Distance estimation and vehicle position detection based on monocular camera

Abstract

This paper describes systematically two methods used in intelligent transportation systems: Distance Estimation using an onboard camera and car position detection. Distance estimation is a method for detecting distance for the preceding vehicles based on monocular camera. Vehicle position detection is a method of specifying the vehicle position relative to the road that can serve as Lane Departure Warning system. These two approaches have been discussed and implemented in this article. For lane detection and tracking, Hough Transform and Kalman filter were adopted. A brief introduction about both lane detection system and object detection is given. Finally, both approaches have been evaluated on a large dataset of videos.