FOREST FIRE DETECTION METHOD BASED ON UAV IMAGE ACQUISITION AND PROCESSING

摘要

With the popularization of the information technology and the continuous innovation and devel-opment of high-tech products in China, drones have become a key direction. The application of drones has involved many industries. The application of drones can not only greatly reduce labor costs but can also improve the efficiency of intelligent infor-mation services and greatly enhance the effective-ness of information monitoring and UAV intelligent applications. Based on drone image acquisition and information processing technology, the detection of forest fire prevention and forest fire prevention has effectively reduced the probability of forest fires. According to the types of UAVimage acquisition, it is divided into different types of UAV operations, which are divided into: long-endurance UAV, height 18000m, activity radius 2000km, endurance greater than 24h; medium-range UAV, height 10,000m, ac-tivity radius 700 -1000km, endurance 24h; short-range UAV, height 8000m, active radius 150-350km, endurance 24h; very short-range UAV, altitude 3000m, active radius 1-50km, endurance 24h. We analyze the effectiveness of four different UAV im-age acquisition and processing methods to deal with forest fire detection. Based on the specific photos of the four types ofUAV monitoring forest fires and the results of empirical analysis, the effective value of UAV image collection and processing functions for forest fire detection and the significance of the eval-uation effect are verified. The four different UAV im-age acquisition results and processing results are ob-viously different. Based on the clarity and effective-ness of the image acquisition, the monitoring data is mainly used to construct a comparative analysis and build a model analysis to verify the gain effect of UAV image acquisition and processing on forest fire detection.