MERFire: towards label-free marine engine room fire detection and tracking

Hanlin Li, Ziwu Zheng and Jinbiao Lin

Cyber-Physical Systems, 2025, vol. 11, issue 4, 446-470

Abstract: Fire detection in marine engine rooms is crucial for safety. Traditional methods using deep learning, like YOLO models, face challenges such as needing large datasets and only analyzing single frames, leading to potential errors. Our proposed method combines background subtraction with the Segment Anything (SAM) model and uses a hidden Markov model (HMM) to filter out noise over time. This hybrid approach effectively detects fire across video frames. We validate its effectiveness on both a simulated marine engine room dataset and a public wildfire dataset, showcasing its ability to generalize in different environments.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/23335777.2025.2458877 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:tcybxx:v:11:y:2025:i:4:p:446-470

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/tcyb20

DOI: 10.1080/23335777.2025.2458877

Access Statistics for this article

Cyber-Physical Systems is currently edited by Yang Xiao

More articles in Cyber-Physical Systems from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().