Novel outline features for pedestrian detection system with thermal images

Chun Fu Lin, Chin Sheng Chen, Wen-Jyi Hwang, Chih Yen Chen, Chi Hung Hwang, Chun Li Chang

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

Recently, the need of pedestrian detection at night has gained more and more interest. However, the performance of traditional nighttime pedestrian detection systems remains poor because region of interest (ROI) generation and feature extraction are designed separately. Thus, this paper presents novel thermal imagery algorithms to enhance the performance of nighttime pedestrian detection systems. The proposed thermal image pedestrian detection system involves novel outline features, developed from the ROI generation method of pedestrians that are different from traditional features. A three-layer back-propagation feed-forward neural network is used as the classifier. Two databases, the OTCBVS database and our own are used to evaluate the performance of the proposed thermal image pedestrian detection algorithm. Experimental results show that the proposed outline features are effective, and the detection performance of a traditional pedestrian detection system at night is improved.

Original languageEnglish
Pages (from-to)3440-3450
Number of pages11
JournalPattern Recognition
Volume48
Issue number11
DOIs
Publication statusPublished - 2015 Nov 1

Fingerprint

Feedforward neural networks
Backpropagation
Feature extraction
Classifiers
Hot Temperature

Keywords

  • Outline features
  • Pedestrian detection at night
  • Thermal images

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Novel outline features for pedestrian detection system with thermal images. / Lin, Chun Fu; Chen, Chin Sheng; Hwang, Wen-Jyi; Chen, Chih Yen; Hwang, Chi Hung; Chang, Chun Li.

In: Pattern Recognition, Vol. 48, No. 11, 01.11.2015, p. 3440-3450.

Research output: Contribution to journalArticle

Lin, Chun Fu ; Chen, Chin Sheng ; Hwang, Wen-Jyi ; Chen, Chih Yen ; Hwang, Chi Hung ; Chang, Chun Li. / Novel outline features for pedestrian detection system with thermal images. In: Pattern Recognition. 2015 ; Vol. 48, No. 11. pp. 3440-3450.
@article{c1e95ee3784d4cd4b74276621a1fea09,
title = "Novel outline features for pedestrian detection system with thermal images",
abstract = "Recently, the need of pedestrian detection at night has gained more and more interest. However, the performance of traditional nighttime pedestrian detection systems remains poor because region of interest (ROI) generation and feature extraction are designed separately. Thus, this paper presents novel thermal imagery algorithms to enhance the performance of nighttime pedestrian detection systems. The proposed thermal image pedestrian detection system involves novel outline features, developed from the ROI generation method of pedestrians that are different from traditional features. A three-layer back-propagation feed-forward neural network is used as the classifier. Two databases, the OTCBVS database and our own are used to evaluate the performance of the proposed thermal image pedestrian detection algorithm. Experimental results show that the proposed outline features are effective, and the detection performance of a traditional pedestrian detection system at night is improved.",
keywords = "Outline features, Pedestrian detection at night, Thermal images",
author = "Lin, {Chun Fu} and Chen, {Chin Sheng} and Wen-Jyi Hwang and Chen, {Chih Yen} and Hwang, {Chi Hung} and Chang, {Chun Li}",
year = "2015",
month = "11",
day = "1",
doi = "10.1016/j.patcog.2015.04.024",
language = "English",
volume = "48",
pages = "3440--3450",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "11",

}

TY - JOUR

T1 - Novel outline features for pedestrian detection system with thermal images

AU - Lin, Chun Fu

AU - Chen, Chin Sheng

AU - Hwang, Wen-Jyi

AU - Chen, Chih Yen

AU - Hwang, Chi Hung

AU - Chang, Chun Li

PY - 2015/11/1

Y1 - 2015/11/1

N2 - Recently, the need of pedestrian detection at night has gained more and more interest. However, the performance of traditional nighttime pedestrian detection systems remains poor because region of interest (ROI) generation and feature extraction are designed separately. Thus, this paper presents novel thermal imagery algorithms to enhance the performance of nighttime pedestrian detection systems. The proposed thermal image pedestrian detection system involves novel outline features, developed from the ROI generation method of pedestrians that are different from traditional features. A three-layer back-propagation feed-forward neural network is used as the classifier. Two databases, the OTCBVS database and our own are used to evaluate the performance of the proposed thermal image pedestrian detection algorithm. Experimental results show that the proposed outline features are effective, and the detection performance of a traditional pedestrian detection system at night is improved.

AB - Recently, the need of pedestrian detection at night has gained more and more interest. However, the performance of traditional nighttime pedestrian detection systems remains poor because region of interest (ROI) generation and feature extraction are designed separately. Thus, this paper presents novel thermal imagery algorithms to enhance the performance of nighttime pedestrian detection systems. The proposed thermal image pedestrian detection system involves novel outline features, developed from the ROI generation method of pedestrians that are different from traditional features. A three-layer back-propagation feed-forward neural network is used as the classifier. Two databases, the OTCBVS database and our own are used to evaluate the performance of the proposed thermal image pedestrian detection algorithm. Experimental results show that the proposed outline features are effective, and the detection performance of a traditional pedestrian detection system at night is improved.

KW - Outline features

KW - Pedestrian detection at night

KW - Thermal images

UR - http://www.scopus.com/inward/record.url?scp=84937812466&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84937812466&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2015.04.024

DO - 10.1016/j.patcog.2015.04.024

M3 - Article

AN - SCOPUS:84937812466

VL - 48

SP - 3440

EP - 3450

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 11

ER -