Monitoring Steel Heating Processes Using Infrared Thermography and Deep Learning-Based Semantic Segmentation.

Autor:	Morales-Cervantes, Antony, Chávez-Campos, Gerardo Marx, Vergara-Hernández, Héctor Javier, Flores, Juan J., Guevara, Edgar
Předmět:	THERMOGRAPHY PROCESS control systems DEEP learning CONVOLUTIONAL neural networks STEEL INDUCTION heating IMAGE processing
Zdroj:	JOM: The Journal of The Minerals, Metals & Materials Society (TMS); Jan2024, Vol. 76 Issue 1, p114-119, 6p
Abstrakt:	The monitoring of the steel heating process is crucial for ensuring the quality and consistency of the final product as well as for optimizing the manufacturing process and reducing costs. This study explored deep learning for the semantic segmentation task of identifying steel heated to high temperatures using infrared thermography. The 1045 steel specimens were heated to 1000°C, and changes in the material were observed using infrared thermography. A SegNet architecture based on convolutional neural networks and used to identify steel in thermograms, was presented for image processing. The region of interest for steel throughout its heating process was established using thermography and image processing described in this article. Thermal patterns in steel heated to high temperatures were accurately identified by the proposed network, which achieved a global accuracy of 91.16%. The use of deep learning applied with infrared thermography has the potential to control and monitor industrial processes without risking operator safety. [ABSTRACT FROM AUTHOR]
Databáze:	Complementary Index
Externí odkaz:	Zobrazit plný text záznamu Full text from SpringerLink