Accident Recognition via 3D CNNs for Automated Traffic Monitoring in Smart Cities

Mikhail Bortnikov; Adil Khan; Asad Masood Khattak; Muhammad Ahmad

doi:10.1007/978-3-030-17798-0_22

Accident Recognition via 3D CNNs for Automated Traffic Monitoring in Smart Cities

Mikhail Bortnikov
, Adil Khan^*
, Asad Masood Khattak
, Muhammad Ahmad

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

29 Scopus citations

Abstract

Automatic recognition of road accidents in traffic videos can improve road safety. Smart cities can deploy accident recognition systems to promote urban traffic safety and efficiency. This work reviews existing approaches for automatic accident detection and highlights a number of challenges that make accident detection a difficult task. Furthermore, we propose to implement a 3D Convolutional Neural Network (CNN) based accident detection system. We customize a video game to generate road traffic video data in a variety of weather and lighting conditions. The generated data is preprocessed using optical flow method and injected with noise to focus only on motion and introduce further variations in the data, respectively. The resulting data is used to train the model, which was then tested on real-life traffic videos from YouTube. The experiments demonstrate that the performance of the proposed algorithm is comparable to that of the existing models, but unlike them, it is not dependent on a large volume of real-life video data for training and does not require manual tuning of any thresholds.

Original language	English
Title of host publication	Advances in Computer Vision - Proceedings of the 2019 Computer Vision Conference CVC
Editors	Supriya Kapoor, Kohei Arai
Publisher	Springer Verlag
Pages	256-264
Number of pages	9
ISBN (Print)	9783030177973
DOIs	https://doi.org/10.1007/978-3-030-17798-0_22
State	Published - 2020
Externally published	Yes
Event	Computer Vision Conference, CVC 2019 - Las Vegas, United States Duration: 25 Apr 2019 → 26 Apr 2019

Publication series

Name	Advances in Intelligent Systems and Computing
Volume	944
ISSN (Print)	2194-5357
ISSN (Electronic)	2194-5365

Conference

Conference	Computer Vision Conference, CVC 2019
Country/Territory	United States
City	Las Vegas
Period	25/04/19 → 26/04/19

Bibliographical note

Publisher Copyright:
© 2020, Springer Nature Switzerland AG.

UN SDGs

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

SDG 3 Good Health and Well-being
SDG 11 Sustainable Cities and Communities

Keywords

3D convolutional neural networks
Accident recognition
Computer vision
Deep learning
Machine learning

ASJC Scopus subject areas

Control and Systems Engineering
General Computer Science

Access to Document

10.1007/978-3-030-17798-0_22

Cite this

Bortnikov, M., Khan, A., Khattak, A. M., & Ahmad, M. (2020). Accident Recognition via 3D CNNs for Automated Traffic Monitoring in Smart Cities. In S. Kapoor, & K. Arai (Eds.), Advances in Computer Vision - Proceedings of the 2019 Computer Vision Conference CVC (pp. 256-264). (Advances in Intelligent Systems and Computing; Vol. 944). Springer Verlag. https://doi.org/10.1007/978-3-030-17798-0_22

@inproceedings{e770fd8986e343249dddc9363751c3b6,

title = "Accident Recognition via 3D CNNs for Automated Traffic Monitoring in Smart Cities",

abstract = "Automatic recognition of road accidents in traffic videos can improve road safety. Smart cities can deploy accident recognition systems to promote urban traffic safety and efficiency. This work reviews existing approaches for automatic accident detection and highlights a number of challenges that make accident detection a difficult task. Furthermore, we propose to implement a 3D Convolutional Neural Network (CNN) based accident detection system. We customize a video game to generate road traffic video data in a variety of weather and lighting conditions. The generated data is preprocessed using optical flow method and injected with noise to focus only on motion and introduce further variations in the data, respectively. The resulting data is used to train the model, which was then tested on real-life traffic videos from YouTube. The experiments demonstrate that the performance of the proposed algorithm is comparable to that of the existing models, but unlike them, it is not dependent on a large volume of real-life video data for training and does not require manual tuning of any thresholds.",

keywords = "3D convolutional neural networks, Accident recognition, Computer vision, Deep learning, Machine learning",

author = "Mikhail Bortnikov and Adil Khan and Khattak, \{Asad Masood\} and Muhammad Ahmad",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.; Computer Vision Conference, CVC 2019 ; Conference date: 25-04-2019 Through 26-04-2019",

year = "2020",

doi = "10.1007/978-3-030-17798-0\_22",

language = "English",

isbn = "9783030177973",

series = "Advances in Intelligent Systems and Computing",

publisher = "Springer Verlag",

pages = "256--264",

editor = "Supriya Kapoor and Kohei Arai",

booktitle = "Advances in Computer Vision - Proceedings of the 2019 Computer Vision Conference CVC",

address = "Germany",

}

Bortnikov, M, Khan, A, Khattak, AM & Ahmad, M 2020, Accident Recognition via 3D CNNs for Automated Traffic Monitoring in Smart Cities. in S Kapoor & K Arai (eds), Advances in Computer Vision - Proceedings of the 2019 Computer Vision Conference CVC. Advances in Intelligent Systems and Computing, vol. 944, Springer Verlag, pp. 256-264, Computer Vision Conference, CVC 2019, Las Vegas, United States, 25/04/19. https://doi.org/10.1007/978-3-030-17798-0_22

Accident Recognition via 3D CNNs for Automated Traffic Monitoring in Smart Cities. / Bortnikov, Mikhail; Khan, Adil; Khattak, Asad Masood et al.
Advances in Computer Vision - Proceedings of the 2019 Computer Vision Conference CVC. ed. / Supriya Kapoor; Kohei Arai. Springer Verlag, 2020. p. 256-264 (Advances in Intelligent Systems and Computing; Vol. 944).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Accident Recognition via 3D CNNs for Automated Traffic Monitoring in Smart Cities

AU - Bortnikov, Mikhail

AU - Khan, Adil

AU - Khattak, Asad Masood

AU - Ahmad, Muhammad

PY - 2020

Y1 - 2020

N2 - Automatic recognition of road accidents in traffic videos can improve road safety. Smart cities can deploy accident recognition systems to promote urban traffic safety and efficiency. This work reviews existing approaches for automatic accident detection and highlights a number of challenges that make accident detection a difficult task. Furthermore, we propose to implement a 3D Convolutional Neural Network (CNN) based accident detection system. We customize a video game to generate road traffic video data in a variety of weather and lighting conditions. The generated data is preprocessed using optical flow method and injected with noise to focus only on motion and introduce further variations in the data, respectively. The resulting data is used to train the model, which was then tested on real-life traffic videos from YouTube. The experiments demonstrate that the performance of the proposed algorithm is comparable to that of the existing models, but unlike them, it is not dependent on a large volume of real-life video data for training and does not require manual tuning of any thresholds.

AB - Automatic recognition of road accidents in traffic videos can improve road safety. Smart cities can deploy accident recognition systems to promote urban traffic safety and efficiency. This work reviews existing approaches for automatic accident detection and highlights a number of challenges that make accident detection a difficult task. Furthermore, we propose to implement a 3D Convolutional Neural Network (CNN) based accident detection system. We customize a video game to generate road traffic video data in a variety of weather and lighting conditions. The generated data is preprocessed using optical flow method and injected with noise to focus only on motion and introduce further variations in the data, respectively. The resulting data is used to train the model, which was then tested on real-life traffic videos from YouTube. The experiments demonstrate that the performance of the proposed algorithm is comparable to that of the existing models, but unlike them, it is not dependent on a large volume of real-life video data for training and does not require manual tuning of any thresholds.

KW - 3D convolutional neural networks

KW - Accident recognition

KW - Computer vision

KW - Deep learning

KW - Machine learning

UR - https://www.scopus.com/pages/publications/85065464687

U2 - 10.1007/978-3-030-17798-0_22

DO - 10.1007/978-3-030-17798-0_22

M3 - Conference contribution

AN - SCOPUS:85065464687

SN - 9783030177973

T3 - Advances in Intelligent Systems and Computing

SP - 256

EP - 264

BT - Advances in Computer Vision - Proceedings of the 2019 Computer Vision Conference CVC

A2 - Kapoor, Supriya

A2 - Arai, Kohei

PB - Springer Verlag

T2 - Computer Vision Conference, CVC 2019

Y2 - 25 April 2019 through 26 April 2019

ER -

Accident Recognition via 3D CNNs for Automated Traffic Monitoring in Smart Cities

Abstract

Publication series

Conference

Bibliographical note

UN SDGs

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this