Flame Recognition Algorithm Research under Complex Background

The paper presents a flame recognition algorithm under complex background. The light changes are mainly studied on the impact of the flame recognition system. An adaptive brightness threshold method which aimed at improving the speed and efficiency of the algorithm is proposed. The static and dynamic flame features are analyzed and the probability model of the flame detection is established based on the trust model. The Suspected probability is used to reflect the video image to appear the probability of fire. The experimental results show that the algorithm remains with stronger adaptive and anti-jamming ability for the white shadow interference of moving objects, static white shadow interference, street lamps, car lamps and lighting changes effects and so on, the suspected probability can always be lower than 5%. Therefore, the algorithm in practice has a broad application prospects.