Issued quarterly (4 issues per year)

ISSN (Online): 2588-123X
ISSN (Print):
Researchgate Linkedin EmailFacebook Twitter
  • Home
  • Editorial Board
  • Guidelines
  • Policies
  • Submissions
  • Search
  • Archives
  • Announcements
  • Statistics

About TDTU

Ton Duc Thang University (TDTU) is a public university with the main campus located in vibrant Ho Chi Minh City, Vietnam’s economic and educational hub. Founded in 1997, TDTU has developed into one of the largest and fastest growing universities in Vietnam with more than 22,000 students, enrolled in undergraduate and graduate programs ranging from science, engineering to business management, law, and humanities. To foster the country’s human resources and best serve the nation in the knowledge based economy of the 21st century, TDTU is combining vocational training with high-level research. The establishment of JAEC is one of TDTU’s efforts in this direction. More

Publication Information


Publisher
Ton Duc Thang University
Honorary Editor-in-Chief
Tran Trong Dao
Executive Editor
Thai Hoang Chien
Chairman of the Editorial Board
Václav Snášel
Vice Chairman of the Editorial Board
Ivan Zelinka
Managing Editor
Nguyen-Thanh Nhon
Editorial Board
Timon Rabczuk
Hari Mohan Srivastava
Seung-Bok Choi
Carlo Cattani
Phan Thien Nhan
Nguyen Minh Tho
Adel M. Alimi
Petr Musílek
Jaroslav Pokorný
Juan Velasquez
Michal Wozniak
Hana Řezanková
Hendrik Richter
Mohammed Chadli
Nikolay V Kuznetsov
Shobhit K. Patel
Miroslav Vozňák
Roman Senkerik
Juan Carlos Burguillo Rial
Akhil Garg
Nguyen Pham Trung Hieu
Nguyen Quoc Hung
Aleš Zamuda
Ngo Son Tung
User

Guide for Authors

  • View 'Guide for Authors' online

Submit Your Paper

In order to submit your paper, please login and navigate to the author page.

If you do not have an account, please consider registering one.

Track Your Paper

Track accepted paper
Once your article has been accepted you will receive an email from Author Services. This email contains a link to check the status of your articles.

Click here to track your accepted papers
Journal Content

Browse
  • By Issue
  • By Author
  • By Title

Abstracting/Indexing 

  • Home
  • Vol 6, No 3 (2022)
  • Sukkar

Improve Detection and Tracking of Pedestrian Subclasses by Pre-Trained Models

Majdi Mustafa Sukkar, Dinesh Kumar, Jigneshsinh Sindha

Abstract


There are sub-classes of pedestrians that can be defined and it is important to distinguish between them for the detection in autonomous vehicle applications, such as elderly, and children, to reduce the risk of collision. It is necessary to talk about effective pedestrian tracking besides detection so that object remains accurately monitored, here the effective pre-trained algorithms come to achieve this goal in real-time. In this paper, we make a comparison between the detection and tracking algorithms, we applied the transfer learning technique to train the detection model on new sub-classes, after making Images augmentation in previous work, we got better results in detection, reached 0.81 mAP in real-time by using Yolov5 model, with a good tracking performance by the tracking algorithm dependent on detection Deep-SORT.


Creative Commons License

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium provided the original work is properly cited.



Keywords


Deep Learning, Yolov5, Deep-SORT, Pedestrian, Detection, Tracking.

Full Text:

PDF

Time cited: 0

Download citation



DOI: http://dx.doi.org/10.55579/jaec.202263.369

Refbacks

  • There are currently no refbacks.


Copyright (c) 2022 Journal of Advanced Engineering and Computation

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.


Owner: Ton Duc Thang University. All rights reserved.
License No: 507/GP-BTTTT, issued: 18th November 2016
Contact address: 19, Nguyen Huu Tho Street, Tan Phong Ward, District 7, Ho Chi Minh City
Tel: +84-28 3775 5037  Fax: +84-28 3775 5055
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.