Usage of Real Time Machine Vision in Rolling Mill

This article deals with the issue of computer vision on a rolling mill. The main goal this is to describe designed and implemented algorithm for automatic identification character string billets allows conversion image information from front billet, which enters process, into characters, further used control technological process. purpose prevent input pieces being confused because different parameters process are set pieces. In solving task, it was necessary design optimal technical equipment capture, choose appropriate lighting, search text recognize individual symbols, insert them system. research methodology based empirical-quantitative principle, basis analysis experimentally obtained data (photographs billet faces) in real operating conditions leading their interpretation (transformation shape digital chain). first part briefly describes system point view technology hardware resources. next parts devoted identification—optical recognition strings characters chain using artificial intelligence. method optical neural networks basic eliminates ambiguities during processing. Successful implementation inspection will increase share operation automation lead ensuring steel according production plan. related trend digitization processes metallurgy also social sustainability processes, means elimination human errors management