A Novel Pedestrian Detector on Low-Resolution Images: Gradient LBP Using Patterns of Oriented Edges

Ahmed BOUDISSA; Joo Kooi TAN; Hyoungseop KIM; Takashi SHINOMIYA; Seiji ISHIKAWA

doi:10.1587/transinf.E96.D.2882

A Novel Pedestrian Detector on Low-Resolution Images: Gradient LBP Using Patterns of Oriented Edges

Ahmed BOUDISSA, Joo Kooi TAN, Hyoungseop KIM, Takashi SHINOMIYA, Seiji ISHIKAWA

Full Text Views

0

Share
Cite this

Summary :

This paper introduces a simple algorithm for pedestrian detection on low resolution images. The main objective is to create a successful means for real-time pedestrian detection. While the framework of the system consists of edge orientations combined with the local binary patterns (LBP) feature extractor, a novel way of selecting the threshold is introduced. Using the mean-variance of the background examples this threshold improves significantly the detection rate as well as the processing time. Furthermore, it makes the system robust to uniformly cluttered backgrounds, noise and light variations. The test data is the INRIA pedestrian dataset and for the classification, a support vector machine with a radial basis function (RBF) kernel is used. The system performs at state-of-the-art detection rates while being intuitive as well as very fast which leaves sufficient processing time for further operations such as tracking and danger estimation.

Publication: IEICE TRANSACTIONS on Information Vol.E96-D No.12 pp.2882-2887

Publication Date: 2013/12/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E96.D.2882

Type of Manuscript: LETTER

Category: Pattern Recognition

Authors

Ahmed BOUDISSA
  Kyushu Institute of Technology
Joo Kooi TAN
  Kyushu Institute of Technology
Hyoungseop KIM
  Kyushu Institute of Technology
Takashi SHINOMIYA
  Japan University of Economics
Seiji ISHIKAWA
  Kyushu Institute of Technology

Keyword

local binary patterns, pedestrian detection, object recognition, support vector machine

Cite this

Copy

Ahmed BOUDISSA, Joo Kooi TAN, Hyoungseop KIM, Takashi SHINOMIYA, Seiji ISHIKAWA, "A Novel Pedestrian Detector on Low-Resolution Images: Gradient LBP Using Patterns of Oriented Edges" in IEICE TRANSACTIONS on Information, vol. E96-D, no. 12, pp. 2882-2887, December 2013, doi: 10.1587/transinf.E96.D.2882.
Abstract: This paper introduces a simple algorithm for pedestrian detection on low resolution images. The main objective is to create a successful means for real-time pedestrian detection. While the framework of the system consists of edge orientations combined with the local binary patterns (LBP) feature extractor, a novel way of selecting the threshold is introduced. Using the mean-variance of the background examples this threshold improves significantly the detection rate as well as the processing time. Furthermore, it makes the system robust to uniformly cluttered backgrounds, noise and light variations. The test data is the INRIA pedestrian dataset and for the classification, a support vector machine with a radial basis function (RBF) kernel is used. The system performs at state-of-the-art detection rates while being intuitive as well as very fast which leaves sufficient processing time for further operations such as tracking and danger estimation.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.E96.D.2882/_p

Copy

@ARTICLE{e96-d_12_2882,
author={Ahmed BOUDISSA, Joo Kooi TAN, Hyoungseop KIM, Takashi SHINOMIYA, Seiji ISHIKAWA, },
journal={IEICE TRANSACTIONS on Information},
title={A Novel Pedestrian Detector on Low-Resolution Images: Gradient LBP Using Patterns of Oriented Edges},
year={2013},
volume={E96-D},
number={12},
pages={2882-2887},
abstract={This paper introduces a simple algorithm for pedestrian detection on low resolution images. The main objective is to create a successful means for real-time pedestrian detection. While the framework of the system consists of edge orientations combined with the local binary patterns (LBP) feature extractor, a novel way of selecting the threshold is introduced. Using the mean-variance of the background examples this threshold improves significantly the detection rate as well as the processing time. Furthermore, it makes the system robust to uniformly cluttered backgrounds, noise and light variations. The test data is the INRIA pedestrian dataset and for the classification, a support vector machine with a radial basis function (RBF) kernel is used. The system performs at state-of-the-art detection rates while being intuitive as well as very fast which leaves sufficient processing time for further operations such as tracking and danger estimation.},
keywords={},
doi={10.1587/transinf.E96.D.2882},
ISSN={1745-1361},
month={December},}

Copy

TY - JOUR
TI - A Novel Pedestrian Detector on Low-Resolution Images: Gradient LBP Using Patterns of Oriented Edges
T2 - IEICE TRANSACTIONS on Information
SP - 2882
EP - 2887
AU - Ahmed BOUDISSA
AU - Joo Kooi TAN
AU - Hyoungseop KIM
AU - Takashi SHINOMIYA
AU - Seiji ISHIKAWA
PY - 2013
DO - 10.1587/transinf.E96.D.2882
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E96-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2013
AB - This paper introduces a simple algorithm for pedestrian detection on low resolution images. The main objective is to create a successful means for real-time pedestrian detection. While the framework of the system consists of edge orientations combined with the local binary patterns (LBP) feature extractor, a novel way of selecting the threshold is introduced. Using the mean-variance of the background examples this threshold improves significantly the detection rate as well as the processing time. Furthermore, it makes the system robust to uniformly cluttered backgrounds, noise and light variations. The test data is the INRIA pedestrian dataset and for the classification, a support vector machine with a radial basis function (RBF) kernel is used. The system performs at state-of-the-art detection rates while being intuitive as well as very fast which leaves sufficient processing time for further operations such as tracking and danger estimation.
ER -