Deep Learning-based Visual Tracking of UAVs using a PTZ Camera System

Halil Utku Unlu; Phillip Stefan Niehaus; Daniel Chirita; Nikolaos Evangeliou; Anthony Tzes

doi:10.1109/IECON.2019.8927731

Deep Learning-based Visual Tracking of UAVs using a PTZ Camera System

Halil Utku Unlu, Phillip Stefan Niehaus, Daniel Chirita, Nikolaos Evangeliou, Anthony Tzes

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

Abstract

The visual tracking problem of Unmanned Aerial Vehicles (UAVs) with a Pan-Tilt-Zoom (PTZ) camera system is the subject of this article. Given the background of an image acquired by a PTZ-camera system, a border encompassing a moving object is computed relying on optical flow and the histogram of oriented gradients. Deep Learning (DL) algorithms are trained off-line to decide on the existence of a UAV within this border. Particularly, the ResNet-50 model was trained using a collected data set with more than 50,000 registered positive images. Having identified a UAV, a visual servoing scheme is employed to adjust the PTZ-parameters in order for the border of a detected UAV to span as large as possible the cameras Field of View. The advocated servoing scheme is robust enough against the UAVs rapid maneuvers. Experimental studies are offered to highlight the efficiency of the suggested scheme.

Original language	English (US)
Title of host publication	Proceedings
Subtitle of host publication	IECON 2019 - 45th Annual Conference of the IEEE Industrial Electronics Society
Publisher	IEEE Computer Society
Pages	638-644
Number of pages	7
ISBN (Electronic)	9781728148786
DOIs	https://doi.org/10.1109/IECON.2019.8927731
State	Published - Oct 2019
Event	45th Annual Conference of the IEEE Industrial Electronics Society, IECON 2019 - Lisbon, Portugal Duration: Oct 14 2019 → Oct 17 2019

Publication series

Name	IECON Proceedings (Industrial Electronics Conference)
Volume	2019-October

Conference

Conference	45th Annual Conference of the IEEE Industrial Electronics Society, IECON 2019
Country/Territory	Portugal
City	Lisbon
Period	10/14/19 → 10/17/19

Keywords

Deep Learning
PTZ camera platform
Unmanned Aerial Systems
Visual Tracking

ASJC Scopus subject areas

Control and Systems Engineering
Electrical and Electronic Engineering

Access to Document

10.1109/IECON.2019.8927731

Cite this

Unlu, H. U., Niehaus, P. S., Chirita, D., Evangeliou, N., & Tzes, A. (2019). Deep Learning-based Visual Tracking of UAVs using a PTZ Camera System. In Proceedings: IECON 2019 - 45th Annual Conference of the IEEE Industrial Electronics Society (pp. 638-644). Article 8927731 (IECON Proceedings (Industrial Electronics Conference); Vol. 2019-October). IEEE Computer Society. https://doi.org/10.1109/IECON.2019.8927731

Deep Learning-based Visual Tracking of UAVs using a PTZ Camera System. / Unlu, Halil Utku; Niehaus, Phillip Stefan; Chirita, Daniel et al.
Proceedings: IECON 2019 - 45th Annual Conference of the IEEE Industrial Electronics Society. IEEE Computer Society, 2019. p. 638-644 8927731 (IECON Proceedings (Industrial Electronics Conference); Vol. 2019-October).

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

Unlu, HU, Niehaus, PS, Chirita, D, Evangeliou, N & Tzes, A 2019, Deep Learning-based Visual Tracking of UAVs using a PTZ Camera System. in Proceedings: IECON 2019 - 45th Annual Conference of the IEEE Industrial Electronics Society., 8927731, IECON Proceedings (Industrial Electronics Conference), vol. 2019-October, IEEE Computer Society, pp. 638-644, 45th Annual Conference of the IEEE Industrial Electronics Society, IECON 2019, Lisbon, Portugal, 10/14/19. https://doi.org/10.1109/IECON.2019.8927731

@inproceedings{64ffce99ee164d5e8e58512a8f229b78,

title = "Deep Learning-based Visual Tracking of UAVs using a PTZ Camera System",

abstract = "The visual tracking problem of Unmanned Aerial Vehicles (UAVs) with a Pan-Tilt-Zoom (PTZ) camera system is the subject of this article. Given the background of an image acquired by a PTZ-camera system, a border encompassing a moving object is computed relying on optical flow and the histogram of oriented gradients. Deep Learning (DL) algorithms are trained off-line to decide on the existence of a UAV within this border. Particularly, the ResNet-50 model was trained using a collected data set with more than 50,000 registered positive images. Having identified a UAV, a visual servoing scheme is employed to adjust the PTZ-parameters in order for the border of a detected UAV to span as large as possible the cameras Field of View. The advocated servoing scheme is robust enough against the UAVs rapid maneuvers. Experimental studies are offered to highlight the efficiency of the suggested scheme.",

keywords = "Deep Learning, PTZ camera platform, Unmanned Aerial Systems, Visual Tracking",

author = "Unlu, {Halil Utku} and Niehaus, {Phillip Stefan} and Daniel Chirita and Nikolaos Evangeliou and Anthony Tzes",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 45th Annual Conference of the IEEE Industrial Electronics Society, IECON 2019 ; Conference date: 14-10-2019 Through 17-10-2019",

year = "2019",

month = oct,

doi = "10.1109/IECON.2019.8927731",

language = "English (US)",

series = "IECON Proceedings (Industrial Electronics Conference)",

publisher = "IEEE Computer Society",

pages = "638--644",

booktitle = "Proceedings",

}

TY - GEN

T1 - Deep Learning-based Visual Tracking of UAVs using a PTZ Camera System

AU - Unlu, Halil Utku

AU - Niehaus, Phillip Stefan

AU - Chirita, Daniel

AU - Evangeliou, Nikolaos

AU - Tzes, Anthony

PY - 2019/10

Y1 - 2019/10

N2 - The visual tracking problem of Unmanned Aerial Vehicles (UAVs) with a Pan-Tilt-Zoom (PTZ) camera system is the subject of this article. Given the background of an image acquired by a PTZ-camera system, a border encompassing a moving object is computed relying on optical flow and the histogram of oriented gradients. Deep Learning (DL) algorithms are trained off-line to decide on the existence of a UAV within this border. Particularly, the ResNet-50 model was trained using a collected data set with more than 50,000 registered positive images. Having identified a UAV, a visual servoing scheme is employed to adjust the PTZ-parameters in order for the border of a detected UAV to span as large as possible the cameras Field of View. The advocated servoing scheme is robust enough against the UAVs rapid maneuvers. Experimental studies are offered to highlight the efficiency of the suggested scheme.

AB - The visual tracking problem of Unmanned Aerial Vehicles (UAVs) with a Pan-Tilt-Zoom (PTZ) camera system is the subject of this article. Given the background of an image acquired by a PTZ-camera system, a border encompassing a moving object is computed relying on optical flow and the histogram of oriented gradients. Deep Learning (DL) algorithms are trained off-line to decide on the existence of a UAV within this border. Particularly, the ResNet-50 model was trained using a collected data set with more than 50,000 registered positive images. Having identified a UAV, a visual servoing scheme is employed to adjust the PTZ-parameters in order for the border of a detected UAV to span as large as possible the cameras Field of View. The advocated servoing scheme is robust enough against the UAVs rapid maneuvers. Experimental studies are offered to highlight the efficiency of the suggested scheme.

KW - Deep Learning

KW - PTZ camera platform

KW - Unmanned Aerial Systems

KW - Visual Tracking

UR - http://www.scopus.com/inward/record.url?scp=85084156996&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85084156996&partnerID=8YFLogxK

U2 - 10.1109/IECON.2019.8927731

DO - 10.1109/IECON.2019.8927731

M3 - Conference contribution

AN - SCOPUS:85084156996

T3 - IECON Proceedings (Industrial Electronics Conference)

SP - 638

EP - 644

BT - Proceedings

PB - IEEE Computer Society

T2 - 45th Annual Conference of the IEEE Industrial Electronics Society, IECON 2019

Y2 - 14 October 2019 through 17 October 2019

ER -

Deep Learning-based Visual Tracking of UAVs using a PTZ Camera System

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this