[1]刘忠宝,王士同.从Parzen窗核密度估计到特征提取方法:新的研究视角[J].智能系统学报,2012,7(06):471-480.
 LIU Zhongbao,WANG Shitong.From Parzen window estimation to feature extraction: a new perspective[J].CAAI Transactions on Intelligent Systems,2012,7(06):471-480.
点击复制

从Parzen窗核密度估计到特征提取方法:新的研究视角(/HTML)
分享到:

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

卷:
第7卷
期数:
2012年06期
页码:
471-480
栏目:
出版日期:
2012-12-25

文章信息/Info

Title:
From Parzen window estimation to feature extraction: a new perspective
文章编号:
1673-4785(2012)06-0471-10
作者:
刘忠宝12王士同1
1. 江南大学 数字媒体学院,江苏 无锡 214122;
2. 中北大学 电子与计算机科学技术学院,山西 太原 030051
Author(s):
LIU Zhongbao12 WANG Shitong1
1. School of Digital Media, Jiangnan University, Wuxi 214122, China;
2. School of Electronics and Computer Science Technology, North University of China, Taiyuan 030051, China
关键词:
特征提取Parzen窗密度估计数据分布特征新视角
Keywords:
feature extraction Parzen window density estimation data distribution characteristics new perspective
分类号:
TP392
文献标志码:
A
摘要:
当前主流特征提取方法大致有2种研究思路:1)从高维数据的几何性质出发,根据某种寻优准则得到基于原始空间特征的一组特征数更少的新特征;2)从降维误差角度出发,保证降维前后数据所呈现的某种偏差达到最小.试图从降维过程中数据分布特征的变化入手,基于广泛使用的Parzen窗核密度估计方法,来审视和揭示Parzen窗估计与典型特征提取方法LPP、LDA和PCA之间的关系,从而说明这些特征提取方法可统一在Parzen窗框架下进行研究,为特征提取方法的研究提供了一个新的视角.
Abstract:
Researches on current feature extraction methods are mainly based on two ways. One originates from geometric properties of highdimensional datasets and attempt to extract fewer features from the original data space according to a certain criterion. The other originates from dimension reduction deviation and try to make the deviation between data before and after dimension reduction be as small as possible. However, there exists almost no study about them from the perspective of the scatter change of a dataset. Based on Parzen window density estimator, we thoroughly revisit the relevant feature extraction methods from a new perspective and the relationships between Parzen window and classical feature extraction methods,ie length of perpendiculars (LPP), linear discriminant analysis (LDA) and principal component analysis (PCA) are built in this paper. Therefore, these feature extraction methods can be researched in the same Parzen window, which provides a new perspective for the research of feature extraction.

参考文献/References:

[1]WANG Qinghua, ZHANG Youyun, LEI Cai, et al. Fault diagnosis for diesel value trains based on nonnegative matrix factorization and neural network ensemble[J]. Mechanical Systems and Signal Processing, 2009, 23(5): 16831695.
[2]CAMACHO J, PIC J, FERRER A. Data understanding with PCA: structural and variance information plots[J]. Chemometrics and Intelligent Laboratory Systems, 2010, 100(1): 4856.
[3]LIPOVETSKY S. PCA and SVD with nonnegative loadings [J]. Pattern Recognition, 2009, 42(1): 6876.
[4]RADULOVIC J, RANKOVIC V. Feedforward neural network and adaptive networkbased fuzzy inference system in study of power lines[J]. Expert Systems with Applications, 2010, 37(1): 165170.
[5]PETER N B, JOAO P H, DAVID J K. Eigenfaces vs Fisherfaces: recognition using class specific linear projection[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 1997, 19(7): 711720.
[6]HU Y H, HE S H. Integrated evaluation method[M]. Beijing: Scientific Press, 2000: 2731.
[7]ROWEIS S T, SAUL L K. Nonlinear dimensionality reduction by locally linear embedding[J]. Science, 2000(290): 23232326.
[8]BELKIN M, IYOGI P. Laplacian eigenmaps and spectral techniques for embedding and clustering[C]// Systems (NIPS). Cambridge: MIT NIYOGI P Press, 2001: 585591.〖JP〗
[9]BORGA M, KNUTSSON H. Canonical correlation analysis in early vision[C]//Proceedings of the 9th European Symposium on Artificial Neural Networks. Bruges, Belgium, 2001: 309314.
[10]HE X F, NIYOGI P. Locality preserving projections[C]// Advances in Neural Information Processing Systems (NIPS). Vancouver, Canada, 2003: 153160.
[11]皋军, 王士同, 邓赵红. 广义的势支撑特征选择GPSFM[J].计算机研究与发展, 2009, 46(1): 4151. 
GAO Jun, WANG Shitong, DENG Zhaohong. Generalized potential support vector machine for feature selection[J]. Computer Research and Development, 2009, 46(1): 4151.
[12]王晓明, 王士同. 广义的监督局部保留投影算法[J]. 电子与信息学报, 2009, 31(8): 18401845.
WANG Xiaoming, WANG Shitong. Generalized supervised locality preserving projection algorithm[J]. Electronics and Information, 2009, 31(8): 18401845.
[13]王超, 王士同. 有局部保持的最大间距准则特征提取方法[J]. 模式识别与人工智能, 2009, 22(6): 898902.
 WANG Chao, WANG Shitong. Maximum margin and locality preserving feature selection method[J]. Artificial Intelligence and Pattern Recognition, 2009, 22(6): 898902.
[14]ODIOWEI P P, YI Cao. Nonlinear dynamic process monitoring using canonical variate analysis and kernel density estimations[J]. IEEE Trans on Industrial Informatics, 2010, 6(1): 3645.
[15]吴葛铭,霍剑青,王晓蒲.一种基于加权Parzen窗的聚类算法[J]. 中国科学技术大学学报, 2002, 32(5): 546551. 
WU Geming, HUO Jianqing, WANG Xiaopu. Weighted Parzen window based clustering[J]. Journal of Chinease University of Sciences and Technologies, 2002, 32(5): 546551.
[16]HE X F, YAN S C, HU Y X, et al. Face recognition using Laplacianfaces[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 2005, 27(3): 328340.
[17]CAMACHO J, PIC J, FERRER A. Data understanding with PCA: structural and variance information plots[J]. Chemometrics and Intelligent Laboratory Systems, 2010, 100(1): 4856.
[18]边肇祺, 张学工. 模式识别[M]. 2版.北京清华大学出版社, 2000: 130. [19]GUO Q C, CHANG X J, CHU H X. Meanshift of variable window based on the Epanechnikov kernel[C]//Proceedings of IEEE Int Conf on Mechatronics and Automation. Piscataway, USA, 2007: 23142319.

相似文献/References:

[1]黄剑华,唐降龙,刘家锋,等.一种基于Homogeneity的文本检测新方法[J].智能系统学报,2007,2(01):69.
 HUANG Jian-hua,TANG Xiang-long,LIU Jia-feng,et al.A new method for text detection based on Homogeneity[J].CAAI Transactions on Intelligent Systems,2007,2(06):69.
[2]谭 营,朱元春.反垃圾电子邮件方法研究进展[J].智能系统学报,2010,5(03):189.
 TAN Ying,ZHU Yuan-chun.Advances in antispam techniques[J].CAAI Transactions on Intelligent Systems,2010,5(06):189.
[3]王斐,张育中,宁廷会,等.脑-机接口研究进展[J].智能系统学报,2011,6(03):189.
 WANG Fei,ZHANG Yuzhong,NING Tinghui,et al.Research progress in a braincomputer interface[J].CAAI Transactions on Intelligent Systems,2011,6(06):189.
[4]刘琚,孙建德.独立分量分析的图像/视频分析与应用[J].智能系统学报,2011,6(06):495.
 LIU Ju,SUN Jiande.Independent component analysisbased image/video analysis and applications[J].CAAI Transactions on Intelligent Systems,2011,6(06):495.
[5]谭营,王军.手指静脉身份识别技术最新进展[J].智能系统学报,2011,6(06):471.
 TAN Ying,WANG Jun.Recent advances in finger vein based biometric techniques[J].CAAI Transactions on Intelligent Systems,2011,6(06):471.
[6]吴家伟,严京旗,方志宏,等.基于图像显著性特征的铸坯表面缺陷检测[J].智能系统学报,2012,7(01):75.
 WU Jiawei,YAN Jingqi,FANG Zhihong,et al.Defect detection on a steel slab surface based on the characteristics of an image’s saliency region[J].CAAI Transactions on Intelligent Systems,2012,7(06):75.
[7]张毅,罗明伟,罗元.脑电信号的小波变换和样本熵特征提取方法[J].智能系统学报,2012,7(04):339.
 ZHANG Yi,LUO Mingwei,LUO Yuan.EEG feature extraction method based on wavelet transform and sample entropy[J].CAAI Transactions on Intelligent Systems,2012,7(06):339.
[8]闫晓波,王士同,郭慧玲.基于Parzen窗的高阶统计量特征降维方法[J].智能系统学报,2013,8(01):1.[doi:10.3969/j.issn.1673-4785.201210046]
 YAN Xiaobo,WANG Shitong,GUO Huiling.Feature reduction of high order statistics based on Parzen window[J].CAAI Transactions on Intelligent Systems,2013,8(06):1.[doi:10.3969/j.issn.1673-4785.201210046]
[9]孙倩茹,王文敏,刘宏.视频序列的人体运动描述方法综述[J].智能系统学报,2013,8(03):189.
 SUN Qianru,WANG Wenmin,LIU Hong.Study of human action representation in video sequences[J].CAAI Transactions on Intelligent Systems,2013,8(06):189.
[10]许可乐,唐涛,蒋咏梅.一种SAR图像稳健特征点提取方法[J].智能系统学报,2013,8(04):287.[doi:10.3969/j.issn.1673-4785.201304038]
 XU Kele,TANG Tao,JIANG Yongmei.A stable feature point extraction approach for SAR image registration[J].CAAI Transactions on Intelligent Systems,2013,8(06):287.[doi:10.3969/j.issn.1673-4785.201304038]

备注/Memo

备注/Memo:
收稿日期:2012-05-14.
网络出版日期:2012-11-16.
基金项目:国家自然科学基金资助项目(61170122, 61272210). 
通信作者:刘忠宝.
E-mail:liu_zhongbao@hotmail.com.
作者简介:
刘忠宝,男,1981年生,博士,主要研究方向为模式识别、机器学习,发表学术论文数十篇.
王士同,男,1964年生,教授,博士生导师,主要研究方向为人工智能与机器学习.
更新日期/Last Update: 2013-03-19