Curve lane detection based on the binary particle swarm optimization

Abstract

The method of B-spline curve fitting linebased on the binary particle swarm optimization is presented in this paper. First, according to the characteristics of the vertical and transverse width of the line must be straight, to extract the lane line feature points, and sorting out the feature points. Then, we select the cubic B-spline curve to fit the curve lane, first the discrete bunary particle swarm algorithm to optimize the number n of the control points, then by the least square method to calculate the control points of B-spline curve, according to the B-spline crrve fitting out the corresponding curve line. In order todetect corners recognition performance in a variety of road conditions has carried on the experimental study and the results show that the method has great adaptability.