Multiple Lane Detection Algorithm Based on Novel Dense Vanishing Point Estimation

Abstract

The detection of multiple curved lane markings is still a challenge for advanced driver assistance systems today, due to interference such as road markings and shadows cast by roadside structures and vehicles. The vanishing point $\mathbf{V}_{p}$ contains the global information of the road image. Hence, $\mathbf{V}_{p}$ -based lane detection algorithms are quite insensitive to interference. When curved lanes are assumed, $\mathbf{V}_{p}$ shifts with respect to the rows of the image. In this paper, a $\mathbf{V}_{p}$ for each individual row of the image is estimated by first extracting a $\mathbf{V}_{py}$ (vertical position of the $\mathbf{V}_{p}$ ) for each individual row of the image from the v-disparity. Then, based on the estimated $\mathbf{V}_{py}$ 's, a 2-D $\mathbf{V}_{px}$ (horizontal position of the $\mathbf{V}_{p}$ ) accumulator is efficiently formed. Thus, by globally optimizing this 2-D $\mathbf{V}_{px}$ accumulator, globally optimum $\mathbf{V}_{p}$ s for the road image are extracted. Then, estimated $\mathbf{V}_{p}$ s are utilized for multiple curved lane marking detection on nonflat road surfaces. The resultant system achieves a detection rate of 99% in 1862 frames of six stereo vision test sequences.