Advanced lane detection technique for structural highway based on computer vision algorithm

Lane departure detection plays a vital role in the Advanced Driver assistive systems and it improves the vehicle's active safe driving. A wholesome lane detection method which is based on computer vision techniques, is introduced. The lane boundaries and its radius of curvatures and lane direction were detected from a stream of videos. This video footage was recorded from a camera mounted on the top of a vehicle. We have corrected the camera distortion in the input frame. HLS thresholding and sobel thresholding techniques are applied to the undistorted image for getting focus on the lane lines in the binary image. Then the resulted frame is warped to the bird's eye by applying the perspective transform. The respective lane line pixels are identified using sliding window search and then left and right lane lines are identified by fitting second-degree polynomials. The lane curvature and deviation from the lane centre are also computed after the identification of the lane. The identified lane boundaries are warped back onto the input image and the radius of lane curvature and vehicle position is calculated and displayed. Hence this technique is enforced using python programming language and for processing the images open CV is utilized. The obtained result illustrates how the proposed method accurately detects the lane line in different lightning conditions. (c) 2021 Elsevier Ltd. All rights reserved. Selection and peer-review under responsibility of the scientific committee of the International Conference on Advances in Materials Research-2019.