本文已被：浏览 1838次 下载 3524次
Received:March 20, 2019 Revised:April 17, 2019
Received:March 20, 2019 Revised:April 17, 2019
中文摘要: 为了提高脉搏波识别的准确率，提出改进的深度融合神经网络MIRNet2.首先，经过主波提取、划分周期和制作hdf5数据集等，获得Caffe可处理的数据集.其次，提出由Inception模块和残差模块构成的融合网络Inception-ResNet （IRNet），包含IRNet1、IRNet2和IRNet3.在此基础上，改进Inception模块、残差模块和池化模块，构造Modified Inception-ResNet （MIRNet），包含MIRNet1和MIRNet2.与本文其它神经网络相比，MIRNet2的分类性能最好，特异性、灵敏度和准确率分别达到87.85%、88.05%和87.84%，参数量和运算量也少于IRNet3.
Abstract:To improve the accuracy of pulse wave recognition, MIRNet2 is proposed, which is a kind of modified deep hybrid neural networks. Firstly, processable data sets of Caffe are obtained by main pulse extraction, segmenting cycle and making hdf5 data sets. Secondly, deep hybrid neural networks are designed. Inception-ResNet (IRNet) is consisted of inception modules and residual modules, containing IRNet1, IRNet2 and IRNet3. Subsequently, Modified Inception-ResNet (MIRNet) composed of modified Inception modules, residual modules and pooling modules (or reduction modules) is proposed, including MIRNet1 and MIRNet2. Compared with other neural networks in the study, MIRNet2 is the best one, with the specificity of 87.85%, the sensitivity of 88.05% and the accuracy of 87.84%, respectively. In addition, parameters and operations of MIRNet2 are also less than that of IRNet3.
keywords: pulse wave recognition convolutional neural networks Google inception net residual neural networks deep learning
文章编号： 中图分类号： 文献标志码：
基金项目:中国科学院苏州纳米技术与纳米仿生研究所资助项目（Y6AAJ11001）
引用文本：
张选,胡晓娟.基于GoogLeNet和ResNet的深度融合神经网络在脉搏波识别中的应用.计算机系统应用,2019,28(10):15-26
ZHANG Xuan,HU Xiao-Juan.Pulse Wave Recognition Using Deep Hybrid Neural Networks Based on GoogLeNet and ResNet.COMPUTER SYSTEMS APPLICATIONS,2019,28(10):15-26
张选,胡晓娟.基于GoogLeNet和ResNet的深度融合神经网络在脉搏波识别中的应用.计算机系统应用,2019,28(10):15-26
ZHANG Xuan,HU Xiao-Juan.Pulse Wave Recognition Using Deep Hybrid Neural Networks Based on GoogLeNet and ResNet.COMPUTER SYSTEMS APPLICATIONS,2019,28(10):15-26
