Synthetic data generation for steel defect detection and classification using deep learning
Boikov A. Payor V. Savelev R. Kolesnikov A.
July 2021MDPI AG
Symmetry
2021#13Issue 7
The paper presents a methodology for training neural networks for vision tasks on synthe-sized data on the example of steel defect recognition in automated production control systems. The article describes the process of dataset procedural generation of steel slab defects with a symmetrical distribution. The results of training two neural networks Unet and Xception on a generated data grid and testing them on real data are presented. The performance of these neural networks was assessed using real data from the Severstal: Steel Defect Detection set. In both cases, the neural networks showed good results in the classification and segmentation of surface defects of steel workpieces in the image. Dice score on synthetic data reaches 0.62, and accuracy—0.81.
Computer vision , Machine learning , Steel defect detection , Synthetic data
Text of the article Перейти на текст статьи
Department of Mineral Process, Automation of Technological Processes and Production, St. Petersburg Mining University, St. Petersburg, 199106, Russian Federation
The Ministry of Education and Science of the Republic of Kazakhstan, M. Auezov South Kazakhstan University, Shymkent, 160012, Kazakhstan
Department of Mineral Process
The Ministry of Education and Science of the Republic of Kazakhstan
10 лет помогаем публиковать статьи Международный издатель
Книга Публикация научной статьи Волощук 2026 Book Publication of a scientific article 2026