Road Boundary Detection Based on the Dynamic Programming and the Randomized Hough Transform

A new vision-based road boundary detection algorithm combining dynamic programming and Hough transform is proposed. The dynamic programming (DP) is known to be a powerful algorithm for optimal path finding on a cost field. Utilizing the intensities of edge image as costs of the DP, most likely road boundaries are obtained in this paper. The DP is further utilized as a spatial filter which removes unrelated edge candidates as road boundaries. Randomized Hough transform is applied on the filtered edge image to detect practically useful road boundaries with straight line segments. Experimental results for road images acquired under situations with shadow, unmarked road painting and illumination variations in rural area are presented.