[1]尤波,李忠杰,黄玲.基于改进型BP神经网络的手部动作识别[J].智能系统学报,2018,13(05):848-854.[doi:10.11992/tis.201703018]
 YOU Bo,LI Zhongjie,HUANG Ling.Hand-motion recognition based on improved BP neural network[J].CAAI Transactions on Intelligent Systems,2018,13(05):848-854.[doi:10.11992/tis.201703018]
点击复制

基于改进型BP神经网络的手部动作识别(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第13卷
期数:
2018年05期
页码:
848-854
栏目:
出版日期:
2018-09-05

文章信息/Info

Title:
Hand-motion recognition based on improved BP neural network
作者:
尤波 李忠杰 黄玲
哈尔滨理工大学 自动化学院, 黑龙江 哈尔滨 150080
Author(s):
YOU Bo LI Zhongjie HUANG Ling
School of Automation, Harbin University of Science and Technology, Harbin 150080, China
关键词:
BP神经网络sEMG信号交叉熵手部动作识别特征提取正则化机器学习模式识别
Keywords:
BP neural networksEMG signalcross entropyhand-motion recognitionfeature extractionregularizationmachine learningpattern recognition
分类号:
TP391
DOI:
10.11992/tis.201703018
摘要:
对手部动作进行模式识别,首先将采集到的肌电信号进行降噪处理,选择时域分析法中的方差算法对采集信号进行特征提取。将特征信号进行归一化处理,实验发现普通BP神经网络分类器出现学习速率慢,泛化能力较差,不同动作识别准确率差别较大等问题。针对以上问题,提出了一种改进型BP神经网络,将神经网络输入数据进行人工升维处理,并对网络学习速率慢的原因进行理论推导,然后引入交叉熵代价函数并对其进行正则化处理,以提高网络的泛化能力以及网络的识别准确率。实验结果表明,改进型BP神经网络的学习速率、泛化能力以及动作分类的准确率均优于普通网络,识别准确率平均为94.34%。
Abstract:
To achieve accurate pattern recognition of hand motions, in this study, we first denoised collected electromyogram (EMG) signals, and then used a variance algorithm in the time domain to extract features from the collected signals. After normalizing the characteristic signal, in the experiment, we found that the general BP neural network classifier has a slow learning rate, poor generalization ability, and large differences in its accuracy of recognizing diverse motions. To address the above problems, we propose an improved BP neural network that processes its input data by artificially increasing the dimensions. It then theoretically determines the reason for the slow network learning rate, and introduces a cross-entropy cost function to regularize it, thereby improving the network’s generalization ability and increasing its reaction speed. Experimental results show that the improved BP neural network has a better learning speed, generalization ability, and accuracy in hand motion classification than the ordinary neural network, with an average recognition accuracy of 94.34%.

参考文献/References:

[1] 尤波, 杨文元. 基于小波理论的sEMG的处理方法[J]. 自动化技术与应用, 2012, 31(2):55-57, 61 YOU Bo, YANG Wenyuan. The processing method about sEMG based on wavelet theory[J]. Techniques of automation and applications, 2012, 31(2):55-57, 61
[2] YOU Bo, ZHOU Lina, HUANG Ling. The design of a recognition system for multi-function electrical rubber hand[J]. Transactions of the institute of measurement and control, 2016, 38(11):1298-1310.
[3] 谢平, 魏秀利, 杜义浩, 等. 基于自排序熵的表面肌电信号特征提取方法[J]. 模式识别与人工智能, 2014, 27(6):496-501 XIE Ping, WEI Xiuli, DU Yihao, et al. Feature extraction method of sEMG based on auto permutation entropy[J]. Pattern recognition and artificial intelligence, 2014, 27(6):496-501
[4] 邱青菊, 朱向阳. 基于双谱分析表面肌电信号特征的提取与模式识别[J]. 中国组织工程研究与临床康复, 2010, 14(4):645-648 QIU Qingju, ZHU Xiangyang. Feature extraction and pattern recognization of surface electromyography signal based on bispectrum analysis[J]. Journal of clinical rehabilitative tissue engineering research, 2010, 14(4):645-648
[5] 张启忠, 席旭刚, 罗志增. 基于非线性特征的表面肌电信号模式识别方法[J]. 电子与信息学报, 2013, 35(9):2054-2058 ZHANG Qizhong, XI Xugang, LUO Zhizeng. A pattern recognition method for surface electromyography based on nonlinear features[J]. Journal of electronics & information technology, 2013, 35(9):2054-2058
[6] NAIK G R, KUMAR D K, JAYADEVA J. Twin SVM for gesture classification using the surface electromyogram[J]. IEEE transactions on information technology in biomedicine-special section on affective and pervasive computing for healthcare, 2010, 14(2):301-308.
[7] CHEN Y, ZHOU Y, CHENG X. Surface EMG based upper limb motion recognition in real-time[J]. Journal of computational information systems, 2013, 9(23):9549-9556.
[8] 席旭刚, 李仲宁, 罗志增. 基于相关性分析和支持向量机的手部肌电信号动作识别[J]. 电子与信息学报, 2008, 30(10):2315-2319 XI Xugang, LI Zhongning, LUO Zhizeng. SEMG movement pattern recognition of hand based on correlation analysis and SVM[J]. Journal of electronics & information technology, 2008, 30(10):2315-2319
[9] LI Yang, TIAN Yantao, CHEN Wanzhong. sEMG pattern recognition based on GRNN and Adaboost[C]//International Conference on Electronics, Communications and Control. Ningbo, China, 2011:1661-1664.
[10] 张毅, 许新丽, 罗元. 基于在线SVM的自适应sEMG人机交互系统[J]. 华中科技大学学报:自然科学版, 2013, 41(4):75-79 ZHANG Yi, XU Xinli, LUO Yuan. Adaptive myoelectric human-machine interface systems using online support vector machines[J]. Journal of huazhong university of science & technology:nature science edition, 2013, 41(4):75-79
[11] KHEZRI M, JAHED M. A neuro-fuzzy inference system for sEMG-based identification of hand motion commands[J]. IEEE transactions on industrial electronics, 2011, 58(5):1952-1960.
[12] NAIK G R, AL-TIMEMY A H, NGUYEN H T. Transradial amputee gesture classification using an optimal number of sEMG sensors:an approach using ICA cluster ing[J]. IEEE transactions on neural systems and rehabilitation engineering, 2016, 24(8):837-846.
[13] KAINZ O, CYMBALÁK D, KARDOS S, et al. Low-cost assistive device for hand gesture recognition using sEMG[C]//Proceedings Volume 10011, First International Workshop on Pattern Recognition. Tokyo, Japan, 2016:100111B.
[14] 林海波, 王浩, 张毅, 等. 改进高斯核函数的人体姿态分析与识别[J]. 智能系统学报, 2015, 10(3):436-441 LIN Haibo, WANG Hao, ZHANG Yi,et al. Human postures recognition based on the improved Gauss kernel function[J]. CAAI transactions on intelligent systems, 2015, 10(3):436-441
[15] 杨帅. 基于AdaBoost算法的手部动作表面肌电信号分类方法研究[D]. 长春:吉林大学, 2015:35-57. YANG Shuai. Research of surface EMG classification of hand movement based on AdaBoost algorithm[D]. Changchun, China:Jilin University, 2015:35-57.
[16] 王尔申, 李兴凯, 庞涛. 基于BP神经网络的粒子滤波算法[J]. 智能系统学报, 2014, 9(6):709-713 WANG Ershen, LI Xingkai, PANG Tao. A particle filtering algorithm based on the BP neural network[J]. CAAI transactions on intelligent systems, 2014, 9(6):709-713
[17] 赵漫丹, 李东旭, 范才智, 等. 基于肌电信号层级分类的手部动作识别方法[J]. 北京生物医学工程, 2014, 33(5):490-496 ZHAO Mandan, LI Dongxu, FAN Caizhi, et al. A method of hand movement pattern recognition based on sEMG hierarchical classification[J]. Beijing biomedical engineering, 2014, 33(5):490-496
[18] 葛哲学, 孙志强. 神经网络理论与MATLAB R2007实现[M]. 北京, 电子工业出版社, 2007:108-112.
[19] 吕国豪, 罗四维, 黄雅平, 等. 基于卷积神经网络的正则化方法[J]. 计算机研究与发展, 2014, 51(9):1891-1900 LÜ Guohao, LUO Siwei, HUANG Yaping, et al. A novel regularization method based on convolution neural network[J]. Journal of computer research and development, 2014, 51(9):1891-1900
[20] 赵谦, 孟德宇, 徐宗本. L1/2正则化Logistic回归[J]. 模式识别与人工智能, 2012, 25(5):721-728 ZHAO Qian, MENG Deyu, XU Zongben. L1/2 regularized Logistic regression[J]. Pattern recognition and artificial intelligence, 2012, 25(5):721-728

相似文献/References:

[1]何世钊,杨宣访,陈晓娟.支持向量机与BP网络在火灾图像探测上的比较[J].智能系统学报,2011,6(04):339.
 HE Shizhao,YANG Xuanfang,CHEN Xiaojuan.Comparisons between a support vector machine and BP neural network for video image fire detection[J].CAAI Transactions on Intelligent Systems,2011,6(05):339.
[2]蒋鸿玲,邵秀丽.基于神经网络的僵尸网络检测[J].智能系统学报,2013,8(02):113.[doi:10.3969/j.issn.1673-4785.201210055]
 JIANG Hongling,SHAO Xiuli.Botnet detection algorithm based on neural network[J].CAAI Transactions on Intelligent Systems,2013,8(05):113.[doi:10.3969/j.issn.1673-4785.201210055]
[3]王尔申,李兴凯,庞涛.基于BP神经网络的粒子滤波算法[J].智能系统学报,2014,9(06):709.[doi:10.3969/j.issn.1673-4785.201310057]
 WANG Ershen,LI Xingkai,PANG Tao.A particle filtering algorithm based on the BP neural network[J].CAAI Transactions on Intelligent Systems,2014,9(05):709.[doi:10.3969/j.issn.1673-4785.201310057]
[4]王伟,周新志.ANFIS微波加热过程分段温度预测模型[J].智能系统学报,2016,11(1):61.[doi:10.11992/tis.201501028]
 WANG Wei,ZHOU Xinzhi.Temperature-sectioned prediction model for microwave heating process based on adaptive network-based fuzzy inference system[J].CAAI Transactions on Intelligent Systems,2016,11(05):61.[doi:10.11992/tis.201501028]
[5]雷森,史振威,石天阳,等.基于递归神经网络的风暴潮增水预测[J].智能系统学报,2017,12(05):640.[doi:10.11992/tis.201706015]
 LEI Sen,SHI Zhenwei,SHI Tianyang,et al.Prediction of storm surge based on recurrent neural network[J].CAAI Transactions on Intelligent Systems,2017,12(05):640.[doi:10.11992/tis.201706015]

备注/Memo

备注/Memo:
收稿日期:2017-03-13。
基金项目:国家“863”计划重大项目(2009AA043803).
作者简介:尤波,男,1962年生,教授,博士生导师,中国自动化学会机器人专业委员会委员,中国宇航学会智能机器人专业委员会委员,享受国务院特殊津贴,主要研究方向为智能机器人技术、模式识别、生产过程自动化。主持多项科研项目,并获得省科技进步二等奖3项和省科技进步三等奖1项。发表学术论文180余篇, 其中SCI收录20余篇,EI收录100余篇,获发明专利10余项;李忠杰,男,1991年生,硕士研究生,主要研究方向为图像处理、深度学习。发表学术论文2篇,EI收录1篇;黄玲,女,1975年生,教授,主要研究方向为T-S模糊系统的分析、控制器设计及在复杂系统处理中的应用。近年来主持多项国家、省、市级项目。发表学术论文30余篇,被SCI、EI收录20余篇。
通讯作者:尤波.E-mail:youbo@hrbust.edu.cn.
更新日期/Last Update: 2018-10-25