The Flame Detection and Background Interference Filtering Algorithm Based on Video Images

This paper gives a new flame detection processing algorithm based on video images by using the threshold and binarization processing to capture the suspicious regions from flame images precisely, and proposes an improved frame-difference method to filter the interference of rapidly moving objects with high brightness. The flame is then detected through using the suspicious region area and circular degree to filter the background interference with fixed high brightness. Based on experiments, it is indicated that this proposed algorithm can filter the background interference effectively and detect the flame accurately.