Advanced

Spatio-temporal patterns of traffic-related air pollutant emissions in different urban functional zones estimated by real-time video and deep learning technique

Song, Jinchao ; Zhao, Chunli LU ; Lin, Tao ; Li, X. and Prishchepov, Alexander V. (2019) In Journal of Cleaner Production 238.
Abstract

The aim of this paper is to explore the relationship between spatial-temporal patterns of vehicles types and numbers in different urban functional zones and traffic-related air pollutant emissions with real-time traffic data collected from traffic surveillance video and image recognition. The data were analyzed by using video-based detection technique, while the air pollution was quantified via pollutant emission coefficients. The results revealed that: (1) the order of traffic-related pollutant emissions was expressway > business zone > industrial zone > residential zone > port; (2) daily maximum emissions of each pollutant occurred in different functional zones on weekdays and weekends. With the exception of expressway,... (More)

The aim of this paper is to explore the relationship between spatial-temporal patterns of vehicles types and numbers in different urban functional zones and traffic-related air pollutant emissions with real-time traffic data collected from traffic surveillance video and image recognition. The data were analyzed by using video-based detection technique, while the air pollution was quantified via pollutant emission coefficients. The results revealed that: (1) the order of traffic-related pollutant emissions was expressway > business zone > industrial zone > residential zone > port; (2) daily maximum emissions of each pollutant occurred in different functional zones on weekdays and weekends. With the exception of expressway, the business zones had the highest emissions of CO, HC and VOC on weekdays, while the highest emissions of all the pollutants (CO, HC, NOx, PM2.5, PM1.0, and VOC) were at the weekend. The industrial zone had the highest emissions of NOx, PM2.5 and PM1.0 on weekdays; (3) pollutant emissions (CO, HC, NOx, PM2.5, PM1.0 and VOC) in all functional zones peaked in the morning and evening peak except at port sites; (4) cars and motorcycles represented the major source of traffic-related pollutant emissions. Collecting data through video-based vehicle detection with finer spatio-temporal resolution represents a cost-effective way of mapping spatio-temporal patterns of traffic-related air pollution to contribute to urban planning and climate change studies.

(Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Deep learning, Pollutant emissions, Urban functional zones, Video-based vehicle detection
in
Journal of Cleaner Production
volume
238
article number
117881
publisher
Elsevier
external identifiers
  • scopus:85070507232
ISSN
0959-6526
DOI
10.1016/j.jclepro.2019.117881
language
English
LU publication?
yes
id
fd1dec38-ba0e-4462-b7c7-09dd341bbc38
date added to LUP
2019-08-29 12:58:17
date last changed
2020-01-13 02:18:37
@article{fd1dec38-ba0e-4462-b7c7-09dd341bbc38,
  abstract     = {<p>The aim of this paper is to explore the relationship between spatial-temporal patterns of vehicles types and numbers in different urban functional zones and traffic-related air pollutant emissions with real-time traffic data collected from traffic surveillance video and image recognition. The data were analyzed by using video-based detection technique, while the air pollution was quantified via pollutant emission coefficients. The results revealed that: (1) the order of traffic-related pollutant emissions was expressway &gt; business zone &gt; industrial zone &gt; residential zone &gt; port; (2) daily maximum emissions of each pollutant occurred in different functional zones on weekdays and weekends. With the exception of expressway, the business zones had the highest emissions of CO, HC and VOC on weekdays, while the highest emissions of all the pollutants (CO, HC, NOx, PM2.5, PM1.0, and VOC) were at the weekend. The industrial zone had the highest emissions of NOx, PM2.5 and PM1.0 on weekdays; (3) pollutant emissions (CO, HC, NOx, PM2.5, PM1.0 and VOC) in all functional zones peaked in the morning and evening peak except at port sites; (4) cars and motorcycles represented the major source of traffic-related pollutant emissions. Collecting data through video-based vehicle detection with finer spatio-temporal resolution represents a cost-effective way of mapping spatio-temporal patterns of traffic-related air pollution to contribute to urban planning and climate change studies.</p>},
  author       = {Song, Jinchao and Zhao, Chunli and Lin, Tao and Li, X. and Prishchepov, Alexander V.},
  issn         = {0959-6526},
  language     = {eng},
  publisher    = {Elsevier},
  series       = {Journal of Cleaner Production},
  title        = {Spatio-temporal patterns of traffic-related air pollutant emissions in different urban functional zones estimated by real-time video and deep learning technique},
  url          = {http://dx.doi.org/10.1016/j.jclepro.2019.117881},
  doi          = {10.1016/j.jclepro.2019.117881},
  volume       = {238},
  year         = {2019},
}