Pedestrian detection with an ensemble of localized features

Shaopeng Tang, Satoshi Goto

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

In this paper, a new human detection approach from still image is proposed. Two vector features are extracted from the image. Histogram of oriented Gradient feature represents the gradient information of human. Histogram of modified local binary pattern is extracted from images convolved with Gabor filter, as a feature vector to represent texture information. It can be seen as a supplement of gradient information. Different support vector machine classifiers are trained by each type of vectors. Finally, two classifiers are combined together for the final result by using the proposed integration method. Because two features contain different information, they have low error dependency and can get high detection rate. Experiment is performed in a large dataset and it shows that this method outperforms state-of-the-art approaches and other combinations of features.

Original languageEnglish
Title of host publicationProceedings - IEEE International Symposium on Circuits and Systems
Pages2838-2841
Number of pages4
DOIs
Publication statusPublished - 2009
Event2009 IEEE International Symposium on Circuits and Systems, ISCAS 2009 - Taipei
Duration: 2009 May 242009 May 27

Other

Other2009 IEEE International Symposium on Circuits and Systems, ISCAS 2009
CityTaipei
Period09/5/2409/5/27

Fingerprint

Classifiers
Gabor filters
Support vector machines
Textures
Experiments

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Tang, S., & Goto, S. (2009). Pedestrian detection with an ensemble of localized features. In Proceedings - IEEE International Symposium on Circuits and Systems (pp. 2838-2841). [5118393] https://doi.org/10.1109/ISCAS.2009.5118393

Pedestrian detection with an ensemble of localized features. / Tang, Shaopeng; Goto, Satoshi.

Proceedings - IEEE International Symposium on Circuits and Systems. 2009. p. 2838-2841 5118393.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Tang, S & Goto, S 2009, Pedestrian detection with an ensemble of localized features. in Proceedings - IEEE International Symposium on Circuits and Systems., 5118393, pp. 2838-2841, 2009 IEEE International Symposium on Circuits and Systems, ISCAS 2009, Taipei, 09/5/24. https://doi.org/10.1109/ISCAS.2009.5118393
Tang S, Goto S. Pedestrian detection with an ensemble of localized features. In Proceedings - IEEE International Symposium on Circuits and Systems. 2009. p. 2838-2841. 5118393 https://doi.org/10.1109/ISCAS.2009.5118393
Tang, Shaopeng ; Goto, Satoshi. / Pedestrian detection with an ensemble of localized features. Proceedings - IEEE International Symposium on Circuits and Systems. 2009. pp. 2838-2841
@inproceedings{940bea08dfd0458f89f98de4aec89444,
title = "Pedestrian detection with an ensemble of localized features",
abstract = "In this paper, a new human detection approach from still image is proposed. Two vector features are extracted from the image. Histogram of oriented Gradient feature represents the gradient information of human. Histogram of modified local binary pattern is extracted from images convolved with Gabor filter, as a feature vector to represent texture information. It can be seen as a supplement of gradient information. Different support vector machine classifiers are trained by each type of vectors. Finally, two classifiers are combined together for the final result by using the proposed integration method. Because two features contain different information, they have low error dependency and can get high detection rate. Experiment is performed in a large dataset and it shows that this method outperforms state-of-the-art approaches and other combinations of features.",
author = "Shaopeng Tang and Satoshi Goto",
year = "2009",
doi = "10.1109/ISCAS.2009.5118393",
language = "English",
isbn = "9781424438280",
pages = "2838--2841",
booktitle = "Proceedings - IEEE International Symposium on Circuits and Systems",

}

TY - GEN

T1 - Pedestrian detection with an ensemble of localized features

AU - Tang, Shaopeng

AU - Goto, Satoshi

PY - 2009

Y1 - 2009

N2 - In this paper, a new human detection approach from still image is proposed. Two vector features are extracted from the image. Histogram of oriented Gradient feature represents the gradient information of human. Histogram of modified local binary pattern is extracted from images convolved with Gabor filter, as a feature vector to represent texture information. It can be seen as a supplement of gradient information. Different support vector machine classifiers are trained by each type of vectors. Finally, two classifiers are combined together for the final result by using the proposed integration method. Because two features contain different information, they have low error dependency and can get high detection rate. Experiment is performed in a large dataset and it shows that this method outperforms state-of-the-art approaches and other combinations of features.

AB - In this paper, a new human detection approach from still image is proposed. Two vector features are extracted from the image. Histogram of oriented Gradient feature represents the gradient information of human. Histogram of modified local binary pattern is extracted from images convolved with Gabor filter, as a feature vector to represent texture information. It can be seen as a supplement of gradient information. Different support vector machine classifiers are trained by each type of vectors. Finally, two classifiers are combined together for the final result by using the proposed integration method. Because two features contain different information, they have low error dependency and can get high detection rate. Experiment is performed in a large dataset and it shows that this method outperforms state-of-the-art approaches and other combinations of features.

UR - http://www.scopus.com/inward/record.url?scp=70350159243&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70350159243&partnerID=8YFLogxK

U2 - 10.1109/ISCAS.2009.5118393

DO - 10.1109/ISCAS.2009.5118393

M3 - Conference contribution

AN - SCOPUS:70350159243

SN - 9781424438280

SP - 2838

EP - 2841

BT - Proceedings - IEEE International Symposium on Circuits and Systems

ER -