In the age of big data, using abundant amounts of information, a model of a specific system can be created based on the analysis of its data. This results in a higher efficiency as the usage of the system can be improved compared to the conventional strategies without creating a complex physical model of it. We are developing a data-driven model for a Syringe Production Line. One of the production steps is the so-called screen printing, where an image is printed on top of the surface of the syringe. After a certain period of time, components must be replaced to ensure the quality of the printed image. In this thesis, a neural network will be developed to accommodate high-resolution inspection image which then used to analyze the quality of the printed image.
Basic in machine learning, deep learning and computer vision. C++ and python programming skill.
Contact person: Yeremia G. Adhisantoso