GOLD - A PARALLEL REAL-TIME STEREO VISION SYSTEM FOR GENERIC OBSTACLEAND LANE DETECTION

This paper describes the Generic Obstacle and Lane Detection system (G OLD), a stereo vision-based hardware and software architecture to be u sed on moving vehicles to increment road safety, Based on a full-custo m massively parallel hardware, it allows to detect both generic obstac les (without constraints on symmetry or shape) and the lane position i n a structured environment (with painted lane markings) at a rate of 1 0 Hz, Thanks to a geometrical transform supported by a specific hardwa re module, the perspective effect is removed from both left and right stereo images; the left is used to detect lane markings with a series of morphological filters, while both remapped stereo images are used f or the detection of free-space in front of the vehicle. The output of the processing is displayed on both an onboard monitor and a control-p anel to give visual feedbacks to the driver, The system was tested on the mobile laboratory (MOB-LAB) experimental land vehicle, which was d riven for more than 3000 km along extra-urban roads and freeways at sp eeds up to 80 km/h, and demonstrated its robustness with respect to sh adows and changing illumination renditions, different road textures, a nd vehicle movement.