• Computer Vision
• defects checking machine
• glass tubes
• Industrial Applications

The thesis is part of a larger development project which involves the design of an inspection machine for automatic quality control of the glass tube in a company that provides the primary packaging for the pharmaceutical industry : vials, test tubes and syringes. After a preliminary study of the existing system at the time of commencement of employment, it has passed the evaluation of its problems and the proposed solutions for the same in order to make the system more robust: in some cases it has paid the adjustment of existing algorithms (tuning) and in others it is designed and manufactured components of the system from scratch. The thesis is mainly on three fronts: improving the images obtained by the machine vision system using digital filters realized ad hoc, in order to make more robust recognition of any faults in the product glass tube; control of the operating conditions of the inspection machine by reading data from temperature sensors on the same, to prevent damage to the electronic components of the machine, which operates in the extreme conditions of a glass foundry; and finally sampling and monitoring of data relating to the production in order to control certain critical parameters, such as a tube bending problem.