Obstacle detection using bi-spectrum CCD camera and image processing

Describes a method to identify cars from the road image obtained by the CCD camera. This method is based on the bi-spectrum combination between the visible and the near infrared ranges. The results allow the system to combine road marking identification for lane localization with obstacle detection. A detection algorithm may be implemented on a real-time machine to react on braking and accelerating pedals. This work contributes to the obstacle detection in the Eureka project PROMETHEUS.<
>