Sandra Paola Hernández-López
◽
Juan Israel Yañez-Vargas
◽
Andrea Gonzalez-Ramirez
◽
Deni Torres-Roman
The increase in the increase in wildfires throughout the world is largely due to increases in temperature and even to an increase in the carelessness of the population in leaving a large amount of the garbage in forests. Using Python and Matlab programs were as working medium. We performed the preprocessing on multispectral images obtained by the Landsat 8 satellite with and without wildfires, which consists of three steps: alignment, characterization and normalization, with the intention of standardization the images. From obtaining the spectral signatures of wildfires and metallic structures, boxes and whiskers diagrams, Shannon entropy and mutual information from the images, there are similar behavior in bands 6, 7, 8, 10 and 11, with more relevant information, taking into account that each image is formed by 11 bands, and in bands 1, 2, 3, 4, 5, 8 and 9 there is less information, SVD decomposition allows to have the best k-rank approximation to the original data matrix. The purpose of this analysis is to reduce the computational complexity.