A Real-time Fire Detection Model Based on Cascade Strategy

Jing Wu, Chunxue Wu, Yan Wu
2018 International Journal of Software & Hardware Research in Engineering  
This paper aims to detect and discover fires through video recognition technology to minimize the harm caused by fires. In this paper, a fire detection model, which is as fast as possible and can be applied to real-time and effective detection of smoke and flame in video surveillance, is present. Firstly, the model locates the dynamic foreground object in the frame of the video, and obtains the areas where smoke or flame may exist, which greatly reduces the number of Regions of Interest (RoIs).
more » ... Next, this model uses the cascade method to apply the color histogram filtering model combined with the SVM and the neural network object recognition model to the fire detection of RoIs. The color histogram filtering model combined with the SVM further reduces the number of RoIs and greatly shortens the recognition time of the neural network model. Finally, considering the continuity character of video sequence image, it is not necessary to detect all video frames, specially, that is, no matter how much the video frame rate is, the number of video frames detected per second in this model will be limited to 10 frames. The model can accurately detect the flame and smoke in the video in real time, and has a strong antiinterference ability. The model in this paper shows good performance in both detection accuracy and speed.
doi:10.26821/ijshre.6.11.2018.61107 fatcat:if5l3modknec3inuyobwv275hy