Optical Flow for Cars Overtaking monitor 1 OPTICAL FLOW FOR CARS OVERTAKING MONITOR: THE REAR MIRROR BLIND SPOT PROBLEM

A fundamental goal of an overtaking monitor system is the segmentation of the overtaking vehicle. This application can be addressed through an optical flow driven scheme. We focus on the rear mirror visual field using a camera on the top of it. If we drive a car, the egomotion optical flow pattern is more or less unidirectional, i.e. all the static objects and landmarks move backwards. On the other hand, an overtaking car generates an optical flow pattern in the opposite direction, i.e. moving forward towards our car. This makes motion processing schemes specially appropriate for an overtaking monitor application. We have implemented a highly parallel bio-inspired optical flow algorithm and tested it with real overtaking sequences in different weather conditions. We have developed a postprocessing optical flow step that allows us to estimate the car position. We have tested it using a bank of overtaking car sequences. The overtaking vehicle position can be used to send useful alert signals to the driver.