{"references": ["Sussman J.: Introduction to transportation systems - Artech House,\n2000", "Faro A., Giordano D., Spampinato C.: Soft-computing agents processing\nweb-cam images to optimize metropolitan traffic systems - Int. Conf. on\nComputer Vision and Graphics (ICCVG04), Warsaw, 2004", "Davis L. et alii: W4 real time surveillance of people and their activities -\nIEEE Trans on Pattern analysis and machine intelligence, v.22, N.8,\n2000", "Fisher R. et alii, Connected Component labeling - http//homepages.inf.\ned.ac.uk/ rbf/HIPR2/label.htm#1, 2003", "Byung-Doo Lee et alii: The application of the fuzzy reasoning to the\nopening games of 19x19Go - CITR-TR-15 2003-\nwww.citr.auckland.ac.nz", "ST Microelectronics: The fuzzy processor ST52F13 - Catania 2002"]}

Traffic management in an urban area is highly facilitated by the knowledge of the traffic conditions in every street or highway involved in the vehicular mobility system. Aim of the paper is to propose a neuro-fuzzy approach able to compute the main parameters of a traffic system, i.e., car density, velocity and flow, by using the images collected by the web-cams located at the crossroads of the traffic network. The performances of this approach encourage its application when the traffic system is far from the saturation. A fuzzy model is also outlined to evaluate when it is suitable to use more accurate, even if more time consuming, algorithms for measuring traffic conditions near to saturation.