Bee2Fire: A deep learning powered forest fire detection system

Rui Valente de Almeida; Fernando Crivellaro; Maria Narciso; Ana Isabel Sousa; Pedro Vieira

doi:10.5220/0008966106030609

Bee2Fire: A deep learning powered forest fire detection system

Rui Valente de Almeida, Fernando Crivellaro, Maria Narciso, Ana Isabel Sousa, Pedro Vieira

DF – Departamento de Física

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

3 Citations (Scopus)

Abstract

Bee2Fire is a commercial system for forest fire detection, inheriting from the Forest Fire Finder System. Designed in Portugal, it aims to address one of Southern Europe's main concern, forest fires. It is a well known fact that the sooner a wildfire is detected, the quicker it can be put out, which highlights the importance of early detection. By scanning the landscape using regular cameras and Deep Artificial Neural Networks, Bee2Fire searches for smoke columns above the horizon with a image classification approach. After these networks were trained, the system was deployed in the field, obtaining a sensitivity score between 74% and 93%, a specificity of more than 99% and a precision of around 82%.

Original language	English
Title of host publication	ICAART 2020 - Proceedings of the 12th International Conference on Agents and Artificial Intelligence
Editors	Ana Rocha, Luc Steels, Jaap van den Herik
Publisher	SciTePress
Pages	603-609
Number of pages	7
Volume	2
ISBN (Electronic)	9789897583957
DOIs	https://doi.org/10.5220/0008966106030609
Publication status	Published - 2020
Event	12th International Conference on Agents and Artificial Intelligence, ICAART 2020 - Valletta, Malta Duration: 22 Feb 2020 → 24 Feb 2020

Conference

Conference	12th International Conference on Agents and Artificial Intelligence, ICAART 2020
Country	Malta
City	Valletta
Period	22/02/20 → 24/02/20

Keywords

Deep Learning
FastAI
Forest Fire Detection
IBM Watson
PyTorch

Access to Document

10.5220/0008966106030609

Cite this

de Almeida, R. V., Crivellaro, F., Narciso, M., Sousa, A. I., & Vieira, P. (2020). Bee2Fire: A deep learning powered forest fire detection system. In A. Rocha, L. Steels, & J. van den Herik (Eds.), ICAART 2020 - Proceedings of the 12th International Conference on Agents and Artificial Intelligence (Vol. 2, pp. 603-609). SciTePress. https://doi.org/10.5220/0008966106030609

@inproceedings{32ab2437ed594138a41dc9d3f3f8d165,

title = "Bee2Fire: A deep learning powered forest fire detection system",

abstract = "Bee2Fire is a commercial system for forest fire detection, inheriting from the Forest Fire Finder System. Designed in Portugal, it aims to address one of Southern Europe's main concern, forest fires. It is a well known fact that the sooner a wildfire is detected, the quicker it can be put out, which highlights the importance of early detection. By scanning the landscape using regular cameras and Deep Artificial Neural Networks, Bee2Fire searches for smoke columns above the horizon with a image classification approach. After these networks were trained, the system was deployed in the field, obtaining a sensitivity score between 74% and 93%, a specificity of more than 99% and a precision of around 82%.",

keywords = "Deep Learning, FastAI, Forest Fire Detection, IBM Watson, PyTorch",

author = "{de Almeida}, {Rui Valente} and Fernando Crivellaro and Maria Narciso and Sousa, {Ana Isabel} and Pedro Vieira",

note = "Publisher Copyright: {\textcopyright} 2020 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved Copyright: Copyright 2020 Elsevier B.V., All rights reserved.; 12th International Conference on Agents and Artificial Intelligence, ICAART 2020 ; Conference date: 22-02-2020 Through 24-02-2020",

year = "2020",

doi = "10.5220/0008966106030609",

language = "English",

volume = "2",

pages = "603--609",

editor = "Ana Rocha and Luc Steels and {van den Herik}, Jaap",

booktitle = "ICAART 2020 - Proceedings of the 12th International Conference on Agents and Artificial Intelligence",

publisher = "SciTePress",

}

de Almeida, RV, Crivellaro, F , Narciso, M , Sousa, AI & Vieira, P 2020, Bee2Fire: A deep learning powered forest fire detection system. in A Rocha, L Steels & J van den Herik (eds), ICAART 2020 - Proceedings of the 12th International Conference on Agents and Artificial Intelligence. vol. 2, SciTePress, pp. 603-609, 12th International Conference on Agents and Artificial Intelligence, ICAART 2020, Valletta, Malta, 22/02/20. https://doi.org/10.5220/0008966106030609

Bee2Fire: A deep learning powered forest fire detection system. / de Almeida, Rui Valente; Crivellaro, Fernando ; Narciso, Maria ; Sousa, Ana Isabel ; Vieira, Pedro.

ICAART 2020 - Proceedings of the 12th International Conference on Agents and Artificial Intelligence. ed. / Ana Rocha; Luc Steels; Jaap van den Herik. Vol. 2 SciTePress, 2020. p. 603-609.

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

TY - GEN

T1 - Bee2Fire: A deep learning powered forest fire detection system

AU - de Almeida, Rui Valente

AU - Crivellaro, Fernando

AU - Narciso, Maria

AU - Sousa, Ana Isabel

AU - Vieira, Pedro

PY - 2020

Y1 - 2020

N2 - Bee2Fire is a commercial system for forest fire detection, inheriting from the Forest Fire Finder System. Designed in Portugal, it aims to address one of Southern Europe's main concern, forest fires. It is a well known fact that the sooner a wildfire is detected, the quicker it can be put out, which highlights the importance of early detection. By scanning the landscape using regular cameras and Deep Artificial Neural Networks, Bee2Fire searches for smoke columns above the horizon with a image classification approach. After these networks were trained, the system was deployed in the field, obtaining a sensitivity score between 74% and 93%, a specificity of more than 99% and a precision of around 82%.

AB - Bee2Fire is a commercial system for forest fire detection, inheriting from the Forest Fire Finder System. Designed in Portugal, it aims to address one of Southern Europe's main concern, forest fires. It is a well known fact that the sooner a wildfire is detected, the quicker it can be put out, which highlights the importance of early detection. By scanning the landscape using regular cameras and Deep Artificial Neural Networks, Bee2Fire searches for smoke columns above the horizon with a image classification approach. After these networks were trained, the system was deployed in the field, obtaining a sensitivity score between 74% and 93%, a specificity of more than 99% and a precision of around 82%.

KW - Deep Learning

KW - FastAI

KW - Forest Fire Detection

KW - IBM Watson

KW - PyTorch

UR - http://www.scopus.com/inward/record.url?scp=85083117544&partnerID=8YFLogxK

U2 - 10.5220/0008966106030609

DO - 10.5220/0008966106030609

M3 - Conference contribution

AN - SCOPUS:85083117544

VL - 2

SP - 603

EP - 609

BT - ICAART 2020 - Proceedings of the 12th International Conference on Agents and Artificial Intelligence

A2 - Rocha, Ana

A2 - Steels, Luc

A2 - van den Herik, Jaap

PB - SciTePress

T2 - 12th International Conference on Agents and Artificial Intelligence, ICAART 2020

Y2 - 22 February 2020 through 24 February 2020

ER -

Bee2Fire: A deep learning powered forest fire detection system

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this