12.07.2015 Views

III WVC 2007 - Iris.sel.eesc.sc.usp.br - USP

III WVC 2007 - Iris.sel.eesc.sc.usp.br - USP

III WVC 2007 - Iris.sel.eesc.sc.usp.br - USP

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

<strong>WVC</strong>'<strong>2007</strong> - <strong>III</strong> Workshop de Visão Computacional, 22 a 24 de Outu<strong>br</strong>o de <strong>2007</strong>, São José do Rio Preto, SP.If no axles are detected then we assume that the detectedobject is not a vehicle, so it is removed from theacquired vehicles set database.7. Results[8] S. O. Rezende. Sistem as Inteligentes - Fundam entos eAplicações.Manule, São Paulo, 1 edition, 2002.[9] P. Viola and M. J. Jones. Rapid object detection using aboosted ca<strong>sc</strong>ade of simple features. IEEE CVPR, 2001.Our first preliminary evaluating tests indicate thatthe proposed system has a good precision and has greatvalue. Tests were made with captured videos with a notsignificant number of vehicle to validate this system asit should be.As a next-step action, the system should be firedto a set of massive tests, that will make it possible tosee what can be done to decrease mistakes done by theclassifier and increasing precision, as well as the systemreliability.These conclusive tests should be presented in a futurework, since a validation of a system like the proposedin this work requires more time, because the systemshould be evaluated under a wide range of trafficconditions - light traffic, congestion, varying speeds,varying weather conditions and different lanes.8. ConclusionAiming at present a good solution to traffic monitoring,a vision-based system offers advantages like goodaccuracy and low cost.The results show that the classification precision ishigh and the instrument has great value to be widelyused.A validation of the proposed system is still requiredand will be presented in a posterior work, comparingwith others solution in same area.References[1] G. Bradski, A. Kaehler, and V. Pisarevsky. Learningbasedcomputer vision with intel’s open source computervision li<strong>br</strong>ary. IntelTechonology Journal, 9:14, 2005.[2] J. Canny. A computational approach to edge detection,.IEEE Trans.on Pattern Analysis and M achine Intelligence,8:6, 1986.[3] D. N. de Infra-Estrutura de Transportes. M anualde E s-tudosde Tráfego. Rio de Janeiro, 2006.[4] Intel. Open source computer vision li<strong>br</strong>ary.http://www.intel.com/research/mrl/research/opencv.[5] F. Keissarian. Computer vision - based imaging systemsfor road traffic monitoring. In A nnualResearch C onferenceatUAEUniversity, 2002.[6] I. M. Monteiro. Lógica fuzzy. Instituto de Informática,Universidade Federal do Rio Grande do Sul, <strong>2007</strong>.[7] S. Pinto and E. Preussler. Pavimentação rodoviária: conceitosfundamentais so<strong>br</strong>e pavimentos flexíveis. Rio deJaneiro, 2002.246

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!