[1]赵杰,丁吉,万遂人,等.全光纤安防系统模式识别混合编程的实现[J].东南大学学报(自然科学版),2011,41(1):41-46.[doi:10.3969/j.issn.1001-0505.2011.01.009]
 Zhao Jie,Ding Ji,Wan Suiren,et al.Implementation of pattern recognition in all fiber security monitoring system using mixed programming[J].Journal of Southeast University (Natural Science Edition),2011,41(1):41-46.[doi:10.3969/j.issn.1001-0505.2011.01.009]
点击复制

全光纤安防系统模式识别混合编程的实现()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
41
期数:
2011年第1期
页码:
41-46
栏目:
自动化
出版日期:
2011-01-20

文章信息/Info

Title:
Implementation of pattern recognition in all fiber security monitoring system using mixed programming
作者:
赵杰1丁吉1万遂人1孙小菡2
(1东南大学生物科学与医学工程学院,南京210096)(2东南大学电子科学与工程学院,南京210096)
Author(s):
Zhao Jie1Ding Ji1Wan Suiren1Sun Xiaohan2
(1School of Biological Science and Medical Engineering, Southeast University, Nanjing 210096, China)
(2School of Electronic Science and Engineering, Southeast University, Nanjing 210096, China)
关键词:
光纤扰动信号模式识别小波包分解BVC-RBF网络混合编程
Keywords:
fiber disturbance signal pattern recognition wavelet packet neural network mixed programming
分类号:
TP277
DOI:
10.3969/j.issn.1001-0505.2011.01.009
摘要:
介绍了宽域全光纤监控预警系统及光纤扰动信号模式识别算法.利用小波包分解对信号进行分解与重构,并提取各频段内信号分量能量谱作为特征向量; 应用综合考虑观测数据和确定性先验知识的BVC-RBF神经网络实现信号分类,提高了网络的泛化能力.利用Matlab引擎、动态链接库(DLL)、组件对象模型(COM)3种混合编程方法将Matlab开发的算法程序集成到VC编写的用户操作平台中.使用文本文件存取网络数据,克服了Matlab无法编译神经网络工具箱函数的缺陷,并从操作复杂度、执行时间、内存消耗等方面对3种方式进行对比,最终选定动态链接库作为系统应用方案.
Abstract:
An all fiber security monitoring system and pattern recognition algorithm on fiber disturbance signal is presented. As a method of analyzing and reconstructing signal wavelet packet is adopted to extract feature vector. Classification is realized by BVC-RBF (boundary value constraints-radial basis function) neural network, which improves generalization by considering deterministic prior knowledge besides observational data. Pattern recognition algorithm is developed with Matlab and integrated into the application developed with VC through three mixed programming methods: Matlab engine, dynamic link library (DLL), component objective model (COM). Network data are read and written through text file to overcome the limitation that Matlab can not compile functions in neural network toolbox. Operation complexity, running time and memory consumption of these methods are compared and dynamic link library is the final choice.

参考文献/References:

[1] Deng K L.Next generation fiber optic sensing and its application[C]//The 14th Annual Wireless and Optical Communications Conference.New Jersey,USA,2005:76-76.
[2] Okazaki Hideki,Nishiyama Michiko,Watanabe Kazuhiro.Specific vibration frequency detection based on hetero-core fiber optic sensing scheme for security environment monitoring[C]//IEEE International Workshop on Robotic and Sensors Environments Proceedings.Lecco,Italy,2009:53-57.
[3] Kim Dae-Hym,Feng Maria Q.A novel optical fiber accelerometer system for monitoring civil infrastructure[C]//Proceedings of the Fourth IEEE Conference on Sensors 2005.Irvine,CA,USA,2005:1107-1111.
[4] Xia Hu,South Richard B,Rutter Malcolm J.Distributed optical fiber sensing system using elliptical core fiber[C]//Conference on Lasers and Electro-Optics Europe.Amsterdam,The Netherlands,1994:330-331.
[5] 尹崇博.分布式光纤传感系统触发模式识别的研究[D].上海:复旦大学材料科学系,2008:28-32.
[6] 郭伟,顾耀娟.基于VC++和COM的复杂系统控制算法的研究[J].计算机应用,2009,31(3):27-31.
  Guo Wei,Gu Yaojuan.Research on complicated system control algorithm based on VC++and COM[J].Computer Application,2009,31(3):27-31.
[7] Zhu Quan,Fu Sheng,Li Jing.Application of wavelet package and neural network in ventilators fault warning[C]//Condition Monitoring and Diagnosis.Beijing,China,2008:1362-1364.
[8] Lyons W B,King D.A 3 sensor multipoint optical fiber water sensor utilizing artificial neural network pattern recognition[C]//Optical Fiber Sensors Conference Technical Digest.Portland,USA,2002,1:463-466.
[9] Han Shuang,Yang Yongping,Liu Yongqian.The comparison of BP network and RBF network in wind power prediction application[C]//Bio-inspired Computing:Theories and Applications.Zhengzhou,China,2007:173-176.
[10] Xia Hong,Chen Sheng.A new RBF neural network with boundary value constrains[J].IEEE Transactions on Systems,Man,and Cybernetics,2009,39(1):298-303.
[11] Fu Hudai,Wang Hua,Yang Wei.Application of mixed programming between Matlab and Visual C++on circles detection[C]//Intelligent Human-machine Systems and Cybernetics.Hangzhou,China,2009:53-57.
[12] 张德丰.Matlab与外部程序接口编程[M].北京:机械工业出版社,2009:116-120.
[13] 赵书兰.Matlab R2008接口技术程序设计实例教程[M].北京:化学工业出版社,2009:375-379.
[14] 陈超.精通Matlab 2008应用程序接口编程技术[M].北京:电子工业出版社,2009:286-290.

相似文献/References:

[1]夏良正,李晓兵,罗庆姚.一种新的基于投影的二维矩不变量快速算法[J].东南大学学报(自然科学版),1993,23(1):41.[doi:10.3969/j.issn.1001-0505.1993.01.006]
 Xia Liangzheng,Li Xiaobing,Luo Qingyao,et al.A New Fast Algorithm for Computation of Two-Dimensional Moment lnvariants Based on Image Projection[J].Journal of Southeast University (Natural Science Edition),1993,23(1):41.[doi:10.3969/j.issn.1001-0505.1993.01.006]
[2]吴乐南.用多样本联合判决提高雷达目标识别正确性[J].东南大学学报(自然科学版),1989,19(4):21.[doi:10.3969/j.issn.1001-0505.1989.04.004]
 Wu Lenan Department of Radio Engineering.Increasing the Correctness in Radar Target Recognition by Multisample Joint Discrimination[J].Journal of Southeast University (Natural Science Edition),1989,19(1):21.[doi:10.3969/j.issn.1001-0505.1989.04.004]
[3]黄仁,时修荣,沙勇,等.磨削烧伤在线辨识方法的研究[J].东南大学学报(自然科学版),1988,18(2):13.[doi:10.3969/j.issn.1001-0505.1988.02.002]
 Huang Ren Shi Xiurong Sha Yong Sun Xiaojun Su Guisheng (Department of Mechanical Engineering).Research on Approach of On-line Identification of the Grinding Burn[J].Journal of Southeast University (Natural Science Edition),1988,18(1):13.[doi:10.3969/j.issn.1001-0505.1988.02.002]

备注/Memo

备注/Memo:
作者简介:赵杰(1986—),男,硕士生;万遂人(联系人),男,博士,教授,博士生导师,suirenwan@hotmail.com.
引文格式: 赵杰,丁吉,万遂人,等.全光纤安防系统模式识别混合编程的实现[J].东南大学学报:自然科学版,2010,41(1):0041-46.[doi:10.3969/j.issn.1001-0505.2011.01.009]
更新日期/Last Update: 2011-01-20