| Grant number: | 16/23763-8 |
| Support type: | Scholarships in Brazil - Doctorate |
| Effective date (Start): | June 01, 2017 |
| Effective date (End): | November 30, 2020 |
| Field of knowledge: | Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques |
| Cooperation agreement: | Coordination of Improvement of Higher Education Personnel (CAPES) |
| Principal Investigator: | Odemir Martinez Bruno |
| Grantee: | Lucas Correia Ribas |
| Home Institution: | Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil |
| Associated scholarship(s): | 19/03277-0 - Pattern recognition in complex networks using distance transform, BE.EP.DR |
Abstract Complex networks have been used as a study tool and contributed in several areas of science due to their interdisciplinary nature and conceptual simplicity. In computer science, networks have been used in problems of architecture, artificial intelligence, computer networks and computer vision. With regard to computational vision, several approaches have been proposed over the past decade to pattern recognition based on complex networks. This fact is motivated by the nonlinear nature that many images present, making the use of complex networks a potential tool as a measure of complexity. Most of these approaches were proposed by researchers from USP of São Carlos. The results show great potential in the use of complex networks as a tool for the development of new methods for computer vision. The objective of this doctorate is to study and develop new methods based on complex networks for computational vision. In particular, two problems of computational vision that are the specialty of the research group and that has been studied for more than a decade will be investigated: texture and shape analysis. In this way, we intend to investigate new methodologies for the modeling of images/videos in networks and new methods of network analysis. In particular, we will investigate the use of automata and deep learning as new analysis tools. In addition, existing approaches will be studied in order to identify possible limitations and when possible, improvements will be proposed. As a way of analyzing the potential of the approaches developed, real problems of biology and nanotechnology in which the researcher's research group is working will be considered for application. In this way, this project presents a great possibility of contribution with new methods for the area of computational vision and, consequently, for the areas of biology and nanotechnology. (AU) | |