Optimal feature selection for pedestrian detection based on logistic regression analysis

Jonghee Kim; Jonghwan Lee; Chungsu Lee; Eunsoo Park; Junmin Kim; Hakil Kim; Jaeeun Lee; Hoeri Jeong

doi:10.1109/SMC.2013.47

Optimal feature selection for pedestrian detection based on logistic regression analysis

Jonghee Kim, Jonghwan Lee, Chungsu Lee, Eunsoo Park, Junmin Kim, Hakil Kim, Jaeeun Lee, Hoeri Jeong

Inha University

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

16 Scopus citations

Abstract

This paper describes a pedestrian detection method using feature selection based on logistic regression analysis. As the parent features, Haar-like and Histograms of Oriented Gradients (HOG) features are used manually. For the statistical analysis, stepwise forward selection, backward elimination, and Least Absolute Shrinkage and Selection Operator (LASSO) methods are applied to our Logistic Regression Model for Pedestrian Detection (LRMPD). The experimental results shows that the average of 48.5% of a full model are selected for LRMPD and this classifier shows performance of up to 95% for detection rate with an approximately 10% false positive rate. Processing time for one test image is about 1.22ms.

Original language	English
Title of host publication	Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013
Pages	239-242
Number of pages	4
DOIs	https://doi.org/10.1109/SMC.2013.47
State	Published - 2013
Event	2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013 - Manchester, United Kingdom Duration: 13 Oct 2013 → 16 Oct 2013

Publication series

Name	Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013

Conference

Conference	2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013
Country/Territory	United Kingdom
City	Manchester
Period	13/10/13 → 16/10/13

Keywords

Feature selection
Logistic regression
Multifeature
Pedestrian detection

Access to Document

10.1109/SMC.2013.47

Cite this

Kim, J., Lee, J., Lee, C., Park, E., Kim, J., Kim, H., Lee, J., & Jeong, H. (2013). Optimal feature selection for pedestrian detection based on logistic regression analysis. In Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013 (pp. 239-242). Article 6721800 (Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013). https://doi.org/10.1109/SMC.2013.47

@inproceedings{81e31407d080497a87cb029f4f49ccb0,

title = "Optimal feature selection for pedestrian detection based on logistic regression analysis",

abstract = "This paper describes a pedestrian detection method using feature selection based on logistic regression analysis. As the parent features, Haar-like and Histograms of Oriented Gradients (HOG) features are used manually. For the statistical analysis, stepwise forward selection, backward elimination, and Least Absolute Shrinkage and Selection Operator (LASSO) methods are applied to our Logistic Regression Model for Pedestrian Detection (LRMPD). The experimental results shows that the average of 48.5\% of a full model are selected for LRMPD and this classifier shows performance of up to 95\% for detection rate with an approximately 10\% false positive rate. Processing time for one test image is about 1.22ms.",

keywords = "Feature selection, Logistic regression, Multifeature, Pedestrian detection",

author = "Jonghee Kim and Jonghwan Lee and Chungsu Lee and Eunsoo Park and Junmin Kim and Hakil Kim and Jaeeun Lee and Hoeri Jeong",

year = "2013",

doi = "10.1109/SMC.2013.47",

language = "English",

isbn = "9780769551548",

series = "Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013",

pages = "239--242",

booktitle = "Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013",

note = "2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013 ; Conference date: 13-10-2013 Through 16-10-2013",

}

Kim, J, Lee, J, Lee, C, Park, E, Kim, J, Kim, H, Lee, J & Jeong, H 2013, Optimal feature selection for pedestrian detection based on logistic regression analysis. in Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013., 6721800, Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013, pp. 239-242, 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013, Manchester, United Kingdom, 13/10/13. https://doi.org/10.1109/SMC.2013.47

Optimal feature selection for pedestrian detection based on logistic regression analysis. / Kim, Jonghee; Lee, Jonghwan; Lee, Chungsu et al.
Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013. 2013. p. 239-242 6721800 (Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013).

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

TY - GEN

T1 - Optimal feature selection for pedestrian detection based on logistic regression analysis

AU - Kim, Jonghee

AU - Lee, Jonghwan

AU - Lee, Chungsu

AU - Park, Eunsoo

AU - Kim, Junmin

AU - Kim, Hakil

AU - Lee, Jaeeun

AU - Jeong, Hoeri

PY - 2013

Y1 - 2013

N2 - This paper describes a pedestrian detection method using feature selection based on logistic regression analysis. As the parent features, Haar-like and Histograms of Oriented Gradients (HOG) features are used manually. For the statistical analysis, stepwise forward selection, backward elimination, and Least Absolute Shrinkage and Selection Operator (LASSO) methods are applied to our Logistic Regression Model for Pedestrian Detection (LRMPD). The experimental results shows that the average of 48.5% of a full model are selected for LRMPD and this classifier shows performance of up to 95% for detection rate with an approximately 10% false positive rate. Processing time for one test image is about 1.22ms.

AB - This paper describes a pedestrian detection method using feature selection based on logistic regression analysis. As the parent features, Haar-like and Histograms of Oriented Gradients (HOG) features are used manually. For the statistical analysis, stepwise forward selection, backward elimination, and Least Absolute Shrinkage and Selection Operator (LASSO) methods are applied to our Logistic Regression Model for Pedestrian Detection (LRMPD). The experimental results shows that the average of 48.5% of a full model are selected for LRMPD and this classifier shows performance of up to 95% for detection rate with an approximately 10% false positive rate. Processing time for one test image is about 1.22ms.

KW - Feature selection

KW - Logistic regression

KW - Multifeature

KW - Pedestrian detection

UR - https://www.scopus.com/pages/publications/84893611800

U2 - 10.1109/SMC.2013.47

DO - 10.1109/SMC.2013.47

M3 - Conference contribution

AN - SCOPUS:84893611800

SN - 9780769551548

T3 - Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013

SP - 239

EP - 242

BT - Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013

T2 - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013

Y2 - 13 October 2013 through 16 October 2013

ER -

Kim J, Lee J, Lee C, Park E, Kim J, Kim H et al. Optimal feature selection for pedestrian detection based on logistic regression analysis. In Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013. 2013. p. 239-242. 6721800. (Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013). doi: 10.1109/SMC.2013.47

Optimal feature selection for pedestrian detection based on logistic regression analysis

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this