Autonomous flame detection in video based on saliency analysis and optical flow

Zhenglin Li; Olga Isupova; Lyudmila Mihaylova; Lucile Rossi

doi:10.1109/MFI.2016.7849492

Autonomous flame detection in video based on saliency analysis and optical flow

Zhenglin Li
, Olga Isupova
, Lyudmila Mihaylova
, Lucile Rossi

Department of Computer Science

University of Corsica
University of Sheffield

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

4 Citations (SciVal)

Abstract

The paper proposes a flame detection method based on saliency analysis, optical flow estimation and temporal wavelet transform. Two separate saliency maps are first obtained based on the grayscale values and optical flow magnitudes of each frame using a saliency detector. Subsequently, the two maps are combined to extract candidate flame regions. To further discard falsely detected pixels, a colour model of flames and temporal wavelet transform are employed. The proposed algorithms can be applied in the autonomous and semi-autonomous systems for environmental surveillance and can reduce the load of human operators. Experiments illustrate the introduced method achieves around 91% true positive rate and 97% true negative rate.

Original language	English
Title of host publication	2016 IEEE lnternational Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2016
Publisher	IEEE
Pages	218-223
Number of pages	6
ISBN (Electronic)	978-1-4673-9708-7
ISBN (Print)	978-1-4673-9709-4
DOIs	https://doi.org/10.1109/MFI.2016.7849492
Publication status	Published - 13 Feb 2017
Event	2016 IEEE lnternational Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2016 - Baden-Baden, Germany Duration: 19 Sept 2016 → 21 Sept 2016

Publication series

Name	IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems
Volume	0

Conference

Conference	2016 IEEE lnternational Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2016
Country/Territory	Germany
City	Baden-Baden
Period	19/09/16 → 21/09/16

Access to Document

10.1109/MFI.2016.7849492

Cite this

Li, Z., Isupova, O., Mihaylova, L., & Rossi, L. (2017). Autonomous flame detection in video based on saliency analysis and optical flow. In 2016 IEEE lnternational Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2016 (pp. 218-223). (IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems; Vol. 0). IEEE. https://doi.org/10.1109/MFI.2016.7849492

Autonomous flame detection in video based on saliency analysis and optical flow. / Li, Zhenglin; Isupova, Olga; Mihaylova, Lyudmila et al.
2016 IEEE lnternational Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2016. IEEE, 2017. p. 218-223 (IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems; Vol. 0).

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

Li, Z, Isupova, O, Mihaylova, L & Rossi, L 2017, Autonomous flame detection in video based on saliency analysis and optical flow. in 2016 IEEE lnternational Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2016. IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, vol. 0, IEEE, pp. 218-223, 2016 IEEE lnternational Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2016, Baden-Baden, Germany, 19/09/16. https://doi.org/10.1109/MFI.2016.7849492

Li Z, Isupova O, Mihaylova L, Rossi L. Autonomous flame detection in video based on saliency analysis and optical flow. In 2016 IEEE lnternational Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2016. IEEE. 2017. p. 218-223. (IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems). doi: 10.1109/MFI.2016.7849492

@inproceedings{29751fa7a7994b5ca63ace1222c84cc8,

title = "Autonomous flame detection in video based on saliency analysis and optical flow",

abstract = "The paper proposes a flame detection method based on saliency analysis, optical flow estimation and temporal wavelet transform. Two separate saliency maps are first obtained based on the grayscale values and optical flow magnitudes of each frame using a saliency detector. Subsequently, the two maps are combined to extract candidate flame regions. To further discard falsely detected pixels, a colour model of flames and temporal wavelet transform are employed. The proposed algorithms can be applied in the autonomous and semi-autonomous systems for environmental surveillance and can reduce the load of human operators. Experiments illustrate the introduced method achieves around 91\% true positive rate and 97\% true negative rate.",

author = "Zhenglin Li and Olga Isupova and Lyudmila Mihaylova and Lucile Rossi",

year = "2017",

month = feb,

day = "13",

doi = "10.1109/MFI.2016.7849492",

language = "English",

isbn = "978-1-4673-9709-4",

series = "IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems",

publisher = "IEEE",

pages = "218--223",

booktitle = "2016 IEEE lnternational Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2016",

address = "USA United States",

note = "2016 IEEE lnternational Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2016 ; Conference date: 19-09-2016 Through 21-09-2016",

}

TY - GEN

T1 - Autonomous flame detection in video based on saliency analysis and optical flow

AU - Li, Zhenglin

AU - Isupova, Olga

AU - Mihaylova, Lyudmila

AU - Rossi, Lucile

PY - 2017/2/13

Y1 - 2017/2/13

N2 - The paper proposes a flame detection method based on saliency analysis, optical flow estimation and temporal wavelet transform. Two separate saliency maps are first obtained based on the grayscale values and optical flow magnitudes of each frame using a saliency detector. Subsequently, the two maps are combined to extract candidate flame regions. To further discard falsely detected pixels, a colour model of flames and temporal wavelet transform are employed. The proposed algorithms can be applied in the autonomous and semi-autonomous systems for environmental surveillance and can reduce the load of human operators. Experiments illustrate the introduced method achieves around 91% true positive rate and 97% true negative rate.

AB - The paper proposes a flame detection method based on saliency analysis, optical flow estimation and temporal wavelet transform. Two separate saliency maps are first obtained based on the grayscale values and optical flow magnitudes of each frame using a saliency detector. Subsequently, the two maps are combined to extract candidate flame regions. To further discard falsely detected pixels, a colour model of flames and temporal wavelet transform are employed. The proposed algorithms can be applied in the autonomous and semi-autonomous systems for environmental surveillance and can reduce the load of human operators. Experiments illustrate the introduced method achieves around 91% true positive rate and 97% true negative rate.

UR - https://www.scopus.com/pages/publications/85015174718

U2 - 10.1109/MFI.2016.7849492

DO - 10.1109/MFI.2016.7849492

M3 - Chapter in a published conference proceeding

AN - SCOPUS:85015174718

SN - 978-1-4673-9709-4

T3 - IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems

SP - 218

EP - 223

BT - 2016 IEEE lnternational Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2016

PB - IEEE

T2 - 2016 IEEE lnternational Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2016

Y2 - 19 September 2016 through 21 September 2016

ER -

Autonomous flame detection in video based on saliency analysis and optical flow

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this