A Robust Lane Detection and Departure Warning System

TL;DR

This paper presents a robust lane detection and departure warning system using a single camera, employing advanced image processing and modeling techniques to operate reliably under challenging lighting and shadow conditions.

Contribution

The work introduces a novel combination of illuminant invariant techniques, steerable filters, and RANSAC for robust lane detection from a single camera in complex environments.

Findings

Effective lane detection under shadow and lighting variations

Reliable lane boundary detection in Indian road conditions

Accurate lane departure warning using optical flow

Abstract

In this work, we have developed a robust lane detection and departure warning technique. Our system is based on single camera sensor. For lane detection a modified Inverse Perspective Mapping using only a few extrinsic camera parameters and illuminant Invariant techniques is used. Lane markings are represented using a combination of 2nd and 4th order steerable filters, robust to shadowing. Effect of shadowing and extra sun light are removed using Lab color space, and illuminant invariant representation. Lanes are assumed to be cubic curves and fitted using robust RANSAC. This method can reliably detect lanes of the road and its boundary. This method has been experimented in Indian road conditions under different challenging situations and the result obtained were very good. For lane departure angle an optical flow based method were used.