Vision-based blind spot monitoring

Abstract

These sections introduce a vision-based system designed for monitoring the area that a driver cannot see from exterior mirrors, usually referred to as blind spot. This is a challenging task that requires to discriminate from vehicles and background when both are not static and also to cope with usual automotive problems like camera vibrations and oscillations. The development of ADAS has involved an improvement of the safety on the road, ensuring vehicle reliability and supporting driver for accidents preventing, as seen in the previous chapters. Particularly, some studies have been focused on the assistance during lane change, monitoring the area covered by the blind spot, which drivers are unable to see in the exterior mirrors (>Fig. 44.1). If the system detects an overtaking vehicle in the dangerous zone, visual and acoustic signals warn the driver about the risk of collision. Radar technology and laser scanner are sometimes employed for the lane changing assistance, although they do not guarantee always a complete representation of the scene: When the vehicle is pitching down or up, the laser beam hits the ground or points to the sky, making data acquisition worthless. Moreover, in complex scenario, like urban road traffic, the radar has limited functionalities because its reliability depends on the radar cross section of the object to identify. For this reason, vision data fusion is necessary to provide high-resolution performance. A common choice for the blind spot monitoring is the using of a camera mounted on the wing mirror (>Fig. 44.2) in order to extend the driver's view.