Design of UAV Single Object Tracking Algorithm Based on Feature Fusion

Changze Li, Xiaoxiong Liu, Xingwang Zhang, Bin Qin

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

3 Scopus citations

Abstract

Considering the real-time performance and robustness of the object tracking algorithm, this paper proposes an improved tracking algorithm for single object tracking based on UAV. The algorithm uses Tiny-YOLOv3 for preliminary detection, and the detection results combine histogram of orientation gradient(HOG) and RGB histogram to extract the features. We use histogram matching to find the highest similarity between the detected candidate object and the object to be tracked to achieve the purpose of tracking. Different tracking strategies are designed when the object is stationary and moving. The experimental results show that the algorithm improves tracking accuracy and robustness while ensuring real-time tracking.

Original languageEnglish
Title of host publicationProceedings of the 40th Chinese Control Conference, CCC 2021
EditorsChen Peng, Jian Sun
PublisherIEEE Computer Society
Pages3088-3092
Number of pages5
ISBN (Electronic)9789881563804
DOIs
StatePublished - 26 Jul 2021
Event40th Chinese Control Conference, CCC 2021 - Shanghai, China
Duration: 26 Jul 202128 Jul 2021

Publication series

NameChinese Control Conference, CCC
Volume2021-July
ISSN (Print)1934-1768
ISSN (Electronic)2161-2927

Conference

Conference40th Chinese Control Conference, CCC 2021
Country/TerritoryChina
CityShanghai
Period26/07/2128/07/21

Keywords

  • histogram matching
  • HOG
  • Single object tracking
  • Tiny-YOLOv3
  • UAV

Fingerprint

Dive into the research topics of 'Design of UAV Single Object Tracking Algorithm Based on Feature Fusion'. Together they form a unique fingerprint.

Cite this