Fire detection system and method basedon visual data

The present invention relates to a vision-based fire detection system and method for detecting a fire zone from the image information. The fire detection system, motion detection module outputs by updating the previous frame background to reflect the current frame after initializing the gray image of the previous frame as a background; Color modeling module for detecting the first fire candidate regions by applying the color model for the frame to the output of the motion detection module; Luminance map processing module for detecting the light a second fire candidate region relative to the surrounding by applying the luminance map for the current frame; It characterized in that it comprises the first and second fire candidate region SVM processing module for using the SVM (Support Vector Machine) to extract the resulting fire detection area according to the variation model of the temporal wavelet transform energy for. Smoke detection, video, color modeling, motion detection, luminance maps, SVM (Support Vector Machine)