Vision-based Object Tracking in Marine Environments using Features from Neural Network Detections

Frederik Emil Thorsson Schöller; Mogens Blanke; Martin Krarup Plenge-Feidenhans'l; Lazaros Nalpantidis

doi:10.1016/j.ifacol.2020.12.1455

Vision-based Object Tracking in Marine Environments using Features from Neural Network Detections

Frederik Emil Thorsson Schöller
, Mogens Blanke
, Martin Krarup Plenge-Feidenhans'l
, Lazaros Nalpantidis

Research output: Contribution to journal › Conference article › Research › peer-review

305 Downloads (Orbit)

Abstract

Autonomous decision support is desired to enable navigation with a temporally unattended bridge or to have the vessel navigated remotely. In order to have safe navigation, it is crucial to correctly interpret the current situation given any scenario. Proper perception of the surrounding environment is essential for good situational awareness. This paper suggests a method for tracking objects that have been detected by a neural network. The method utilises features that have been computed during the detection step, thereby ensuring good features that are representative for the given objects while saving the time it would take to compute new features. The suggested method is evaluated on data acquired in Danish near-coastal waters. Evaluation shows that the tracking method is able to track the detections well with few switches of object identity. The method is shown to outperform a similar tracking algorithm, while keeping the speed needed for real-time applications.

Original language	English
Book series	IFAC-PapersOnLine
Volume	53
Issue number	2
Pages (from-to)	14517-14523
ISSN	2405-8963
DOIs	https://doi.org/10.1016/j.ifacol.2020.12.1455
Publication status	Published - 2021
Event	IFAC World Congress 2020 - Virtual, Berlin, Germany Duration: 13 Jul 2020 → 17 Jul 2020 https://www.ifac2020.org/

Conference

Conference	IFAC World Congress 2020
Location	Virtual
Country/Territory	Germany
City	Berlin
Period	13/07/2020 → 17/07/2020
Internet address	https://www.ifac2020.org/

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 14 Life Below Water

Keywords

Autonomous Marine Vessels
Navigation
Computer Vision
Object Tracking

Access to Document

10.1016/j.ifacol.2020.12.1455

FulltextFinal published version, 657 KB

OpenUrl availability

Check availability in DTU Findit

ShippingLab Autonomy
Blanke, M. (PI), Galeazzi, R. (CoPI), Dittmann, K. (CoPI), Hansen, S. (CoPI), Papageorgiou, D. (Supervisor), Nalpantidis, L. (Supervisor), Schöller, F. E. T. S. (PhD Student), Plenge-Feidenhans'l, M. K. (PhD Student), Hansen, N. (PhD Student), Andersen, R. H. (Project Participant), Becktor, J. B. (PhD Student), Enevoldsen, T. T. (PhD Student), Dagdilelis, D. (PhD Student), Karstensen, P. I. H. (Project Participant), Nielsen, R. E. (Project Participant), Garde, J. (Project Participant), Ravn, O. (Supervisor), Christin, L. P. E. (PI) & Nielsen, R. E. (Project Participant)
01/04/2019 → 31/12/2022
Project: Research

Cite this

@inproceedings{e7361cee7e7d417c8d48f919bcf2e431,

title = "Vision-based Object Tracking in Marine Environments using Features from Neural Network Detections",

abstract = "Autonomous decision support is desired to enable navigation with a temporally unattended bridge or to have the vessel navigated remotely. In order to have safe navigation, it is crucial to correctly interpret the current situation given any scenario. Proper perception of the surrounding environment is essential for good situational awareness. This paper suggests a method for tracking objects that have been detected by a neural network. The method utilises features that have been computed during the detection step, thereby ensuring good features that are representative for the given objects while saving the time it would take to compute new features. The suggested method is evaluated on data acquired in Danish near-coastal waters. Evaluation shows that the tracking method is able to track the detections well with few switches of object identity. The method is shown to outperform a similar tracking algorithm, while keeping the speed needed for real-time applications.",

keywords = "Autonomous Marine Vessels, Navigation, Computer Vision, Object Tracking",

author = "Sch{\"o}ller, \{Frederik Emil Thorsson\} and Mogens Blanke and Plenge-Feidenhans'l, \{Martin Krarup\} and Lazaros Nalpantidis",

year = "2021",

doi = "10.1016/j.ifacol.2020.12.1455",

language = "English",

volume = "53",

pages = "14517--14523",

journal = "IFAC-PapersOnLine",

issn = "2405-8963",

publisher = "Elsevier",

number = "2",

note = "IFAC World Congress 2020, IFAC 2020 ; Conference date: 13-07-2020 Through 17-07-2020",

url = "https://www.ifac2020.org/",

}

TY - GEN

T1 - Vision-based Object Tracking in Marine Environments using Features from Neural Network Detections

AU - Schöller, Frederik Emil Thorsson

AU - Blanke, Mogens

AU - Plenge-Feidenhans'l, Martin Krarup

AU - Nalpantidis, Lazaros

PY - 2021

Y1 - 2021

N2 - Autonomous decision support is desired to enable navigation with a temporally unattended bridge or to have the vessel navigated remotely. In order to have safe navigation, it is crucial to correctly interpret the current situation given any scenario. Proper perception of the surrounding environment is essential for good situational awareness. This paper suggests a method for tracking objects that have been detected by a neural network. The method utilises features that have been computed during the detection step, thereby ensuring good features that are representative for the given objects while saving the time it would take to compute new features. The suggested method is evaluated on data acquired in Danish near-coastal waters. Evaluation shows that the tracking method is able to track the detections well with few switches of object identity. The method is shown to outperform a similar tracking algorithm, while keeping the speed needed for real-time applications.

AB - Autonomous decision support is desired to enable navigation with a temporally unattended bridge or to have the vessel navigated remotely. In order to have safe navigation, it is crucial to correctly interpret the current situation given any scenario. Proper perception of the surrounding environment is essential for good situational awareness. This paper suggests a method for tracking objects that have been detected by a neural network. The method utilises features that have been computed during the detection step, thereby ensuring good features that are representative for the given objects while saving the time it would take to compute new features. The suggested method is evaluated on data acquired in Danish near-coastal waters. Evaluation shows that the tracking method is able to track the detections well with few switches of object identity. The method is shown to outperform a similar tracking algorithm, while keeping the speed needed for real-time applications.

KW - Autonomous Marine Vessels

KW - Navigation

KW - Computer Vision

KW - Object Tracking

U2 - 10.1016/j.ifacol.2020.12.1455

DO - 10.1016/j.ifacol.2020.12.1455

M3 - Conference article

SN - 2405-8963

VL - 53

SP - 14517

EP - 14523

JO - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

IS - 2

T2 - IFAC World Congress 2020

Y2 - 13 July 2020 through 17 July 2020

ER -

Vision-based Object Tracking in Marine Environments using Features from Neural Network Detections

Abstract

Conference

UN SDGs

Keywords

Access to Document

OpenUrl availability

Fingerprint

Projects

ShippingLab Autonomy

Cite this