In public transportation there's more and more need to analyse passenger flow in order to enhance communication on each line - setting out more vehicles in rush hours, or reducing their number during lower congestion period.

Typically in public transportation passengers are being counted with a help of IR detectors. They're very expensive, hence it is always a tough decision whether to install such system, and if so, only a small percentage of vehicles end up with it.

The alternative for IR based motion detectors is image analysis software combined with a DVR and small cameras mounted above each entrance.

After the vehicle comes back at the end of the day to a depot, all video data is transfered to a local server, and appropriately configured software performs analysis over recorded video files, tracking passenger movement and counts both entrances and exits through each door. Time stamps on video frames, combined with desired line timetable makes it possible to create a graph with total number of passengers on Y- and consecutive stops on X-axis.

As the system uses a DVR for data acquisition, it is possible to combine passenger counting with video surveillance using same DVR with more input channels, hence the total cost of the system is just a fraction of the IR based ones price.