To achieve safe and effective management of transportation networks a knowledge of the level of pedestrian activity is essential. This is however lacking or rudimentary at most existing signalised junctions. This paper describes results from a research program, funded by the EPSRC, to develop algorithms for the automatic visual sensing of pedestrian demand at pedestrian crossings. The measurement of pedestrian demand is achieved using small low-cost image sensors together with the automatic computer analysis of the sensed images. Such a pedestrian monitoring system will also find applications in a variety of related transportation areas including volumetric estimation of demand at bus stops and in metro systems. This paper begins with an introduction to the problem of pedestrian detection at PUFFIN (pedestrian user friendly intelligent) pedestrian crossings. It then goes on to describe the development of a new algorithm for pedestrian sensing which incorporates a model of observed pedestrian characteristics to improve detection performance. Finally results from the analysis of video recordings from initial site trials covering a range of flow and environmental conditions are presented.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Detection of pedestrians at Puffin crossings using computer vision


    Contributors:

    Publication date :

    1996-01-01


    Size :

    6 pages



    Type of media :

    Article (Journal)


    Type of material :

    Print


    Language :

    English




    The puffin pedestrian crossing

    Williams, D. / Phull, S. | Tema Archive | 1995


    Pedestrians behaviour at traffic-actuated crossings

    Schirmacher, G. | Online Contents | 2002



    Pedestrians' and cyclists' attitudes to Toucan crossings

    Taylor, S. B. / Halliday, M. E. | TIBKAT | 1997


    Sight Distance for Disabled Pedestrians at Crossings

    Schoon, John G. | Transportation Research Record | 2013