An Improved Algorithm Based on D-S Evidence Theory for Accurate Wildfire Detection

Accurately predicting wildfire before they occur is a challenging task. D-S evidence theory can make use of collected data to predict wildfire. Due to the possible conflict of the collected data, the accuracy of the theory is not high in predicting the mountain fire. In order to solve this problem, this paper improved the traditional D-S evidence theory, combined with the theory of hierarchical analysis to reduce the conflict between data, and finally improve the accuracy of mountain fire prediction. In this paper, the accuracy rate of risk prediction and cost expenditure are used as the measurement basis for accurately predicting mountain fires. When the accuracy of risk prediction is higher, the accuracy of mountain fire prediction algorithm is higher. The lower the cost used, the higher the accuracy of the fire prediction algorithm. The risk prediction accuracy of the improved D-S evidence theory is 1, which is 40% higher than the traditional accuracy. The cost of the improved D-S evidence theory is 11 units less than that of the traditional theory. Therefore, the improved D-S evidence theory algorithm can predict the mountain fire more accurately.