Application of inpanting techniques and segmentation removal in digital images of partially detected cartographic features

• Grant number: 20/07709-9
• Support type: Scholarships in Brazil - Scientific Initiation
• Effective date (Start): October 01, 2020
• Effective date (End): September 30, 2021
• Field of knowledge: Physical Sciences and Mathematics - Geosciences
• Principal researcher: Erivaldo Antonio da Silva
• Grantee: Allan Alves Lopes Ferreira
• Home Institution: Faculdade de Ciências e Tecnologia (FCT). Universidade Estadual Paulista (UNESP). Campus de Presidente Prudente. Presidente Prudente , SP, Brazil

Abstract

This project consists of the study and application of techniques for the reconstruction of digital images and the reduction of excessive segmentation in partially detected cartographic features resulting from the application of morphological routines, aiming to make this process fast and accurate. For this, two inpainting techniques were selected that will be used to facilitate the reconstruction of partially detected features along with a segmentation technique proposed by the fellow. Initially, test areas of cartographic features will be selected whose topology is the result of anthropic actions, such as roads, roads and streets in urban subdivisions in order to select samples that represent the region of interest to be worked. From that, the inpainting techniques that will be studied and applied are based on the inpainting techniques of Criminisi et al [1] and Huang et al [2] and that can be used in binary digital images, in levels of gray and color ( RGB). In parallel with the execution of the inpainting techniques, a technique will be developed by the fellow to remove excessive segmentation from the use of Mathematical Morphology, which in his initial idea, will be based on the Watershed segmentation technique. Such a project is of great importance, because in a country with vast territory there is a need to create and keep cartographic products updated at different scales, where the results of this project can contribute to the cartographic area with the improvement in the results obtained in processes extraction of cartographic features from morphological routines. (AU)