Advanced search
Start date

Watching the cities: big data and computer vision for analysis of urban environments

Grant number: 15/03475-5
Support type:Scholarships in Brazil - Scientific Initiation
Effective date (Start): April 01, 2015
Effective date (End): June 30, 2017
Field of knowledge:Physical Sciences and Mathematics - Computer Science
Principal Investigator:Roberto Marcondes Cesar Junior
Grantee:Gabriel Augusto Bianchi Azevedo Ferreira
Home Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil
Associated research grant:11/50761-2 - Models and methods of e-Science for life and agricultural sciences, AP.TEM
Associated scholarship(s):16/12667-8 - Data-driven city model representation and analysis using visual and non-visual information, BE.EP.IC


This work aims to develop and to apply methods for studying the urban environment through video. This study will be divided in two parts (or steps), related to the analysis of urban scenes.In the first part, we plan to constitute a database, which will gather information, in the form of videos and metadata, about urban environments. The first step includes designing the database, developing robots to capture data, along with using models and algorithms to assure the quality of the collected information.The second part will deal with data mining, which includes the interpretation of that huge amount of data obtained on the first step. The image analysis will focus on detecting three main elements: i) vehicles, ii) buildings and iii) pedestrians. For this purpose, we intend to use computer vision methods and algorithms, along with pattern recognition techniques. Most of the studies regarding urban environments and video cameras make use of these aforementioned basic elements. Therefore, there is a natural interest in detecting them.