Research of recognition algorithm for sheet metal marking
I. S. Nikitin, Yu. A. Usachev · 2017
The problem of automation of the product traceability processes within the technological process has now become of major importance. The most common traceability method is based on optical identification implemented by various markings application to the product surface. Such markings can contain a wide variety of information that allows one to determine the parameters of the raw materials, the manufacturer, the technological operations already performed and much more. In this case, the degree of readability of such markings can be used by a human operator as a recognition criterion. The paper presents an algorithm for optical character recognition to determine the marking of steel sheets used in the pipe and tube rolling industry. This algorithm was used for the development of the system of technical vision (STV) and implemented in the C# programming language with the use of specialized Emgu CV libraries. The main difficulties encountered in the recognition of marking elements as well as the methods for their elimination are considered in the paper.