吉林大学学报（理学版）2018，Vol.56Issue(3)：669-675,7.DOI:10.13413/j.cnki.jdxblxb.2018.03.33
基于改进的稀疏降噪自编码网络的三维模型识别方法
3D Model Recognition Method Based on Improved Sparse Denoising Autoencoder Network
摘要
Abstract
Aiming at the problem of low accuracy of 3D model feature recognition in massive data mining ,we proposed an improved sparse denoising autoencoder neural network model .Firstly ,based on the improved sparse denoising autoencoder method ,we constructed a deep neural network model , then used the unsupervised pre-training method and restricted quasi New ton method to train the autoencoder neural network .Finally ,the softmax regression and the obtained features were used to train the final classifier . The results show that the method has well robustness to the feature information of the three-dimensional model with noise . Compared with the stack of autoencoder neural network and the self-learning neural network ,the method has better recognition rate .关键词
三维模型识别/稀疏降噪自编码/softmax分类器Key words
3D model recognition/sparse denoising autoencoder/softmax classifier分类
信息技术与安全科学引用本文复制引用
刘钢,王慧,王新颖..基于改进的稀疏降噪自编码网络的三维模型识别方法[J].吉林大学学报（理学版）,2018,56(3):669-675,7.基金项目
国家自然科学基金(批准号:61303132)、吉林省教育厅"十三五"科学技术研究项目(批准号:JJKH20170574KJ)、吉林省科技厅重大科技招标专项基金(批准号:20160203010GX)和吉林省发改委产业创新专项基金(批准号:20170505MA2). （批准号:61303132）
