| Grant number: | 16/04553-2 |
| Support type: | Scholarships abroad - Research Internship - Doctorate |
| Effective date (Start): | February 01, 2017 |
| Effective date (End): | January 31, 2018 |
| Field of knowledge: | Physical Sciences and Mathematics - Geosciences |
| Principal Investigator: | Erivaldo Antonio da Silva |
| Grantee: | Guilherme Pina Cardim |
| Supervisor abroad: | Ignacio Bravo Muñoz |
| Home Institution: | Faculdade de Ciências e Tecnologia (FCT). Universidade Estadual Paulista (UNESP). Campus de Presidente Prudente. Presidente Prudente , SP, Brazil |
| Local de pesquisa : | Universidad de Alcalá (UAH), Spain |
| Associated to the scholarship: | 14/24392-8 - Proposal of co-design platform for processing remote sensing images, BP.DR |
Abstract The digital image processing (DIP) consists of an area of great interest in studies, which usually requires a large amount of information processing. In Cartography, the DIP is widely used in remote sensing studies to detect interest features in the images. Nowadays, with great technological advances, the remote sensing images have high spatial resolution with large amount of information to be processed during the studies developed using these images. Therefore, the processing of high spatial resolution images needs high time to complete all the processing. For that reason, several libraries of digital image processing have emerged in the literature to facilitate the use and improve the performance of the DIP functions, such as OpenCV, DipImage and CARTOMORPH. Among the mentioned libraries, it is worth mentioning the importance of the CARTOMORPH library for the Cartography area, since it was developed to studies of cartographic feature detection and it is focused on mathematical morphology techniques, which are highlighting in related studies. However, sometimes the processing of cartographic images of high resolution, using DIP libraries, is still slow because of the large amount of information to be processed and the complexity of some DIP algorithms. Therefore, the current project proposes a methodology to analyze routines of cartographic features detection, developed with the CARTOMORPH library, split them and control the flow of operations. After the routine division, the proposal continues with the possibility of sending part of the processing to a hardware platform, which will be defined during the project development and will be responsible for the processing that requires high computational cost. This way, there is a proposal for a co-designed system combining hardware and software for processing of high resolution images, from remote sensors, for achievement of expected results in a shorter processing time. The project is justified by the need of studies to enhance the performance and reduce the waiting time of researchers, in the Cartography area, developing computational routines to detect cartographic features in remote sensing images. (AU) | |