Lane Mark Identification Algorithm Based on Fast Line Detection
Han Sun · Computer Technology and Development · 2009
In order to get an ideal lane mark's edge in a road image paper,Canny operator is used to acquire an edge image.Then,line detection algorithm based on chain code is used to wipe off the line segment whose camber is higher.Meanwhile,length discrimination and orientation discrimination are used to get credible lane mark's edge line candidates.In further processing,linear connection algorithm based on minimizing connective potential energy is proposed in this paper to get more reliable lane mark's edge line candidates.Finally,we make use of POV(Point Orientation Vote) and barycenter to determine where the road boundary is.The result shows that the method proposed in this article is valid.