A Bus Crowdedness Sensing System Using Deep-Learning Based Object Detection

Wenhao HUANG; Akira TSUGE; Yin CHEN; Tadashi OKOSHI; Jin NAKAZAWA

doi:10.1587/transinf.2022PCP0007

A Bus Crowdedness Sensing System Using Deep-Learning Based Object Detection

Wenhao HUANG, Akira TSUGE, Yin CHEN, Tadashi OKOSHI, Jin NAKAZAWA

Full Text Views

0

Cite this

Summary :

Crowdedness of buses is playing an increasingly important role in the disease control of COVID-19. The lack of a practical approach to sensing the crowdedness of buses is a major problem. This paper proposes a bus crowdedness sensing system which exploits deep learning-based object detection to count the numbers of passengers getting on and off a bus and thus estimate the crowdedness of buses in real time. In our prototype system, we combine YOLOv5s object detection model with Kalman Filter object tracking algorithm to implement a sensing algorithm running on a Jetson nano-based vehicular device mounted on a bus. By using the driving recorder video data taken from real bus, we experimentally evaluate the performance of the proposed sensing system to verify that our proposed system system improves counting accuracy and achieves real-time processing at the Jetson Nano platform.

Publication: IEICE TRANSACTIONS on Information Vol.E105-D No.10 pp.1712-1720

Publication Date: 2022/10/01

Publicized: 2022/06/23

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2022PCP0007

Type of Manuscript: Special Section PAPER (Special Section on Picture Coding and Image Media Processing)

Category

Authors

Wenhao HUANG
  Keio University
Akira TSUGE
  Keio University
Yin CHEN
  Keio University,Reitaku University
Tadashi OKOSHI
  Keio University
Jin NAKAZAWA
  Keio University

Keyword

object detection, bus crowdedness sensing, deep learning, smart cities, image processing, edge computing

Cite this

Copy

Wenhao HUANG, Akira TSUGE, Yin CHEN, Tadashi OKOSHI, Jin NAKAZAWA, "A Bus Crowdedness Sensing System Using Deep-Learning Based Object Detection" in IEICE TRANSACTIONS on Information, vol. E105-D, no. 10, pp. 1712-1720, October 2022, doi: 10.1587/transinf.2022PCP0007.
Abstract: Crowdedness of buses is playing an increasingly important role in the disease control of COVID-19. The lack of a practical approach to sensing the crowdedness of buses is a major problem. This paper proposes a bus crowdedness sensing system which exploits deep learning-based object detection to count the numbers of passengers getting on and off a bus and thus estimate the crowdedness of buses in real time. In our prototype system, we combine YOLOv5s object detection model with Kalman Filter object tracking algorithm to implement a sensing algorithm running on a Jetson nano-based vehicular device mounted on a bus. By using the driving recorder video data taken from real bus, we experimentally evaluate the performance of the proposed sensing system to verify that our proposed system system improves counting accuracy and achieves real-time processing at the Jetson Nano platform.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022PCP0007/_p

Copy

@ARTICLE{e105-d_10_1712,
author={Wenhao HUANG, Akira TSUGE, Yin CHEN, Tadashi OKOSHI, Jin NAKAZAWA, },
journal={IEICE TRANSACTIONS on Information},
title={A Bus Crowdedness Sensing System Using Deep-Learning Based Object Detection},
year={2022},
volume={E105-D},
number={10},
pages={1712-1720},
abstract={Crowdedness of buses is playing an increasingly important role in the disease control of COVID-19. The lack of a practical approach to sensing the crowdedness of buses is a major problem. This paper proposes a bus crowdedness sensing system which exploits deep learning-based object detection to count the numbers of passengers getting on and off a bus and thus estimate the crowdedness of buses in real time. In our prototype system, we combine YOLOv5s object detection model with Kalman Filter object tracking algorithm to implement a sensing algorithm running on a Jetson nano-based vehicular device mounted on a bus. By using the driving recorder video data taken from real bus, we experimentally evaluate the performance of the proposed sensing system to verify that our proposed system system improves counting accuracy and achieves real-time processing at the Jetson Nano platform.},
keywords={},
doi={10.1587/transinf.2022PCP0007},
ISSN={1745-1361},
month={October},}

Copy

TY - JOUR
TI - A Bus Crowdedness Sensing System Using Deep-Learning Based Object Detection
T2 - IEICE TRANSACTIONS on Information
SP - 1712
EP - 1720
AU - Wenhao HUANG
AU - Akira TSUGE
AU - Yin CHEN
AU - Tadashi OKOSHI
AU - Jin NAKAZAWA
PY - 2022
DO - 10.1587/transinf.2022PCP0007
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E105-D
IS - 10
JA - IEICE TRANSACTIONS on Information
Y1 - October 2022
AB - Crowdedness of buses is playing an increasingly important role in the disease control of COVID-19. The lack of a practical approach to sensing the crowdedness of buses is a major problem. This paper proposes a bus crowdedness sensing system which exploits deep learning-based object detection to count the numbers of passengers getting on and off a bus and thus estimate the crowdedness of buses in real time. In our prototype system, we combine YOLOv5s object detection model with Kalman Filter object tracking algorithm to implement a sensing algorithm running on a Jetson nano-based vehicular device mounted on a bus. By using the driving recorder video data taken from real bus, we experimentally evaluate the performance of the proposed sensing system to verify that our proposed system system improves counting accuracy and achieves real-time processing at the Jetson Nano platform.
ER -