Lane detection algorithm based on top-view image using random sample consensus algorithm and curve road model

Recently, Lane Detection technology has been used for passenger safety systems such as the Lane Departure Warning System and Lane Keeping assist system to the most of the recently launched vehicles. There are many researches for lane detection algorithm but approaches of the previous studies such as template matching method, probabilistic method, color model method, etc. have limitations that are high sensitivity to noise similar to lane shape and non-uniform illumination. In this paper, we proposed lane detection algorithm based on generated Top-View image through Inverse Perspective Mapping using Random Sample Consensus algorithm. Moreover, the detected lane is extended to the bottom of the Region of Interest by applying the Curve road model. The proposed algorithm has been tested in various environment conditions. Experimental results show that the proposed algorithm can detect both straight and curve lane and can process about 25 frames per second.