Computer vision based industrial process monitoring

OData support
Supervisor:
Dr. Csorba Kristóf
Department of Automation and Applied Informatics

In recent years, more and more factories have been started using intelligent hardware and software to meet the needs of economy. However, the increasing production speed may also affect the quality of the product. That is the reason why we should integrate real-time, automated quality control system into the production chain.

Systems that are able to monitor the products on the production line, and then take a decision about these products, based on the observation, are the product control systems. Nowadays, computers have high computational capacity to provide real-time information about the captured pictures. The so called machine vision, which is an active research area of information technology, can be used for this purpose. Not only high computational capacity computers are needed, but also high-resolution cameras for effective machine vision.

The goal of my thesis is to design and implement a framework based on machine vision which can control the production line. In my thesis, two production lines will be examined on which matches and boxes are made. First, I will present the mechanism of machine vision to design the system, then I will survey the main functions of a widely used open source computer vision library and I will design algorithms required for my framework. Finally, I will create a test environment in order to check the developed algorithms.

Downloads

Please sign in to download the files of this thesis.