no title

Abstract

 Wildfires pose a serious threat to biodiversity, the envi
ronment, and human communities. An automated system 
that can forecast forest fires using satellite data would pro
vide crucial early warning for disaster mitigation and preven
tion. Early detection of wildfires is critical to avoid significant 
human, economic and ecological losses. However, traditio
nal detection methods can be slow or inefficient, resulting 
in delayed responses and extensive damage. In the present 
research, a predictive intelligent automaton model for fire 
early warning using satellite information is proposed. The 
methodology used to develop this project is an adaptation 
of the extreme programming methodologies implemented 
in two stages. In the first stage, the following phases are 
carried out in a cyclic manner: Phase one: data search and preprocessing, Phase two: design of the intelligent system. 
Phase three: coding, Phase four: testing. Subsequently, in 
the second half, the launching phase is implemented. The 
data were taken from the Tutiempo Network database that 
provides information on atmospheric data synchronized 
with the Corponor database that provides data for a conti
nuous period of thirty days. An acceptable 92.75% accuracy 
rate was obtained.