Real-time Traffic Analysis Using Deep Learning Techniques And UAV Based Video

HUAIZHONG ZHANG; MARK LIPTROTT; NIKOLAOS BESSIS; Jianquan Chen

doi:10.1109/AVSS.2019.8909879

Real-time Traffic Analysis Using Deep Learning Techniques And UAV Based Video

HUAIZHONG ZHANG, MARK LIPTROTT, NIKOLAOS BESSIS, Jianquan Chen

Computer Science

Manchester Metropolitan University

Research output: Contribution to journal › Article (journal) › peer-review

37 Citations (Scopus)

424 Downloads (Pure)

Abstract

In urban environments there are daily issues of traffic
congestion which city authorities need to address. Realtime
analysis of traffic flow information is crucial for
efficiently managing urban traffic. This paper aims to
conduct traffic analysis using UAV-based videos and deep
learning techniques. The road traffic video is collected by
using a position-fixed UAV. The most recent deep learning
methods are applied to identify the moving objects in
videos. The relevant mobility metrics are calculated to
conduct traffic analysis and measure the consequences of
traffic congestion. The proposed approach is validated with
the manual analysis results and the visualization results.
The traffic analysis process is real-time in terms of the pretrained
model used.

Original language	English
Number of pages	5
Journal	2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)
Early online date	25 Nov 2019
DOIs	https://doi.org/10.1109/AVSS.2019.8909879
Publication status	E-pub ahead of print - 25 Nov 2019
Event	16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) - Taipei, Taiwan, Province of China Duration: 18 Sept 2019 → 21 Sept 2019

Keywords

Traffic congestion
UAV
Deep learning

Research Centres

Data and Complex Systems Research Centre

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/AVSS.2019.8909879

PID6215077Accepted author manuscript, 529 KBLicence: CC BY-NC-ND

Cite this

@article{2776ca88a21242039ffd6411859fb1ba,

title = "Real-time Traffic Analysis Using Deep Learning Techniques And UAV Based Video",

abstract = "In urban environments there are daily issues of trafficcongestion which city authorities need to address. Realtimeanalysis of traffic flow information is crucial forefficiently managing urban traffic. This paper aims toconduct traffic analysis using UAV-based videos and deeplearning techniques. The road traffic video is collected byusing a position-fixed UAV. The most recent deep learningmethods are applied to identify the moving objects invideos. The relevant mobility metrics are calculated toconduct traffic analysis and measure the consequences oftraffic congestion. The proposed approach is validated withthe manual analysis results and the visualization results.The traffic analysis process is real-time in terms of the pretrainedmodel used.",

keywords = "Traffic congestion, UAV, Deep learning",

author = "HUAIZHONG ZHANG and MARK LIPTROTT and NIKOLAOS BESSIS and Jianquan Chen",

year = "2019",

month = nov,

day = "25",

doi = "10.1109/AVSS.2019.8909879",

language = "English",

journal = "2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)",

issn = "2643-6213",

publisher = "IEEE",

note = "16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) ; Conference date: 18-09-2019 Through 21-09-2019",

}

TY - JOUR

T1 - Real-time Traffic Analysis Using Deep Learning Techniques And UAV Based Video

AU - ZHANG, HUAIZHONG

AU - LIPTROTT, MARK

AU - BESSIS, NIKOLAOS

AU - Chen, Jianquan

PY - 2019/11/25

Y1 - 2019/11/25

N2 - In urban environments there are daily issues of trafficcongestion which city authorities need to address. Realtimeanalysis of traffic flow information is crucial forefficiently managing urban traffic. This paper aims toconduct traffic analysis using UAV-based videos and deeplearning techniques. The road traffic video is collected byusing a position-fixed UAV. The most recent deep learningmethods are applied to identify the moving objects invideos. The relevant mobility metrics are calculated toconduct traffic analysis and measure the consequences oftraffic congestion. The proposed approach is validated withthe manual analysis results and the visualization results.The traffic analysis process is real-time in terms of the pretrainedmodel used.

AB - In urban environments there are daily issues of trafficcongestion which city authorities need to address. Realtimeanalysis of traffic flow information is crucial forefficiently managing urban traffic. This paper aims toconduct traffic analysis using UAV-based videos and deeplearning techniques. The road traffic video is collected byusing a position-fixed UAV. The most recent deep learningmethods are applied to identify the moving objects invideos. The relevant mobility metrics are calculated toconduct traffic analysis and measure the consequences oftraffic congestion. The proposed approach is validated withthe manual analysis results and the visualization results.The traffic analysis process is real-time in terms of the pretrainedmodel used.

KW - Traffic congestion

KW - UAV

KW - Deep learning

U2 - 10.1109/AVSS.2019.8909879

DO - 10.1109/AVSS.2019.8909879

M3 - Article (journal)

SN - 2643-6213

JO - 2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)

JF - 2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)

T2 - 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)

Y2 - 18 September 2019 through 21 September 2019

ER -

Real-time Traffic Analysis Using Deep Learning Techniques And UAV Based Video

Abstract

Keywords

Research Centres

UN SDGs

Access to Document

Fingerprint

Cite this