Machine vision integration and information support: methods, models and tools

Abstract

In recent years, machine vision systems have been introduced into the electronics manufacturing industry for control and inspection applications. For some applications, great benefit has been derived, but serious limitations affecting flexibility and the system life cycle have been exposed. This paper proposes a model for building integratable vision machines capable of deriving benefit from support information, providing facilities to ease change and improve flexibilily. A proof of concept implementation is discussed using information storage based on both the relational and the object-oriented models. The paper presupposes the need for an integration infrastructure such as that specified by CIM-OSA or as implemented at Loughborough University of Technology by the Systems Integration Group as the CIM-BIOSYS (Computer Integrated Manufacturing Building Integrated Open Systems) integration infrastructure.