Road Boundary Detection for Straight Lane Lines Using Automatic Inverse Perspective Mapping

This paper addresses the difficulty of detecting and/or verifying road boundary markings from road images. The 2-D road images usually contain the perspective effect/distortion because of the acquisition process resulting in an inaccuracy of the lane detection. We thus propose in this paper an approach to automatically select four points for performing the Inverse Perspective Mapping (IPM) in order to diminish such effect/distortion. Our proposed method is efficient and very practical to implement with the existing lane detection method, as shown in the experimental results.