Visual identification of industrial components using part family classification coding system

A novel method of identifying industrial components based on group technology concept and computer vision is proposed in this paper. The vision system is used to extract the features of the component. A software has been developed to generate the group technology (GT) classification code of that component from the vision data. The classification code generated from the vision data is then compared with the GT code of the component in the CAD database. If the matching occurs, the component is identified and the geometrical detail of the component is extracted from the database. The methodology algorithm and experimental results are presented in the paper.