In recent years, Lane Departure Warning System for security has attracted great attention of driving assistance in order to enhance safety and security usage on road driving. This paper presents the implementation of a real-time monocular vision based lane departure warning system. In order to make it useful for real situation, our proposed system is suggested to be real time. Otsu's threshold method and Hough Transform are the main algorithms used to deal with the lighting problem, extract the lane boundaries and monitor the vehicle's position with respect to the lane and provides warning if the vehicle is about to leave the lane. Vanishing point detection plays an essential role in lane detection to extract the Region of Interest and reduce computational complexity. Our proposed algorithm works accurately in various lighting conditions.