Illumination robust road detection using geometric information

Abstract

Road detection is an important task in intelligent transportation system (ITS). Over the past few decades, several vision-based approaches for road detection have been proposed and most of them are based on color information. However, color information may result in false road detection under variation of illumination conditions. To deal with illumination problems, we propose an illumination invariant road detection method using geometric information. By incorporating geometric information with a color-based road probability map, the proposed method robustly detect road regions on real scene containing variation of illumination such as shadow and mixed artificial lights. Experimental results show that the proposed method outperforms the conventional methods.