Banca de DEFESA: LUIZ FERNANDO VIRGINIO DA SILVA

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
DISCENTE : LUIZ FERNANDO VIRGINIO DA SILVA
DATA : 31/07/2017
HORA: 09:00
LOCAL: Auditório II do DIMAp
TÍTULO:

Extracting Traffic Data from Videos in Real-Time


PALAVRAS-CHAVES:

Traffic Monitoring, Image Processing, Computer Vision, Machine Learning, Detection, Tracking.


PÁGINAS: 65
GRANDE ÁREA: Ciências Exatas e da Terra
ÁREA: Ciência da Computação
SUBÁREA: Metodologia e Técnicas da Computação
ESPECIALIDADE: Processamento Gráfico (Graphics)
RESUMO:

Some of the major problems in large cities are related to traffic in public streets. Problems such as traffic jams and vehicle accidents directly impact society in a negative way, and are usually attributed to lack of urban planning from government, the lack of public policy or researches aimed at solving this problems, even if partially. These researches depend on data that must be collected in loco on the main avenues and streets of a city, demanding a big human force to collected them manually, also possibly incurring in error because of its manual characteristic. It is common to see CFTV systems around the city being used as the main mean of traffic monitoring. Thus, we devised a solution capable of collecting these data automatically in real-time, using videos captured using these cameras. The proposed solution consists of the following steps: (i) detect objects usign motion segmentation; (ii) apply pattern recognition methods such as machine learning mehtods or point descriptors to identify and classify vehicles. In this stage, we deal with the problem of occluded vehicles, e; (iii) track each object, now using Kalman filters or the mehtod of Senior in order to obtain relevant traffic data. Initially, we are collecting direction, velocity and flow intensity trhough volumetric counting, that can be used for other purposes, and not only the one handled here. Finally, we demostrate the efficiency of the proposed approach on experiments using videos acquired under different lightning conditions.

 


MEMBROS DA BANCA:
Presidente - 2177445 - BRUNO MOTTA DE CARVALHO
Interno - 1350250 - ANNE MAGALY DE PAULA CANUTO
Externo à Instituição - ALUISIO IGOR REGO FONTES - IFRN
Notícia cadastrada em: 25/07/2017 16:23
SIGAA | Superintendência de Informática - (84) 3215-3148 | Copyright © 2006-2017 - UFRN - sigaa05-producao.info.ufrn.br.sigaa05-producao