CBERS-4A IMAGERY FOR MAPPING URBAN LAND COVER IN THE AMAZON

The primary method for collecting information about the Earth's surface in recent decades, notably for developing nations, has been remote sensing. Despite this, Amazonian cities lack databases and cartographic publications. Considering Santarém as the study site, this paper proposes to create a classification model for mapping the land cover of an Amazonian city. Using imagery from the CBERS-4A satellite's WPM sensor, we created a classification model that combines the Geographic Object-Based Image Analysis (GEOBIA) method, data mining strategies, and the Random Forest machine learning algorithm. The results are promising in discerning different intra-urban cover classes, with an overall accuracy level in the validation samples of over 98%.