ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Multi-lane Detection Robust to Complex Illumination Variations and Noise Sources

Srivastava, S and Maiti, R (2019) Multi-lane Detection Robust to Complex Illumination Variations and Noise Sources. In: 2019 IEEE International Conference on Electrical, Control and Instrumentation Engineering, ICECIE 2019, 25 November 2019, Malaysia.

[img] PDF
int_con_ele_con_ins_eng_nov_2019.pdf - Published Version
Restricted to Registered users only

Download (1MB) | Request a copy
Official URL: https://dx.doi.org/10.1109/ICECIE47765.2019.897479...

Abstract

Lane detection is the active research area in the field of Advanced driver assistance system (ADAS) and autonomous vehicle. However, it is not easy to detect lane accurately when road lanes are affected by occlusion, weather conditions and illumination variations. In this paper, we propose a multi-lane detection robust to complex illumination variations and noise sources. The first step of this method is to generate adaptive Region of Interest (ROI) by using the combination of Hough lines and vanishing point. To improve the accuracy vanishing point detection, noise removal method is employed. Second, mask is generated to extract the road boundary region, which will be used as a search space for the detection of candidate lane mark points present on the road. Outlying points in mask generation process is eliminated using Density-Based Spatial Clustering of Applications with Noise (DBSCAN). Finally, lane is detected by fitting second order polynomial curve on extracted lane mark points by using Random Sample Consensus (RANSAC). Experiment results on the dataset shows that proposed method efficiently extracts lane features and detects the multi-lane in real world driving scenarios under various complex illumination variations and noise sources. © 2019 IEEE.

Item Type: Conference Paper
Publication: 2019 IEEE International Conference on Electrical, Control and Instrumentation Engineering, ICECIE 2019 - Proceedings
Publisher: Institute of Electrical and Electronics Engineers Inc.
Additional Information: cited By 0; Conference of 2019 IEEE International Conference on Electrical, Control and Instrumentation Engineering, ICECIE 2019 ; Conference Date: 25 November 2019; Conference Code:157264
Keywords: Automobile drivers; Curve fitting; Image segmentation; Polynomial approximation; Roads and streets, Adaptive region of interest; Density-based spatial clustering of applications with noise; Hough lines; Illumination variation; Random sample consensus; Second-order polynomial; Vanishing point; Vanishing point detection, Advanced driver assistance systems
Department/Centre: Division of Mechanical Sciences > Centre for Product Design & Manufacturing
Date Deposited: 13 Aug 2020 10:30
Last Modified: 13 Aug 2020 10:30
URI: http://eprints.iisc.ac.in/id/eprint/64664

Actions (login required)

View Item View Item