Computer Vision based autonomous robot navigation has caught the attention of many researchers because of its huge potential in various industrial and social applications. This paper focuses on various approaches used to detect road region which is fundamental requirement in applications like intelligent vehicles, lane detection and tracking and driver assistance systems. Broadly speaking techniques proposed for road detection are classified in three categories: activity driven, feature driven and model driven. This paper surveys development of last decade in the field of vision based road region detection. Two major components discussed in this paper are methods for structured and unstructured road detection. Keywords— Computer vision, autonomous navigation, structured roads, unstructured roads.
Download Full PDF Version (Non-Commercial Use)