An Approach to Linear Lane Mark Identifying and Tracking using Zernike Moments

In order to identify navigation lane for an intelligent vehicle more rapidly and exactly, the lane mark edge lines are firstly detected based on Zernike moments, and then lane mark feature points are gained according to certain characters. Based on these points, lane mark line parameters can be acquired rapidly and exactly by using linear regression algorithm. Finally lane mark real-time tracking is realized by applying trapezoid AOI method. Experimental results show that the proposed method has better reliability, robustness and real time.