Colour based fire detection method with temporal intensity variation filtration

2014 Joint IMEKO TC1-TC7-TC13 Symposium: Measurement Science Behind Safety and Security

Trambitckii, K.^1 ; Anding, K.^2 ; Musalimov, V.^1 ; Linß, G.^2

Faculty of Fine Mechanics and Technology, Department of Mechatronics, National Research University of Information Technologies, Saint-Petersburg, Russia^1

Faculty of Mechanical Engineering, Department of Quality Assurance and Industrial Image Processing, Ilmenau University of Technology, Ilmenau, Germany^2

关键词: Colour informations;    Computing technology;    Fire detection;    Fire detection algorithms;    Fire-detection method;    Intensity variations;    Shooting conditions;    Video information;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/588/1/012038/pdf
DOI  :  10.1088/1742-6596/588/1/012038

来源: IOP

【 摘 要 】

Development of video, computing technologies and computer vision gives a possibility of automatic fire detection on video information. Under that project different algorithms was implemented to find more efficient way of fire detection. In that article colour based fire detection algorithm is described. But it is not enough to use only colour information to detect fire properly. The main reason of this is that in the shooting conditions may be a lot of things having colour similar to fire. A temporary intensity variation of pixels is used to separate them from the fire. These variations are averaged over the series of several frames. This algorithm shows robust work and was realised as a computer program by using of the OpenCV library.