People pay more and more attention on the active safety technologies in Modern traffic. This paper develops a driving assistant system with the lane departure warning and forward collision warning functions. The system uses data fused from millimeter-waves radar and machine vision to implement detecting the front lane and vehicles ahead, to meet the requirement of accuracy, real-time and robustness in actual road environment. After analyzing information of targets ahead using for target location information, through establishing a unified coordinate system model, ROI (Region of interest) is defined according to the different candidate characteristics. Therefore, the search speed is improved greatly and the interference is reduced significantly. Finally, experiments demonstrate the feasibility of this system.
展开▼