Lane Departure Warning through Message and Alarm & Vehicle Position Tracking using Video Camera based Driver Assist System

( Vol-2,Issue-7,July 2016 )

Author(s): Tangina Sultana, Md. Delowar Hossain, Asfia Nuzat, Nadira Islam Ruku

Total View : 1136
Downloads : 164
Page No: 1054-1058
DOI:

Keywords:

Geometric Translation, Haugh Transform, Kalman Filter, Lane detparture, Tracking

Abstract:

In modern competitive era, people are so much desperate to reach their goal in a short time. They are not even concerned much about their lives. When these people are driving on road, they have more attention about the time to reach the destination. As a result, they are risking their and others lives on road due to driver’s inattention or incompetence or drowsiness. If driver have an assist system in their vehicle which alert him when he drives car out of lane, then he can save himself and others from accident. This paper proposes a video based driver assist system which alerts the driver with audio alarm and visual message about lane departure as well as can track the specific vehicle using license plate extraction. The proposed system uses camera sensor to get the real time input data of road environment which is then passes through the 2D FIR filter and thresholding process. After that Hough transform and Hough line is used to detect lane marker and line. Kalman filter is employed for lane tracking. Furthermore, the system searches a specific region to detect vehicle based on license plate extraction. Geometrical translator is used to reduce the unwanted motion of the video. The algorithm of our proposed system is implemented by using MATLAB. The driver assist system which is proposed in this paper is very efficient, low cost and robust against noise. Through this system, the accuracy of lane detection is almost 96% in different environment.

Cite This Article:

Share: