An Urban Lane Detection Method Based on Inverse Perspective Mapping

Abstract

Lane markings can provide drivers significant warning instruction about current/ approaching road condition, in that case precise lane marking detection results are important assistance for safety issue in the driving assistance system (DAS). In this paper, we mainly focus on urban lane marking detection. We used sampling peaks from IPM image for lane markings' extraction and description. The proposed method is applied to various video images from black box, and is verified to be robust.

Road Detection through Supervised Classification

“…Detection and tracking of lane marking is essential for driving safety and intelligent vehicle [5], [6]. Offline road understanding and Lane detection algorithms are generally composed of multiple modules: image preprocessing, feature extraction and model-fitting [7], [8], [9], [10]. Low-level feature extraction (feature level processing) in every single frame is usually not practical in real-time scenarios due to complexity issues.…”

“…Detection and tracking of lane marking is essential for driving safety and intelligent vehicle [5], [6]. Offline road understanding and Lane detection algorithms are generally composed of multiple modules: image preprocessing, feature extraction and model-fitting [7], [8], [9], [10].…”