Extended feature descriptor and vehicle motion model with tracking-by-detection for pedestrian active safety

Hirokatsu Kataoka, Kimimasa Tamura, Kenji Iwata, Yutaka Satoh, Yasuhiro Matsui, Yoshimitsu Aoki

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

The percentage of pedestrian deaths in traffic accidents is on the rise in Japan. In recent years, there have been calls for measures to be introduced to protect vulnerable road users such as pedestrians and cyclists. In this study, a method to detect and track pedestrians using an in-vehicle camera is presented. We improve the technology of detecting pedestrians by using the highly accurate images obtained with a monocular camera. In the detection step, we employ ECoHOG as the feature descriptor; it accumulates the integrated gradient intensities. In the tracking step, we apply an effective motion model using optical flow and the proposed feature descriptor ECoHOG in a tracking-by-detection framework. These techniques were verified using images captured on real roads.

Original languageEnglish
Pages (from-to)296-304
Number of pages9
JournalIEICE Transactions on Information and Systems
VolumeE97-D
Issue number2
DOIs
Publication statusPublished - 2014

Fingerprint

Cameras
Highway accidents
Optical flows

Keywords

  • ECoHOG
  • Particle filter
  • Pedestrian active safety
  • Tracking-by-detection
  • Vehicle motion model

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Software
  • Artificial Intelligence
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition

Cite this

Extended feature descriptor and vehicle motion model with tracking-by-detection for pedestrian active safety. / Kataoka, Hirokatsu; Tamura, Kimimasa; Iwata, Kenji; Satoh, Yutaka; Matsui, Yasuhiro; Aoki, Yoshimitsu.

In: IEICE Transactions on Information and Systems, Vol. E97-D, No. 2, 2014, p. 296-304.

Research output: Contribution to journalArticle

Kataoka, Hirokatsu ; Tamura, Kimimasa ; Iwata, Kenji ; Satoh, Yutaka ; Matsui, Yasuhiro ; Aoki, Yoshimitsu. / Extended feature descriptor and vehicle motion model with tracking-by-detection for pedestrian active safety. In: IEICE Transactions on Information and Systems. 2014 ; Vol. E97-D, No. 2. pp. 296-304.
@article{03f3004602e74b729f190d93f92eb41f,
title = "Extended feature descriptor and vehicle motion model with tracking-by-detection for pedestrian active safety",
abstract = "The percentage of pedestrian deaths in traffic accidents is on the rise in Japan. In recent years, there have been calls for measures to be introduced to protect vulnerable road users such as pedestrians and cyclists. In this study, a method to detect and track pedestrians using an in-vehicle camera is presented. We improve the technology of detecting pedestrians by using the highly accurate images obtained with a monocular camera. In the detection step, we employ ECoHOG as the feature descriptor; it accumulates the integrated gradient intensities. In the tracking step, we apply an effective motion model using optical flow and the proposed feature descriptor ECoHOG in a tracking-by-detection framework. These techniques were verified using images captured on real roads.",
keywords = "ECoHOG, Particle filter, Pedestrian active safety, Tracking-by-detection, Vehicle motion model",
author = "Hirokatsu Kataoka and Kimimasa Tamura and Kenji Iwata and Yutaka Satoh and Yasuhiro Matsui and Yoshimitsu Aoki",
year = "2014",
doi = "10.1587/transinf.E97.D.296",
language = "English",
volume = "E97-D",
pages = "296--304",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "2",

}

TY - JOUR

T1 - Extended feature descriptor and vehicle motion model with tracking-by-detection for pedestrian active safety

AU - Kataoka, Hirokatsu

AU - Tamura, Kimimasa

AU - Iwata, Kenji

AU - Satoh, Yutaka

AU - Matsui, Yasuhiro

AU - Aoki, Yoshimitsu

PY - 2014

Y1 - 2014

N2 - The percentage of pedestrian deaths in traffic accidents is on the rise in Japan. In recent years, there have been calls for measures to be introduced to protect vulnerable road users such as pedestrians and cyclists. In this study, a method to detect and track pedestrians using an in-vehicle camera is presented. We improve the technology of detecting pedestrians by using the highly accurate images obtained with a monocular camera. In the detection step, we employ ECoHOG as the feature descriptor; it accumulates the integrated gradient intensities. In the tracking step, we apply an effective motion model using optical flow and the proposed feature descriptor ECoHOG in a tracking-by-detection framework. These techniques were verified using images captured on real roads.

AB - The percentage of pedestrian deaths in traffic accidents is on the rise in Japan. In recent years, there have been calls for measures to be introduced to protect vulnerable road users such as pedestrians and cyclists. In this study, a method to detect and track pedestrians using an in-vehicle camera is presented. We improve the technology of detecting pedestrians by using the highly accurate images obtained with a monocular camera. In the detection step, we employ ECoHOG as the feature descriptor; it accumulates the integrated gradient intensities. In the tracking step, we apply an effective motion model using optical flow and the proposed feature descriptor ECoHOG in a tracking-by-detection framework. These techniques were verified using images captured on real roads.

KW - ECoHOG

KW - Particle filter

KW - Pedestrian active safety

KW - Tracking-by-detection

KW - Vehicle motion model

UR - http://www.scopus.com/inward/record.url?scp=84893242144&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893242144&partnerID=8YFLogxK

U2 - 10.1587/transinf.E97.D.296

DO - 10.1587/transinf.E97.D.296

M3 - Article

AN - SCOPUS:84893242144

VL - E97-D

SP - 296

EP - 304

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 2

ER -