TY - GEN
T1 - The research of real-time forest fire alarm algorithm based on video
AU - Song, Lu
AU - Wang, Bo
AU - Zhou, Zhiqiang
AU - Wang, Hailuo
AU - Wu, Shujie
N1 - Publisher Copyright:
© 2014 IEEE.
PY - 2014/10/7
Y1 - 2014/10/7
N2 - Forest fire has brought incalculable economic losses due to its strong abruptness and huge destructiveness. Forest real-time monitoring and fire alarm can effectively reduce the loss of life and property, but the traditional detection devices are not widely used for its limited detection range. This paper proposes a real-time forest fire alarm algorithms based on video. Extracting the moving region on the basis of background modeling, and then we use the mixed color space feature analysis to detect the flame, and use adaptive threshold segmentation and color moments analysis to detect smoke. The experiments has shown that our algorithms have both a higher detection rate and a lower false alarm rate, and the average processing speed reaches 40 ms/frame, and the detection of flame and smoke under different environments and lighting conditions can also work accurately and quickly.
AB - Forest fire has brought incalculable economic losses due to its strong abruptness and huge destructiveness. Forest real-time monitoring and fire alarm can effectively reduce the loss of life and property, but the traditional detection devices are not widely used for its limited detection range. This paper proposes a real-time forest fire alarm algorithms based on video. Extracting the moving region on the basis of background modeling, and then we use the mixed color space feature analysis to detect the flame, and use adaptive threshold segmentation and color moments analysis to detect smoke. The experiments has shown that our algorithms have both a higher detection rate and a lower false alarm rate, and the average processing speed reaches 40 ms/frame, and the detection of flame and smoke under different environments and lighting conditions can also work accurately and quickly.
KW - Background modeling
KW - Color moments
KW - adaptive threshold segmentation
KW - mixed color space feature
UR - http://www.scopus.com/inward/record.url?scp=84910104999&partnerID=8YFLogxK
U2 - 10.1109/IHMSC.2014.34
DO - 10.1109/IHMSC.2014.34
M3 - Conference contribution
AN - SCOPUS:84910104999
T3 - Proceedings - 2014 6th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2014
SP - 106
EP - 109
BT - Proceedings - 2014 6th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2014
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2014 6th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2014
Y2 - 26 August 2014 through 27 August 2014
ER -