Title: Lane detection method based on improved Hough transform

Authors: Yimin Yang

Addresses: The Higher Educational Key Laboratory for Flexible Manufacturing Equipment Integration of Fujian Province, Xiamen Institute of Technology, Xiamen, Fujian, China

Abstract: In intelligent driving, how to keep the vehicle on the road safely and accurately without deviating from the road, is an important topic. In practice, machine vision is commonly used to effectively detect lane lines, so as to alarm vehicles that deviate from lane lines. In this paper, the detection of lane lines includes image pre-processing to obtain areas of interest, histogram enhancement for low-contrast images, median filtering to remove image noise while preserving details, and Otsu threshold segmentation method to separate targets in images. After image pre-processing, the Laplacian of Gaussian operator is selected for edge detection by comparing and analysing several operators. Finally, the improved Hough transform is used to realise the lane detection within the limited parameters, reducing the computation and saving the running time. Experimental results show that the proposed algorithm can effectively detect lane lines in normal weather or under low contrast.

Keywords: lane detect; image enhancement; edge detection; Hough transform.

DOI: 10.1504/IJSPM.2023.139791

International Journal of Simulation and Process Modelling, 2023 Vol.21 No.1, pp.14 - 21

Received: 05 Sep 2023
Accepted: 07 Dec 2023
Published online: 05 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article