Research On CAD Component Retrieval Method Based On Model Similarity

With the rapid development of industry and the extensive use of industrial products, the retrieval technology of three-dimensional CAD model is widely used in our daily life. In the era of rapid development of the retrieval technology, we should learn how to efficiently use CAD model retrieval methods and reuse those existing CAD models. It can improve the development of economy and industry. It has become a hot issue to retrieve 3D models correctly in field of CAD model retrieval. Because the size of the CAD model database is relatively large, the retrieval complexity is increasing. At present, the existing model retrieval method can not solve the users' requirements in various fields, which brings inconvenience to users.This paper studies CAD model retrieval methods based on greedy strategy and bipartite graph. These methods have good flexibility and scalability. And they are able to cope with the development of the CAD model retrieval methods. They have become the mainstream CAD model retrieval methods. The research content of this paper is mainly divided into the following parts:Firstly, this paper introduces the research status of three-dimensional CAD model retrieval at home and abroad, and the retrieval systems of the authoritative CAD model are described. The various classification methods and characteristics of CAD model retrieval are studied. Those possible problems and solutions in the process of CAD model retrieval are given.Secondly, the feature extraction of the model is reseached. In the adjacency graph, similarities between faces are calculated by using the number of the edges. So, the similarity calculation between the models is converted to the similarity calculation between the faces. At the same time, the CAD model retrieval methods based on greedy strategy and bipartite graph are researched. This paper focuses on the establishment process of face matching matrix and the procedures of these two matching algorithms. The retrieval of CAD model is achieved according to the two methods.Finally, experimental data of these two methods are obtained. The effects of greedy strategy and bipartite graph on CAD model retrieval are researched. Although the research is not perfect, it can also provide a solution for CAD model retrieval.