Implementation of an optimized distance function for retrieval and similarity comparison of non-rotational parts

Abstract

In the recent decades, advances in design of 3D solid models have been obtained due to the novel software and computing developments in addition to innovative design possibilities by using Computer Aid Design (CAD) in new product design. One of the superior methods to optimize the design process of 3D shapes is reusing of existing knowledge to assist in the design of new parts. In this concept, a classification system makes use of this huge amount of existing knowledge available and possible for the design as well as for the manufacturing. The advantage of using these knowledge results in reducing time and cost in the design process of a part. To achieve this goal, the philosophy of Group Technology that aims to group and classify parts has been applied. As a well-known approach of Group Technology, the Opitz Code System has been selected for this project which classifies manufacturing features of rotational and non-rotational solid models generating a shape signature. This Thesis aims to analyze, develop and realize the non-rotational section of the Opitz Code. To recognize and extract the features of non-rotational parts, IGES format has been implemented. This format identifies the boundary representation of a shape and generates a file with all their values. For the similarity comparison, a new distance function to compare Opitz codes has been implemented. To conclude this research, a Graphic User Interface (GUI) has been implemented to manage properly the full developed system including the database.