Visual lane analysis and higher-order tasks: a concise review

Abstract

Lane detection, lane tracking, or lane departure warning have been the earliest components of vision-based driver assistance systems. At first (in the 1990s), they have been designed and implemented for situations defined by good viewing conditions and clear lane markings on highways. Since then, accuracy for particular situations (also for challenging conditions), robustness for a wide range of scenarios, time efficiency, and integration into higher-order tasks define visual lane detection and tracking as a continuing research subject. The paper reviews past and current work in computer vision that aims at real-time lane or road understanding under a comprehensive analysis perspective, for moving on to higher-order tasks combined with various lane analysis components, and introduces related work along four independent axes as shown in Fig. 2. This concise review provides not only summarizing definitions and statements for understanding key ideas in related work, it also presents selected details of potentially applicable methods, and shows applications for illustrating progress. This review helps to plan future research which can benefit from given progress in visual lane analysis. It supports the understanding of newly emerging subjects which combine lane analysis with more complex road or traffic understanding issues. The review should help readers in selecting suitable methods for their own targeted scenario.