Vessel Bearing Estimation Using Visible and Thermal Imaging
Loading...
Access rights
openAccess
URL
Journal Title
Journal ISSN
Volume Title
A4 Artikkeli konferenssijulkaisussa
This publication is imported from Aalto University research portal.
View publication in the Research portal (opens in new window)
View/Open full text file from the Research portal (opens in new window)
Other link related to publication (opens in new window)
View publication in the Research portal (opens in new window)
View/Open full text file from the Research portal (opens in new window)
Other link related to publication (opens in new window)
Date
2023
Major/Subject
Mcode
Degree programme
Language
en
Pages
9
373-381
373-381
Series
Image Analysis - 23rd Scandinavian Conference, SCIA 2023, Proceedings, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Volume 13886 LNCS
Abstract
Maritime awareness and autonomous navigation can be enabled by state-of-the-art deep learning methods, by monitoring and tracking the position of maritime vessels. In our experiment, we acquire ship dataset from Megastar cruise ferry campaigns in Baltic Sea. We detect the nearby vessels using visible and infrared imaging sensors and object detectors You-Only-Look-Once v5 (YOLOv5), and Detectron2 RCNN network and use that information along with DeepSORT method to track the position of the vessel. We obtain the bearing of the vessels detected from both infrared and visible image sequences of ship and fuse them using a Kalman filter (KF) and Rauch-Tung-Striebel (RTS) smoother. We then compare the result to the bearing obtained from the automatic identification system (AIS) of the vessel, and also compare it among object detectors. We obtained a root-mean-square error of 0.17∘ in vessel bearing tracking as compared to AIS.Description
Funding Information: We thank European Space Agency (ESA) for funding and Tallink Megastar crew for enabling the measurements onboard the vessel. We also thank Henrik Ramm-Schmidt from Fleetrange, Toni Hammarberg, and Martta-Kaisa Olkkonen from Finnish Geospatial Research Institute (FGI) for aiding in collecting the data from the campaign. Publisher Copyright: © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
Keywords
Bearing estimation, Maritime awareness, Ship detection
Other note
Citation
Gorad, A, Hassan, S S & Särkkä, S 2023, Vessel Bearing Estimation Using Visible and Thermal Imaging . in R Gade, M Felsberg & J-K Kämäräinen (eds), Image Analysis - 23rd Scandinavian Conference, SCIA 2023, Proceedings . Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13886 LNCS, Springer, pp. 373-381, Scandinavian Conference on Image Analysis, Kittilä, Finland, 18/04/2023 . https://doi.org/10.1007/978-3-031-31438-4_25