Title :
3D feature recognition base on CNN
Author :
Pin, Tao ; Bo, Zhang ; Zhen, Ye
Author_Institution :
State Key Lab. of Intelligent Technol. & Syst., Tsinghua Univ., Beijing, China
Abstract :
A 3D model feature recognition is an important problem in CIMS, and act as the bridge between CAD and CAM. With the topologic and complicated data of the 3D model, it is very difficult to find the feature. A constructive neural network is applied to the problem in the paper. The algorithm can recognize eight different simple features from each other, and can deal with some cross-feature. The result obtained indicates that the constructive neural network has significant ability in solving the problem of the feature recognition.
Keywords :
CAD/CAM; computer integrated manufacturing; image recognition; neural nets; 3D model feature recognition; CAD; CAM; CIMS; constructive neural network; Automation; Bridges; CADCAM; Cellular neural networks; Computer aided manufacturing; Computer integrated manufacturing; Intelligent control; Intelligent systems; Laboratories; Neural networks;
Conference_Titel :
Intelligent Control and Automation, 2002. Proceedings of the 4th World Congress on
Print_ISBN :
0-7803-7268-9
DOI :
10.1109/WCICA.2002.1020088
