METHOD OF IDENTIFICATION OF SPACE IMAGES FOR FORECASTING FOREST FIRES

Abstract

Space forest fire monitoring methods help to detect them at an early stage and provide prompt decision-making, which facilitates further monitoring and evaluation of the consequences. The use of space data to monitor fire conditions allows you to quickly and economically obtain objective and independent information to make quick decisions to deal with the elements. The method of fire identification using low-resolution space images obtained from TERRA MODIS and NOAA AVHHR satellites is developed in the work. One of the problematic aspects in the method of determining fires is the masking of clouds and water. Therefore, in order to identify "fire" pixels, it is important to exclude from the analysis fragments of images that are covered with clouds and occupied by water bodies. The algorithm requires a significant increase in radiation in the range microns, as well as the relative observed radiation in the range of 11 microns. The algorithm examines each pixel of the scene, which as a result is assigned one of the following classes: missing data, cloud, water, potentially fire or undefined. Pixels of clouds and water bodies, determined by the technique of masking clouds and water bodies, belong to the classes of clouds and water, respectively. The fire detection algorithm examines only those pixels of the earth's surface that are classified as potentially fire or undefined. The method was implemented using the Power Builder visual programming tool in the Erdas Imaging remote sensing system. As a result of using the identification method, fire-hazardous places in the Chornobyl Exclusion Zone were identified. The use of satellite fire identification is important for the rapid detection of fires in remote forests that are poorly controlled by terrestrial monitoring methods.