Forest fire point secondary screening method and system

Abstract

The embodiments of the invention disclose a forest fire point secondary screening method and a system, which are used to perform secondary screening on satellite monitoring hotspots, reduce misjudgment in satellite-based fire point monitoring, and enhance the accuracy of transmission line forest fire warning. The method disclosed by one embodiment of the invention comprises the following steps: S1, acquiring historical satellite hotspot information as sample data, and performing characteristic quantity extraction on sample data under both conditions of real fire points and false fire points to get extracted characteristic quantity; S2, normalizing the extracted characteristic quantity, and constructing an input space vector according to the normalized characteristic quantity; S3, building an SVM classification model with the input space vector as input and the result of real fire points or false fire points corresponding to the input space vector as output, and training the SVM classification model; and S4, inputting satellite remote sensing hotspot information according to the trained SVM classification model to get a fire point screening output result.

Description

technical field [0001] The invention relates to the technical field of mountain fire identification, in particular to a method and system for secondary identification of mountain fire points. Background technique [0002] Wildfire is one of the important factors affecting the safe and stable operation of the power grid. Wildfire near the transmission line corridor can easily cause line tripping, and it is difficult to successfully reclose after a fault occurs, affecting the reliability of power supply. Using satellites to monitor the occurrence of wildfires can detect wildfires on transmission lines in time and provide guidance for emergency response. [0003] However, after using satellite remote sensing technology to obtain hotspot information, a second screening of hotspots is required. First of all, due to factors such as clouds, flares, and urban heat sources, satellite hotspots are not necessarily all real fires. Secondly, the algorithm for extracting fire points is ...

AI Technical Summary

Problems solved by technology

The existing fire point secondary identification method generally adopts the expert scoring method, which largely relies on expert experience, and it is difficult to continuously adjust according to different regions and different climates, and cannot meet the accuracy requirements of existing mountain fire monitoring

Images