Satellite images provide us with information of vital importance, in order to analyze large tracts of land, so their analysis has a degree of complexity characterized by the weight of the image and its size, analyzing large tracts of land originates analyzing the image in all its extension, there are now intelligent algorithms capable of classifying the images, these can analyze the images causing a decrease in the analysis time and improving the result of the analysis of the images. The vegetal cover in our planet is suffering great changes produced by phenomena caused by man, by effects of deforestation, illegal mining among others, that are originating great changes in the terrestrial cover, the evaluation of these changes can be realized by the analysis of satellite images with which you can classify and then locate the area, for this purpose the chromatic characteristics of the images are analyzed with the help of artificial intelligence techniques. In this work, the chromatic characteristics of an image dataset are analyzed. They correspond areas that belong to vegetal cover and areas that do not correspond to the vegetal cover, with the intention of analyzing if these two classes are linearly separable.