Advances in automated pedestrian counting

Recent developments in automated building management and road traffic monitoring systems have highlighted a need for accurate real time data relating to occupancy and flow of pedestrians in relatively unconstrained physical areas. The ESPRIT PEDMON (PEDestrian MONitoring) project, which was funded through the Central European Commission, focused on the development of technologies to enable the automated counting of pedestrians in unconstrained areas and making that information available to users on a real time basis. The principal threads of the project involved research and development of the following: a suitable human computer interface, a simple low cost communications network, non-video sensing nodes and video based sensing nodes. This report describes the video image processing work, hardware and network development performed by the authors in fulfilment of the project. In the first section aspects of the image processing algorithms are discussed. In the second section the hardware and network developed for the intelligent sensors are detailed. In the third section the system performance is discussed based on real data taken from the installation at St Lazare Railway Station in Paris.