Lane detection algorithm based on density clustering and RANSAC
Authors:
- 1 June 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 919-924
Abstract
In order to improve the accuracy, real-time and robustness of video lane detection, in this paper, we propose a fast lane detection algorithm based on DBSCAN (Density Based Spatial Clustering of Applications with Noise) clustering algorithm and improved RANSAC (random sample consensus). Firstly, the image is preprocessed by graying and binarizing in the inverse perspective mapping image. Secondly, eliminating noise points and smoothing lane boundaries by using morphological erosion and opening operation in the binary image, then traversing the binary image and extracting the feature points. Then, the feature points are clustered by DBSCAN clustering algorithm, which are then divided into clusters. Finally, the points in every class was fitted a straight line or curve line by using improved RANSAC algorithm based on parabola. The experimental results show that the improved RANSAC algorithm accelerates the speed of lane detection. The proposed algorithm has good real-time, robustness and accuracy.Keywords
This publication has 5 references indexed in Scilit:
- Lane Detection Method Based on Improved RANSAC Algorithm2015
- Sharp curve lane boundaries projective model and detection2012
- Lane Detection Algorithm at Night Based-on Distribution Feature of Boundary Dots for Vehicle Active SafetyInformation Technology Journal, 2012
- An extended hyperbola model for road tracking for video-based personal navigationKnowledge-Based Systems, 2008
- Fast lane detection with Randomized Hough Transform2008