In order to gain higher efficiency in obstacle avoidance task in autonomous vehicles from the aspect of processing cost and operating in real-time, it`s critical to find a region of interest (ROI) which obstacles are more possible to appear and degrade the obstacle`s search zone to it. In this paper we propose novel methods to find this ROI using computer vision technologies. The road scenes are acquired with a monocular camera. Current lane of autonomous vehicle is recognized by detection of lane markings. Adjacent lanes are also estimated based on some geometric calculations. A novel lane matching mechanism is suggested to validate detected lane markings. Finally a method for lane departure warning is proposed. The experimental results show that the proposed algorithms correctly find lanes region with high accuracy in real-time, are robust to noise and shadows, testing on Hemmat highway in Tehran and another dataset in the daytime.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.