Posterior probability based Multi-Classifier fusion in pedestrian detection

Jialu Zhao; Yan Chen; Xuanyi Zhuang; Yong Xu

doi:10.1007/978-3-319-01796-9_35

Posterior probability based Multi-Classifier fusion in pedestrian detection

Jialu Zhao
, Yan Chen
, Xuanyi Zhuang
, Yong Xu

Harbin Institute of Technology Shenzhen

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

6 Scopus citations

Abstract

This paper presents a novel method for pedestrian detection at measurement level. At feature extraction stage, we use Histogram of Oriented Gradient to describe the feature of pedestrian and non-pedestrian. To decrease the time cost, we reduce the dimension by using PCA. The base classifiers used in posterior probability based multi-classifier fusion are posterior probability based SVM, Naıve Bayesian and Minimum Distance Classifier, respectively. To estimate the accuracy of fusion result, stratified cross-validation is used. Experimental results on pedestrian databases prove the efficiency of this work.

Original language	English
Title of host publication	Genetic and Evolutionary Computing - 7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013, Proceedings
Editors	Jeng-Shyang Pan, Pavel Krömer, Václav Snášel
Publisher	Springer Verlag
Pages	323-329
Number of pages	7
ISBN (Electronic)	9783319017808
DOIs	https://doi.org/10.1007/978-3-319-01796-9_35
State	Published - 2014
Externally published	Yes
Event	7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013 - Prague, Czech Republic Duration: 25 Aug 2013 → 27 Aug 2013

Publication series

Name	Advances in Intelligent Systems and Computing
Volume	238
ISSN (Print)	2194-5357

Conference

Conference	7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013
Country/Territory	Czech Republic
City	Prague
Period	25/08/13 → 27/08/13

Keywords

Multi-classifier fusion
Pedestrian detection
Posterior probability
Stratified cross-validation

Access to Document

10.1007/978-3-319-01796-9_35

Cite this

Zhao, J., Chen, Y., Zhuang, X., & Xu, Y. (2014). Posterior probability based Multi-Classifier fusion in pedestrian detection. In J.-S. Pan, P. Krömer, & V. Snášel (Eds.), Genetic and Evolutionary Computing - 7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013, Proceedings (pp. 323-329). (Advances in Intelligent Systems and Computing; Vol. 238). Springer Verlag. https://doi.org/10.1007/978-3-319-01796-9_35

Zhao, Jialu ; Chen, Yan ; Zhuang, Xuanyi et al. / Posterior probability based Multi-Classifier fusion in pedestrian detection. Genetic and Evolutionary Computing - 7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013, Proceedings. editor / Jeng-Shyang Pan ; Pavel Krömer ; Václav Snášel. Springer Verlag, 2014. pp. 323-329 (Advances in Intelligent Systems and Computing).

@inproceedings{89f0b42f4b244cc7b3773189a82ec2ea,

title = "Posterior probability based Multi-Classifier fusion in pedestrian detection",

abstract = "This paper presents a novel method for pedestrian detection at measurement level. At feature extraction stage, we use Histogram of Oriented Gradient to describe the feature of pedestrian and non-pedestrian. To decrease the time cost, we reduce the dimension by using PCA. The base classifiers used in posterior probability based multi-classifier fusion are posterior probability based SVM, Naıve Bayesian and Minimum Distance Classifier, respectively. To estimate the accuracy of fusion result, stratified cross-validation is used. Experimental results on pedestrian databases prove the efficiency of this work.",

keywords = "Multi-classifier fusion, Pedestrian detection, Posterior probability, Stratified cross-validation",

author = "Jialu Zhao and Yan Chen and Xuanyi Zhuang and Yong Xu",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2014.; 7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013 ; Conference date: 25-08-2013 Through 27-08-2013",

year = "2014",

doi = "10.1007/978-3-319-01796-9\_35",

language = "英语",

series = "Advances in Intelligent Systems and Computing",

publisher = "Springer Verlag",

pages = "323--329",

editor = "Jeng-Shyang Pan and Pavel Kr{\"o}mer and V{\'a}clav Sn{\'a}{\v s}el",

booktitle = "Genetic and Evolutionary Computing - 7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013, Proceedings",

address = "德国",

}

Zhao, J, Chen, Y, Zhuang, X & Xu, Y 2014, Posterior probability based Multi-Classifier fusion in pedestrian detection. in J-S Pan, P Krömer & V Snášel (eds), Genetic and Evolutionary Computing - 7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013, Proceedings. Advances in Intelligent Systems and Computing, vol. 238, Springer Verlag, pp. 323-329, 7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013, Prague, Czech Republic, 25/08/13. https://doi.org/10.1007/978-3-319-01796-9_35

Posterior probability based Multi-Classifier fusion in pedestrian detection. / Zhao, Jialu; Chen, Yan; Zhuang, Xuanyi et al.
Genetic and Evolutionary Computing - 7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013, Proceedings. ed. / Jeng-Shyang Pan; Pavel Krömer; Václav Snášel. Springer Verlag, 2014. p. 323-329 (Advances in Intelligent Systems and Computing; Vol. 238).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Posterior probability based Multi-Classifier fusion in pedestrian detection

AU - Zhao, Jialu

AU - Chen, Yan

AU - Zhuang, Xuanyi

AU - Xu, Yong

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2014.

PY - 2014

Y1 - 2014

N2 - This paper presents a novel method for pedestrian detection at measurement level. At feature extraction stage, we use Histogram of Oriented Gradient to describe the feature of pedestrian and non-pedestrian. To decrease the time cost, we reduce the dimension by using PCA. The base classifiers used in posterior probability based multi-classifier fusion are posterior probability based SVM, Naıve Bayesian and Minimum Distance Classifier, respectively. To estimate the accuracy of fusion result, stratified cross-validation is used. Experimental results on pedestrian databases prove the efficiency of this work.

AB - This paper presents a novel method for pedestrian detection at measurement level. At feature extraction stage, we use Histogram of Oriented Gradient to describe the feature of pedestrian and non-pedestrian. To decrease the time cost, we reduce the dimension by using PCA. The base classifiers used in posterior probability based multi-classifier fusion are posterior probability based SVM, Naıve Bayesian and Minimum Distance Classifier, respectively. To estimate the accuracy of fusion result, stratified cross-validation is used. Experimental results on pedestrian databases prove the efficiency of this work.

KW - Multi-classifier fusion

KW - Pedestrian detection

KW - Posterior probability

KW - Stratified cross-validation

UR - https://www.scopus.com/pages/publications/84902651889

U2 - 10.1007/978-3-319-01796-9_35

DO - 10.1007/978-3-319-01796-9_35

M3 - 会议稿件

AN - SCOPUS:84902651889

T3 - Advances in Intelligent Systems and Computing

SP - 323

EP - 329

BT - Genetic and Evolutionary Computing - 7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013, Proceedings

A2 - Pan, Jeng-Shyang

A2 - Krömer, Pavel

A2 - Snášel, Václav

PB - Springer Verlag

T2 - 7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013

Y2 - 25 August 2013 through 27 August 2013

ER -

Zhao J, Chen Y, Zhuang X, Xu Y. Posterior probability based Multi-Classifier fusion in pedestrian detection. In Pan JS, Krömer P, Snášel V, editors, Genetic and Evolutionary Computing - 7th International Conference on Genetic and Evolutionary Computing, ICGEC 2013, Proceedings. Springer Verlag. 2014. p. 323-329. (Advances in Intelligent Systems and Computing). doi: 10.1007/978-3-319-01796-9_35

Posterior probability based Multi-Classifier fusion in pedestrian detection

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this