A COMPUTER VISION BASED APPROACH FOR ACCURATE ROAD LANE DETECTION

Abstract

This project focuses on detecting road lane lines using Python programming and the OpenCV library. It is mainly useful for driver assistance and basic self-driving systems. By analyzing the road images or videos taken from a front-facing camera, the system can identify lane markings and draw lines over them. This helps the vehicle stay within its lane and improves road safety for both drivers and passengers.We start by converting the road image into a simpler format using grayscale and removing noise using filters like Gaussian blur. Then, we apply edge detection to highlight the lane boundaries. A region of interest is selected to focus only on the road part, and then we use the Hough Line Transform to detect the actual lane lines. The detected lines are then drawn clearly on the image, showing the path of the lane. This method is low-cost and works well in good lighting and road conditions. It can be improved further by adding curve detection and handling shadows or faded lane marks. This project shows how image processing in Python with OpenCV can be used to solve real-world traffic problems in a simple and effective way.