing (NDT). The methodology utilizes an open-source image dataset to categorize submerged arc weld conditions into three defect types: cracks, lack of penetration and porosity, and ideal welds (no defects). To enhance image analysis, we employ preprocessing and feature extraction in both spatial and frequency domains using segmentation techniques like grey level difference method (GLDM), grey level co-occurrence matrix (GLCM), discrete wavelet transform (DWT), Fast Fourier Transform (FFT), and texture analysis. Following image preparation, the preprocessed data undergoes training and testing in our proposed ensemble-based deep learning model. The model's performance is thoroughly assessed using key metrics such as precision, recall, F1 score, receiver operating characteristics (ROC) curve, and a confusion matrix, resulting in an impressive accuracy rate of 93.12% for detecting and classifying weld faults. In comparing our model to state-of-the-art techniques in existing literature, our ensemble-based deep learning approach consistently demonstrates a high fault detection and classification performance in the face of a simplistic MLP-based core architecture, ratifying the robustness of our feature ensemble approach. The study concludes that this model has significant potential for integration into current inspection systems, offering an efficient and robust solution for real-time condition monitoring of welds along weldment joints.

Computer vision–based surface defect identification method for weld images

Welding is the main connection type used in steel structure bridges. As the weld quality significantly affects the performance of a structure, weld defect detection is crucial in the performance evaluation of bridges. With the continuous development of computer vision (CV) techniques and image acquisition equipment, CV-based methods are being increasingly applied to weld defect detection. However, accurate identification of welding defects from several images is challenging. This paper proposes a CV-based method for detecting and quantifying surface porosity defects in welds. CV techniques were used to identify and extract weld defects, and the pixel-level sizes of the weld defects were obtained. The actual sizes of the weld defects were calculated based on a conversion between the image size and actual size. The results show that the proposed image processing method has low complexity and high accuracy in detecting weld defects, providing accurate measurement of the defect size. It is a fast and non-contact detection method for weld defect detection.

Overview of welding defect detection utilising metal magnetic memory technology

Welded joints frequently endure the composite stress of steel components. The presence of defects within these welded joints can significantly jeopardise the safety and performance of the welded structure. Magnetic memory testing technology has garnered substantial attention due to its ability to evaluate welding defects. However, the conventional zero-point pole theory, which serves as the foundation for defect assessment in practical detection, may lead to defect location and omission errors. In response to this challenge, scholars have conducted extensive research to accurately pinpoint the location and identify the types of defect within welds. This paper systematically reviews the mechanisms of magnetic memory welding defect detection, the factors that influence it, signal characteristic parameters, noise reduction in magnetic memory signals and the application of machine learning for quantitative assessment. By summarising these research advancements, this paper aims to address the current issues and provide guidance for the precise quantitative evaluation of welding defects in the future using metal magnetic memory technology.

A Lightweight Deep-Learning Algorithm for Welding Defect Detection in New Energy Vehicle Battery Current Collectors

A Lightweight Deep-Learning Algorithm for Welding Defect Detection in New Energy Vehicle Battery Current Collectors

Weld Defect Inspection of Cylindrical Secondary Batteries Based on Active Thermographic Measurement with Thermal Magnification

This paper presents an active thermographic measurement-based inspection tech-nique for detecting weld defects in the battery cap region of cylindrical secondary batteries. As the proliferation of electric vehicles continues, safety issues related to secondary batteries have become prominent. In this study, a defect detection algorithm was developed by aligning visual images and active thermographic im-ages of the cylindrical secondary battery cap region. Lock-in amplitude images during the heating-cooling process using a laser were extracted to measure the difference in temperature patterns arising from internal structural variations. Ad-ditionally, to minimize the influence of surface conditions and temperature varia-tions other than laser-induced, a 10Hz periodic heating was employed, and ther-mal magnification in the heating frequency was achieved through data processing. This approach allowed the measurement of temperature changes induced by laser heating and cooling while minimizing the impact of surface conditions and tem-perature variations unrelated to the laser. A dataset of 1000 specimens was con-structed by aligning thermographic data with vision images acquired through this method. defective samples. Using this thermographic-vision alignment dataset, a defect detection algorithm was trained to achieve a more accurate and faster in-spection performance compared to existing non-destructive testing technologies.

Significant effect of image contrast enhancement on weld defect detection

Weld defect inspection is an essential aspect of testing in industries field. From a human viewpoint, a manual inspection can make appropriate justification more difficult and lead to incorrect identification during weld defect detection. Weld defect inspection uses X-radiography testing, which is now mostly outdated. Recently, numerous researchers have utilized X-radiography digital images to inspect the defect. As a result, for error-free inspection, an autonomous weld detection and classification system are required. One of the most difficult issues in the field of image processing, particularly for enhancing image quality, is the issue of contrast variation and luminosity. Enhancement is carried out by adjusting the brightness of the dark or bright intensity to boost segmentation performance and image quality. To equalize contrast variation and luminosity, many different approaches have recently been put forth. In this research, a novel approach called Hybrid Statistical Enhancement (HSE), which is based on a direct strategy using statistical data, is proposed. The HSE method divided each pixel into three groups, the foreground, border, and problematic region, using the mean and standard deviation of a global and local neighborhood (luminosity and contrast). To illustrate the impact of the HSE method on the segmentation or detection stage, the datasets, specifically the weld defect image, were used. Bernsen and Otsu’s methods are the two segmentation techniques utilized. The findings from the objective and visual elements demonstrated that the HSE approach might automatically improve segmentation output while effectively enhancing contrast variation and normalizing luminosity. In comparison to the Homomorphic Filter (HF) and Difference of Gaussian (DoG) approaches, the segmentation results for HSE images had the lowest result according to Misclassification Error (ME). After being applied to the HSE images during the segmentation stage, every quantitative result showed an increase. For example, accuracy increased from 64.171 to 84.964. In summary, the application of the HSE method has resulted in an effective and efficient outcome for background correction as well as improving the quality of images.

An Improved YOLOv5x-Based Algorithm for IC Pin Welding Defects Detection

This study suggests an integrated circuit (IC) pin welding defect detection algorithm based on improved YOLOv5x to address the issues of low detection accuracy caused by small target size and dense pin arrangement in IC pin welding defects identification. The ability of the network to extract features is improved by effective fusing of features with various receptive fields through the inclusion of the D-SPP module to merge different channel information. The introduction of the mask self-attention mechanism module increases the network’s capacity to recognize global feature correlations and raises the algorithm’s detection precision. In order to speed up the model’s convergence and tackle the issue of sample imbalance in BBox regression, the Focal-EIoU loss function is applied. The detection accuracy and speed are increased by using the k-means++ clustering algorithm to create nine clustering centers to figure out the size of the prior box. According to the results of the experiment, the new method achieves average precisions for short-circuit, missing pin, pin-cocked, and little tin faults in IC pin welding of 96.7%, 94.5%, 95.6%, and 93.3%, respectively. The mean average precision increases to 95.0% at a threshold of 0.5, which is 13.3% and 8.9% greater than YOLOv3 and YOLOv5x, respectively. A reference value for IC pin welding defect identification is provided by the improved algorithm, which has a detection time of 0.142 seconds per image. This meets the speed requirements of IC quality inspection.

OCT monitoring data processing method of laser deep penetration welding based on HDBSCAN

To address the challenge of low accuracy in the measurement of laser deep penetration welding depth, this study introduces optical coherence tomography (OCT) technology for the direct measurement of this parameter, which is designed to enable real-time measurement and the accurate acquisition of depth information. This study has developed a laser welding depth inspection system that integrates Spectral-Domain OCT (SD-OCT), utilizing 304 stainless steel as the test material for inspecting laser deep penetration welding. A comparative analysis evaluated the impact of Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN), Density-Based Spatial Clustering of Applications with Noise (DBSCAN), and K-means clustering (K-mean) on the clustering of OCT raw data, in addition to the use of percentile filters combined with the exponential moving average method for extracting the depth-of-melt curve. The depth-of-melt curve, obtained via the HDBSCAN algorithm, more accurately reflects the actual depth curve, with an average error remaining below 5%. This error rate is 46% lower than that of DBSCAN and 42% lower than that of K-mean. The results of the experiments with continuously varying process parameters show that a high degree of accuracy can be demonstrated even when the welding process parameters are continuously varied. Experimental results indicate that the combination of the HDBSCAN algorithm with percentile filtering and the exponential moving average method significantly enhances the recognition accuracy of the deep penetration laser welding depth curve. This method provides more precise feature parameters for the real-time detection and identification of defects in deep penetration laser welding, thereby offering strong support for optimizing and controlling the laser welding process.

Advanced ANN Techniques for Precise Detection and Classification of Welding Defects

Advanced ANN Techniques for Precise Detection and Classification of Welding Defects

An X-Ray-Based Automatic Welding Defect Detection Method for Special Equipment System

An X-Ray-Based Automatic Welding Defect Detection Method for Special Equipment System