Machine Learning Based Classification of Welded Components

Kulkarni, Sudheer D.; Selvi, S. Sethu; Momin, Mohammed Zuber M; Bharadwaj, Nandan S; Navya, S R; Sudesh,; Thanki, Shiv Kumar S

doi:10.1109/i4c57141.2022.10057885

Cited by 3 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several works have proposed an automated image-based quality control using Xray [8][9][10][11][12] and visual images. [13][14][15] Yang et al 11 proposed a unified deep neural network with multi-level features for weld defect classification in radiographic images. With eleven weld defect features as inputs, the model used fine-tuning strategies to improve the generalization performance with a small dataset and to classify defects like porosity, slag inclusion, and crack.…”

Section: Non Destructive Testing Of Weldingmentioning

confidence: 99%

“…Also in this case a CNN was used to extract high-level features from the images and the features were then used to train a random forest classifier, enabling the detection of defects such as pores, spatters, or overlaps. Finally, the work proposed by Kulkarni et al 13 used machine learning algorithms trained on visual data to classify welding grades. The welds were classified into grades A to D, where grade A represents best quality welds and grade D is the worst.…”

Section: Non Destructive Testing Of Weldingmentioning

confidence: 99%

See 1 more Smart Citation

Synthetic data generation for improvement of machine learning-based optical quality control: a practical approach in the welding context

Mitri¹,

Frommknecht²,

Huber³

et al. 2023

Multimodal Sensing and Artificial Intelligence: Technologies and Applications III

View full text Add to dashboard Cite

Weld quality inspection allows the detection of defects that may compromise the quality and strength of the weld. Although visual optical inspection offers lower reliability than other non-destructive methods, it enables weld analysis at a significantly lower cost. In this context, developing machine learning-based algorithms for automatic optical weld quality recognition requires acquiring large amounts of data for training. This entails high costs in terms of time, material and energy required for test preparation. However, one possible approach to tackling the problem with limited datasets is to use synthetic data. Using such data increases the amount and variety of data available to the detection algorithm. With a focus on the context of welding, this paper presents an approach that uses synthetic data as a form of data augmentation to improve the performance of the optical detection of weld seams. Specifically, we propose a generative neural network for semantic image synthesis using a limited starting dataset. The network generates new data instances by receiving as input a semantic map of the image to be represented. Weld defects such as porosity or weld spatter are added to the semantic map so that the network synthesizes corresponding defect images. Analysing the performance on a segmentation network, experimental results show how adding synthetic data to the original data can ensure improvements in network performance.

show abstract

Section: Non Destructive Testing Of Weldingmentioning

confidence: 99%

Section: Non Destructive Testing Of Weldingmentioning

confidence: 99%

Synthetic data generation for improvement of machine learning-based optical quality control: a practical approach in the welding context

Mitri¹,

Frommknecht²,

Huber³

et al. 2023

Multimodal Sensing and Artificial Intelligence: Technologies and Applications III

View full text Add to dashboard Cite

show abstract

“…To determine the type of weld fault and classify it, the retrieved characteristics are fed into a random forest algorithm. The random forest algorithm was also regarded as one of the classification approaches to identify and detect the kind of weld flaw to accomplish Non-destructive welding (Kulkarni, et al, 2022). The CNN and SVM classifier based on the radial basis function (RBF) are used to identify aggressive and benign breast cancer.…”

Section: Related Workmentioning

confidence: 99%

Classification of weld defects using machine vision using convolutional neural network

Kanthalakshmi

Pradeepa

2023

The Scientific Temper

View full text Add to dashboard Cite

Welding is an important aspect in commercial use of almost every industry. Because weld flaws can cause irregularities or inconsistencies during welding process, welding quality control is a critical step in ensuring the product’s quality and overall longevity. This study focuses on recognizing contamination defects, lack of fusion defects, or if the weld belongs to the good weld category among the defects that occur during the welding process. This category categorization is carried out for the Convolutional Neural Network (CNN) algorithm and the accuracy metric is obtained to evaluate the efficiency of the algorithm for the 3 – class dataset. According to this research, the pure CNN approach gave an accuracy result of 96.1%.

show abstract

A Systematic Literature Review on Artificial Intelligence and Explainable Artificial Intelligence for Visual Quality Assurance in Manufacturing

Hoffmann,

Reich

2023

Electronics

View full text Add to dashboard Cite

Quality assurance (QA) plays a crucial role in manufacturing to ensure that products meet their specifications. However, manual QA processes are costly and time-consuming, thereby making artificial intelligence (AI) an attractive solution for automation and expert support. In particular, convolutional neural networks (CNNs) have gained a lot of interest in visual inspection. Next to AI methods, the explainable artificial intelligence (XAI) systems, which achieve transparency and interpretability by providing insights into the decision-making process of the AI, are interesting methods for achieveing quality inspections in manufacturing processes. In this study, we conducted a systematic literature review (SLR) to explore AI and XAI approaches for visual QA (VQA) in manufacturing. Our objective was to assess the current state of the art and identify research gaps in this context. Our findings revealed that AI-based systems predominantly focused on visual quality control (VQC) for defect detection. Research addressing VQA practices, like process optimization, predictive maintenance, or root cause analysis, are more rare. Least often cited are papers that utilize XAI methods. In conclusion, this survey emphasizes the importance and potential of AI and XAI in VQA across various industries. By integrating XAI, organizations can enhance model transparency, interpretability, and trust in AI systems. Overall, leveraging AI and XAI improves VQA practices and decision-making in industries.

show abstract

Machine Learning Based Classification of Welded Components

Cited by 3 publications

References 12 publications

Synthetic data generation for improvement of machine learning-based optical quality control: a practical approach in the welding context

Synthetic data generation for improvement of machine learning-based optical quality control: a practical approach in the welding context

Classification of weld defects using machine vision using convolutional neural network

A Systematic Literature Review on Artificial Intelligence and Explainable Artificial Intelligence for Visual Quality Assurance in Manufacturing

Contact Info

Product

Resources

About