Automatic Passenger Counting in the High Occupancy Vehicle Lane

Project summary:

We propose to carry out a feasibility study regarding the automatic counting of car passengers in the High Occupancy Vehicle Lane (HOVL) of freeways. The study will be carried out in cooperation with Honeywell. We plan to investigate the combined use (fusion) of infrared and visible spectrum sensors. In the particular application, infrared systems can overcome problems caused by the fact that the interior of a vehicle is nearly always much darker than the exterior. In addition, infrared systems give distinct signatures for dummies and humans. Visible spectrum cameras are also useful because they can enhance vehicle detection and registration. Appropriate fusion of the infrared and visible range signals is expected to combine their relative merits and produce a superior result. The clarity of the fused signal will determine the computer vision methodology for the passenger-counting task. The merits of our approach will be validated during a lab demonstration at the end of the project period.