基于主成分分析进行特征融合的心拍分类

王迪; 司玉娟; 刘通; 郭永丛; 刘书文

doi:10.3788/OPE.20152313.0453

您当前的位置：

首页 >

文章列表页 >

基于主成分分析进行特征融合的心拍分类

更新时间：2020-08-13

- 基于主成分分析进行特征融合的心拍分类
- Heart beat classification based on feature fusion by principle component analysis
- 光学精密工程 2015年23卷第10z期页码：453-458
- 作者机构：
  
  1. 吉林大学通信工程学院,吉林长春,中国,130012
  2. 吉林大学珠海学院,广东珠海,519041
  3. 北京大学软件与微电子学院北京,100871
- 作者简介：
- 基金信息：
  
  吉林省重点科技攻关项目(No.20150204039GX)();吉林省长春市重大科技攻关专项资助项目(No.14KG064)();广东省科技计划资助项目(No.2013
- DOI：10.3788/OPE.20152313.0453
  中图分类号： R540.4+1;TP391.4
- 收稿日期：2015-04-15，
  
  修回日期：2015-05-07，
  
  纸质出版日期：2015-11-14
- 稿件说明：
移动端阅览
王迪, 司玉娟, 刘通等. 基于主成分分析进行特征融合的心拍分类[J]. 光学精密工程, 2015,23(10z): 453-458

WANG Di, SI Yu-juan, LIU Tong etc. Heart beat classification based on feature fusion by principle component analysis[J]. Editorial Office of Optics and Precision Engineering, 2015,23(10z): 453-458
王迪, 司玉娟, 刘通等. 基于主成分分析进行特征融合的心拍分类[J]. 光学精密工程, 2015,23(10z): 453-458 DOI： 10.3788/OPE.20152313.0453.

WANG Di, SI Yu-juan, LIU Tong etc. Heart beat classification based on feature fusion by principle component analysis[J]. Editorial Office of Optics and Precision Engineering, 2015,23(10z): 453-458 DOI： 10.3788/OPE.20152313.0453.

摘要

针对心拍分类中单一特征分类准确率低而组合特征存在特征冗余和维度过高的问题

提出了一种基于主成分分析(PCA)进行特征融合的心拍分类算法。该算法在对单一特征进行规范化后

将多种特征组合为一个具有大量信息的高维特征;之后引入PCA算法去除冗余成分

得到低维融合特征;最后

利用支持向量机(SVM)为分类器完成心拍分类。以心拍时域特征、离散小波特征、离散傅里叶特征为例

在MIT-BIH数据库上进行了实验。实验结果表明

融合特征维度为100时

分类准确率可达97.389%。与单一特征相比

融合特征提高了分类准确率;与组合特征相比

融合特征降低了特征维度。

Abstract

For lower accuracy by a single feature and too high dimensionality by a combined feature in heartbeat classification

a new heartbeat classification algorithm was proposed based on the feature fusion by using Principle Component Analysis(PCA). With proposed algorithm

each single feature was normalized

then all kinds of features were combined together to a new one with high dimensionality and more information. In order to reduce dimensionality of combined features

the PCA was employed to remove redundant components. Finally

the Support Vector Machine(SVM) was used as a classifier to classify different heartbeats. By taking time domain feature

Discrete Wavelet Trans form(DWT) feature and Discrete Fourier Transform(DFT) feature as examples

the experiments were performed in MIT-BIH database. This study was compared with the new feature mentioned above

three single features and combined feature without Principle Component Analysis. The experiments results indicate that the accuracy of the new features is 97.389% when its dimension is 100. The new fusion feature has lower dimension than combined feature and higher classification accuracy than single feature.

关键词

Keywords

references

PHILIP C,MARIA O'D,RICHARD B. Automatic classification of heartbeart using ECG morphology and heartbeat interval features[J].IEEE Transactions on Biomedical Engineering,2004(51):1196-1206.

HARI M R,ANURAG T,SHAILJA S. ECG signal processing for abnormalities detection using multi-resolution wavelet transform and artificial neural network classifier[J].Measurement,2013(46):3238-3246.

林海明,杜子芳. 主成分分析综合评价应该注意的问题[J]. 统计研究, 2013,30(8):25-31. LIN H M,DU Z F.Some problems in comprehensive evaluation in the principal component analysis[J].Statistical Research,2013,30(8):25-31.(in Chinese)

唐红,郑文斌,李宪霞. 主成分分析在光全散射特征波长选择中的应用[J]. 光学精密工程,2010,18(8):1691-1698. TANG H,ZHENG W B,LI X X.Application of principal component analysis to selection of characteristic wavelengths with total light scattering[J].Opt. Precision Eng.,2010,18(8):1691-1698.(in Chinese).

王丽荣,王建蕾. 基于主成分分析的唇部轮廓建模[J]. 光学精密工程,2012,20(12):2768-2772. WANG L R, WANG J L.Lip contour modeling based on PCA[J].Opt. Precision Eng.,2012,20(12):2768-2772.(in Chinese)

付小宁,张涛,万里. 基于多光谱分离的烟雾检测[J]. 光学精密工程,2013,21(11):2798-2802. FU X N,ZHANG T,WAN L.Smoke detection based on multispectral separation[J].Opt. Precision Eng.,2013,21(11):2798-2802.(in Chinese)

范增飞. 基于频域特征的心电图分类研究[D].天津:天津大学,2006. FAN Z F.ECG classification based on frequency domain features[D].Tianjin:Tianjin University,2006.(in Chinese)

LIU T,SI Y J,WEN D W,et al.. Vector quantization for ECG beats classification[C]. 17th IEEE International Conference on Compututational Science and Engineering,Chengdu, Computational Science and Engineering(CSE),2014:13-20.

YU S N, CHEN Y H. Electrocardiogram beat classification based on wavelet transformation and probabilistic neural network[J].Pattern Recognition Letters,2007,28(10):1142-1150.

KHORRAMI H, MOAVENIAN M. A comparative study of dwt, cwt and dct transformations in ecg arrhythmias classification[J].Expert Systerms with Applications,2010,37(8):5751-5757.

蔡哲元,余建国,李先鹏,等.基于核空间距离测度的特征选择[J]. 模式识别与人工智能,2010,23(2):235-240. CAI Z Y,YU J G,LI X P,et al..Feature selection algorithm based on kernel distance measure[J].Pattern Recognition and Artificial Intelligence,2010,23(2):235-240.(in Chinese)

韩景梅. 支持向量机决策树算法研究及其应用[D].上海:上海交通大学,2007. HAN J M. Research on support vector machine-decision tree arithmetic and its application[D].Shanghai:Shanghai Jiao Tong University,2007.(in Chinese)

WANG D K,LI D W, LIN Y. A new method of face recognition with data field and PCA[C].2013 IEEE International Conference on Granular Computing(Grc),Beijing, 2013:320-325.

ZHAI M H, SHI F Y,Duncan D.Convariance-Based PCA for multi-size data[C].2014 22nd International Conference on Pattern Recognition, Stockholm, 2014:1603-1608.

LIM S T,YAP D F W,MANAP N A. Medical image compression using block-based PCA algorithm[C].2014 International Conference on Computer, Communications, and Control Technology, Langkawei, Kedah, 2014:171-175.

浏览量

369

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

强化边缘特征提取的双分支融合真实图像去雾

基于混合时空卷积的轻量级视频超分辨率重建

联合自注意力和分支采样的无人机图像目标检测

多维注意力机制与选择性特征融合的图像超分辨率重建

动态特征优化机制下的跨尺度红外行人检测