Deep convolutional neural network for weld defect classification in radiographic images

The quality of welds is critical to the safety of structures in construction, so early detection of irregularities is crucial. Advances in machine vision inspection technologies, such as deep learning models, have improved the detection of weld defects. This paper presents a new CNN model based on ResNet50 to classify four types of weld defects in radiographic images: crack, pore, non-penetration, and no defect. Stratified cross-validation, data augmentation, and regularization were used to improve generalization and avoid over-fitting. The model was tested on three datasets, RIAWELC, GDXray, and a private dataset of low image quality, obtaining an accuracy of 98.75 %, 90.255 %, and 75.83 %, respectively. The model proposed in this paper achieves high accuracies on different datasets and constitutes a valuable tool to improve the efficiency and effectiveness of quality control processes in the welding industry. Moreover, experimental tests show that the proposed approach performs well on even low-resolution images.

Advancements in Radiographic Evaluation Through the Migration into NDE 4.0

The challenges presented by the global pandemic and slump in oil prices have imposed costly avoidance measures and delayed project timeliness, but it also has created the opportunity for innovation conditions in industrial non-destructive testing. The evolutional path leveraged by Industry 4.0, present sustainable offerings of robotic platforms, digital solutions and connected devices commonly known as the Internet of Things (IoT) that may assist in recapturing some of the current losses. The landscape is broad with staggered adoptions. An overview of the Industrial Revolutions, related developments in NDE 4.0 and specific focus on North American radiography in the petroleum industry is highlighted. Additionally, focusing on the importance of shared transparency and burden throughout the value-chain to ensure efficacy throughout the migration, and the human contributor for collaborative transition and skills required for the future. The evolutional path of Industrial revolutions leads to Industry 4.0 which presents opportunities with Artificial Intelligence, and connected devices commonly known as the Internet of Things (IoT). This article focuses on three major issues for consideration in the development of a strategic plan to capitalize on the advancements of digital radiography in the petroleum sector. The components are the technological NDE 4.0 transformation, the need for a new and realistic perspective of digital radiography, and the challenges of developing a workforce to adapt to the transition of NDE 4.0.