Abstract:
Our dissertation describes in-depth the algorithm intended to identify lane lines on streets and
highways in different conditions. Lane detection enhances the protection of the independent
system to some extent. The autonomous or self-sufficient vehicle is an independent device that
senses conditions and determines accordingly without any human intervention. Our emphasis was
on the use of a vehicle prototype to recognize lanes that could be used later on a standard vehicle.
To capture real-time video, PiCamera is embedded with a RaspberryPi 3.0 Model B for processing
purposes. RaspberryPi along with battery, motors are mounted in a prototype constructed using
CNC machine with sheer perfection. The real-time video captured by Picamera is sufficient
enough to evaluate the performance of the algorithms used. The algorithms that have been used
are the concepts of OpenCV, Hough transformation, canny edge detection algorithm and
elementary algebra to compute and draw the lines. Python 2.7 was used to write the code for the
relevant algorithm. The accuracy level of the algorithm used is quite astonishing. Our research
works significantly in the field of autonomous vehicles. The comprehensive approach shown here
offers a fairly precise and efficient solution for tracking lines. This can make a big difference in
the driving experience in every way possible if used properly.
Description:
Supervised by
Dr. Golam Sarowar
Associate Professor,
Department of Electrical and Electronic Engineering,
Islamic University of Technology (IUT),
Boardbazar, Gazipur-1704.
