Fire Detection Based on Infrared Filter and Deep Learing

The use of video detection to detect fire has become an important research topic, which helps reduce personal and property losses. Detecting early fire helps prevent disaster, and the key factors for detecting fire are high accuracy, low false alarm rate, and fast response time. This paper uses infrared detection to improve the accuracy of fire detection, and at the same time uses convolutional neural networks to reduce the false alarm rate of detection. The detection performance of this model is better than some existing fire detection algorithms, and the algorithm model is deployed in the equipment, the function of real-time fire detection is realized.