DIGITIZATION OF EMBOSSED NUMBERS ON CONTINUOUS STEEL CASTING BILLETS

1 DAVID Jiří
Co-authors:
1 GARZINOVÁ Romana 1 PRAŽÁKOVÁ Veronika 1 SLÁČALA Jaroslav
Institution:
1 VSB - Technical University of Ostrava, Ostrava, Czech Republic, EU, j.david@vsb.cz
Conference:
28th International Conference on Metallurgy and Materials, Hotel Voronez I, Brno, Czech Republic, EU, May 22nd - 24th 2019
Proceedings:
Proceedings 28th International Conference on Metallurgy and Materials
Pages:
1892-1897
ISBN:
978-80-87294-92-5
ISSN:
2694-9296
Published:
4th November 2019
Proceedings of the conference were published in Web of Science and Scopus.
Metrics:
34 views / 13 downloads
Abstract

Computer Vision is currently one of the most advanced and fastest growing areas of computing and software development. It can be used to recognize objects from the captured image. It is a visual image and video recognition system coupled with artificial intelligence. Industrial vision using industrial cameras is currently used in many industrial areas. Initial capital investment into the vision system has fast economic return, depending on the cost of the system, the number of human operators replaced, production capacity and other parameters. In the case of a properly designed and configured system that can often fully eliminate the human factor. Typical tasks in machine vision can be recognition and counting of products using industrial cameras, positioning, dimensioning, or optical quality control.The paper will describe the use of machine vision in the metallurgical industry - specifically for the numerical identification of embossed numbers on continuous steel casting billets. The basic requirement of the operation was to create a system for billet identification and archiving of collected data in order to eliminate inaccurate control causing billets to be replaced of each other and fatal manufacturing defects with considerable financial losses. The solution uses a combination of machine vision and neural networks. Combined with automation, advanced data analytics and production management systems, it creates a unique concept of smart metallurgical operation.

Keywords: Metallurgy, machine vision, neural networks, control, applications
Scroll to Top