Research on key technologies of UAV real-time recognition and tracking based on YOLOv4

  • Hongming Liang
  • , Tao Hong
  • , Zhihua Chen
  • , Michel Kadoch

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

With the development of the Internet of Things (IoT) technology and the fifth generation of mobile communication technology (5G), unmanned aerial vehicle (UAV) has been widely used in urban life, such as UAV photography and monitoring, and UAV delivery. However, UAV targets are generally small in size, and characteristics between multiple UAVs are not obvious. Interference from other flying objects and complex electromagnetic environment bring great challenges to accurate detection and stable tracking of UAV targets. Therefore, according to the characteristics of urban environment, it is very necessary to establish a high-efficiency UAV target real-time detection and tracking system. In our study, a UAV detection and tracking method with YOLOv4 as the target detector and Deep-SORT as the target tracker is proposed, which can realize the high-efficiency and high-precision identification and tracking of multiple UAV targets.

Original languageEnglish
Title of host publicationProceedings - 2022 International Conference on Information Processing and Network Provisioning, ICIPNP 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages107-110
Number of pages4
ISBN (Electronic)9781665464055
DOIs
StatePublished - 2022
Event2022 International Conference on Information Processing and Network Provisioning, ICIPNP 2022 - Beijing, China
Duration: 20 Oct 202221 Oct 2022

Publication series

NameProceedings - 2022 International Conference on Information Processing and Network Provisioning, ICIPNP 2022

Conference

Conference2022 International Conference on Information Processing and Network Provisioning, ICIPNP 2022
Country/TerritoryChina
CityBeijing
Period20/10/2221/10/22

Keywords

  • DeepSORT
  • UAV
  • YOLOv4
  • multi-target detection
  • multi-target tracking

Fingerprint

Dive into the research topics of 'Research on key technologies of UAV real-time recognition and tracking based on YOLOv4'. Together they form a unique fingerprint.

Cite this