Proposition of methodology for the extraction of cartographic features in digital media and high spatial resolutions aiming at updating cartographic products from urban areas (condominiums and lots)

• Grant number: 20/07357-5
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Effective date (Start): September 01, 2020
• Effective date (End): December 31, 2021
• Field of knowledge: Physical Sciences and Mathematics - Geosciences
• Principal Investigator: Erivaldo Antonio da Silva
• Grantee: Eduardo Soares Nascimento
• Host Institution: Faculdade de Ciências e Tecnologia (FCT). Universidade Estadual Paulista (UNESP). Campus de Presidente Prudente. Presidente Prudente , SP, Brazil
• Associated scholarship(s): 21/12584-3 - Development of methodology for the extraction of cartographic features aiming at updating urban areas using remote sensing images from medium and high spatial resolutions, BE.EP.IC

Abstract

Currently, even with the advancement of technologies, the updated representation by means of maps, is still seen as a major problem in a country that is known for its vast and unexplored territory, which is Brazil. Bearing in mind that representations of the territory, in an updated way, and that have quality, it is extremely important for decision making, in areas of deforestation, flooding, updating of roads, health, among other important areas. Cartography has these deficiencies, due to the constant change that the 'environment' lives, (such as road works, large constructions, natural disasters, among others). Knowing this problem, cartography works side by side with Remote Sensing, to obtain orbital images, which using the Digital Image Processing (PDI), the manipulation of images becomes possible. Knowing that there are several PDI techniques, Mathematical Morphology was chosen, which through it, it is possible to detect features of interest and extract it, thus having the result a future cartographic product. With MATLAB software and the application of PDI techniques, using Image Prossessing tooboxes, it is possible to detect and / or extract features of interest. Based on the assumption that a quality control is necessary, using the CARTOMORPH Software, the quality analysis of the cartographic product will be made, with the metrics called quality, correctness and completness. This project aims, in addition to the detection and/or extraction of cartographic features, to compare images from the same place, from different times, aiming at the changes suffered in the places of interest.