The paper presents a part of an advanced system for railway and road traffic monitoring in metropolitan areas utilizing analyses of sound, vision and air pollution based on a ubiquitous computing approach. This system is a functional part of a more complex solution designed for environmental monitoring in cities. The system consists of many autonomous, universal measuring units and a multimedia server, which gathers, processes and presents data obtained from the distributed measuring units. The results are visualized on numerical maps. The paper contains a functional and a technical description of the monitoring system. It describes also the algorithm for moving vehicle detection in video sequences based on a pixel-level difference among the image frames and a continually updated background model utilizing mixtures of Gaussians. The experiments carried out involve the implemented algorithm to the detection of vehicles in the recorded video sequences. The results obtained are illustrated with some examples and discussed.