Machine Learning and 3D Reconstruction of Materials Surface for Nondestructive Inspection

Abstract

During the steel pipeline installation, special attention is paid to the butt weld control performed by fusion welding. The operation of the currently popular automated X-ray and ultrasonic testing complexes is associated with high resource and monetary costs. In this regard, this work is devoted to the development of alternative and cost-effective means of preliminary quality control of the work performed based on the visual testing method. To achieve this goal, a hardware platform based on a single board Raspberry Pi4 minicomputer and a set of available modules and expansion cards is proposed, and software whose main functionality is implemented based on the systemic application of computer vision algorithms and machine learning methods. The YOLOv5 object detection algorithm and the random forest machine learning model were used as a defect detection and classification system. The mean average precision (mAP) of the trained YOLOv5 algorithm based on extracted weld contours is 86.9%. A copy of YOLOv5 trained on the images of control objects showed a mAP result of 96.8%. Random forest identifying of the defect precursor based on the point clouds of the weld surface achieved a mAP of 87.5%.

Intelligent Manufacturing Technology in the Steel Industry of China: A Review

“…Online detection in steel firms has an impact on accuracy and intelligence control [ 82 ]. The first issue is that crucial parameters of the manufacturing process cannot be identified in a timely manner, requiring smelting control to rely on professional knowledge and, on occasion, batch-quality incidents.…”

A global centralized magnetic flux leakage small defect detection network

“…1. The dataset is hybrid expanded by superposition of simulated magnetic flux leakage defects with actual detected defects, and the existing data enhancement method [14] is subsequently used to expand the dataset, thereby improving the relatively monotonous shape of small defects in the dataset and improving the robustness of the model;…”

Assessing switch weld quality with 3D sensing and machine learning

“…The continuous demand to reach high product quality and reliability in industrial manufacturing is favoring the development of systems and technologies for the non-destructive inspection [1][2][3][4] of components such as thermomagnetic switches. These components need to meet high-quality assurance and reliable performance, especially after the micro welding operations on the printed circuit board (PCB).…”