改进的K主曲线去噪性能研究

Journal of Systems Engineering and Electronics ›› 2009, Vol. 31 ›› Issue (5): 1047-1050.

改进的K主曲线去噪性能研究

陈慧贤, 吴彦华, 钟子发

1. 合肥电子工程学院信息系309室, 安徽, 合肥, 230037;
2. 安徽省电子制约技术重点实验室, 安徽, 合肥, 230037

收稿日期:2008-01-11 修回日期:2008-06-24 出版日期:2009-05-20 发布日期:2010-01-03
作者简介:陈慧贤(1981- ),女,博士研究生,主要研究方向为电磁监测,数据挖掘,聚类分析.E-mail:chhx916@163.com

Research on improved K principal curves denoise performance

CHEN Hui-xian, WU Yan-hua, ZHONG Zi-fa

1. Dept. of Information, Electronic Engineering Inst., Hefei 230037, China;
2. Key Lab. of Electronic Restriction, Anhui Province, Hefei 230037, China

Received:2008-01-11 Revised:2008-06-24 Online:2009-05-20 Published:2010-01-03

摘要/Abstract

摘要： 依据主曲线的思想,重点对主曲线算法在信号去噪上的应用进行了研究.首先介绍了K主曲线的理论和实际算法,论述了原始算法的局限性,提出了对原始主曲线的改进算法,并和原始算法在处理能力上进行了性能比较,最后将改进算法进行去噪实验.实验数据表明改进的K主曲线算法在一定信噪比限制下去噪性能优良,并且对噪声种类影响不敏感.

Abstract: According to the principal curves theory,the denoise technology is investigated by the principal curves algorithm.First,the paper introduces the theory and algorithm of K principal curves.Then several improvements are proposed to improve the limitations of the original algorithm.And the performances of both algorithms are compared.Finally,the improved algorithm is applied to a denoise experiment.Experiment results indicate that the performance of denoise is excellent under certain signal-to-noise ratio.And it is not sensitive to noise varieties.

中图分类号:

TP314

陈慧贤, 吴彦华, 钟子发. 改进的K主曲线去噪性能研究[J]. Journal of Systems Engineering and Electronics, 2009, 31(5): 1047-1050.

CHEN Hui-xian, WU Yan-hua, ZHONG Zi-fa. Research on improved K principal curves denoise performance[J]. Journal of Systems Engineering and Electronics, 2009, 31(5): 1047-1050.

参考文献

[1] Hastie T.Principal curve and surfaces[D].Laboratory for Computational Statistics,Stanford University,America,1984.
[2] Hastie T,Stuetzle W.Principal curves[J].Journal of the American Statistical Association,1988,84(406):502-516.
[3] Banfielf J D,Raftery A E.lee floe identification in satellite images using mathematical morphology and clustering about principal curves[J].Journal of American Statistical Association,1992,87(417):7-16.
[4] Kegl B.Principal curves:learning,design,and applications[D].Concordia University,Canada,1999.
[5] Kegl B,Krzyzak A.A polygonal line algorithm for constructing principal curves[J].Proceedings of Neural Information Processing Systems,Denver Colorado,USA,1999:501-507.
[6] Kegl B,Krzyzak A.Learning and design of principal curves[J].IEEE Trans.on Pattern Analysis and Machine Intelligence,2000,22(3):281-297.
[7] Reinhard K,Niranjan M.Subspaee models for speech transitions using principal curves[J].Proceedings of Institute of Acoustics,1998,20(6):53-60.
[8] Reinhard K,Niranjan M.Parametric subspaee modeling of speech transitions[J].Speech Communication,1999,27(1):19-24.
[9] Hermann T,Meinieke P,Ritter H.Principal curve sonification[J].Proceedings of International Conference on Auditory Display,Atlanda,USA,2000:81-86.
[10] De'ath G.Principal curves:a new technique for indirect and direct gradient analysis[J],Ecology,1999,80(7):2237-2253.
[11] Dong D,Mcavoy T J.Nonlinear principal component analysis based on principal curves and neural networks[J].Computers Chemistry Engineering,1995,20(1):65-87.
[12] Wilson D J H,Irwin G W,Lightbody G.RBF principal manifolds for process monitoring[J].IEEE Trans.of Neural Networks,1999,10(6):1424-1434.
[13] 杨诗琴,须文波,孙俊.用于多峰函数优化的改进小生境微粒群算法[J].计算机应用,2007,27(5):1191-1200.
[14] 陈国良,王煦法,庄镇泉,等.遗传算法及其应用[M].北京:人民邮电出版社,1996:111-116.