Affiliation:
1. Guilin University of Electronic Technology
2. Guilin College of Aerospace Technology
Abstract
To accelerate the development of new product, one important method is to use the information of the existed products. In this paper, Fuzzy C-Means clustering method (FCM) is proposed to retrieve the desired model in the given model database. In the retrieval process, first, the shape distribution histograms of the model are established and transformed into the proper format of FCM. Then, the shape distribution histograms of the model are served as inputs of FCM and classified into several groups by this algorithm. Last, all the models that belong to the classification of the query model are returned as the retrieval results. To validate the proposed method, a case study is presented. The results show that the result of proposed method is better than that of Shape Distribution. Thus, this method is suitable for 3D mechanical models retrieval.
Publisher
Trans Tech Publications, Ltd.
Reference7 articles.
1. Michael. M. Kazhdan: Doctor Thesis of Princeton University, Princeton, U.S. (2004).
2. Iyer N, et al. Computer-Aided Design, 37(5): 509-30(2005).
3. Osada. R. et al: ACM Transaction on Graphics, Vol. 21 (2002), pp.807-832.
4. Meifa Huang and Hui Jing: Virtual and Physical Prototyping, Vol. 2 (2009), pp.105-112.
5. Subramaniam Jayanti. et al: Computer-Aided Design, Vol. 38(2006), pp.939-953.