| 摘 要: 本文针对五类简单几何体三维模型，设计了完整的卷积神经网络模型以实现自动分类识别。首先本文运 用CATIA二次开发技术完成了模型的生产并收集了原始数据，其次运用图像处理的理论对原始数据进行了预处理，然 后运用卷积神经网络理论完成了本文的卷积神经网络模型设计，最后进行了实验分析并验证了本文模型的可行性。 |
| 关键词: 卷积神经网络；识别与分类；简单几何体；三维模型 |
|
中图分类号: TP391.41
文献标识码: A
|
|
| Research on Automatic Classification and Recognition of Simple Geometric 3D Models Based on Convolutional Neural Network |
|
DANG Yingcong,CHEN Jinjie
|
( University of Shanghai for Science and Technology, Shanghai 200093, China)
|
| Abstract: In this paper,a complete convolution neural network model is designed for automatic classification and recognition of five kinds of simple geometric three-dimensional models.Firstly,the CATIA secondary development technology is used to complete the production of the model and collect the original data.Secondly,the theory of image processing is used to preprocess the original data.Thirdly,the convolutional neural network model is designed by using the convolutional neural network theory.Finally,the feasibility of the model is analyzed and verified through experiments. |
| Keywords: convolutional neural network;classification and recognition;simple geometric;3D models |
