An Improved Lane Detection and Tracking Method for Lane Departure Warning Systems

An Improved Lane Detection and Tracking Method for Lane Departure Warning Systems

Mohamed Hammami (MIRACL, Sfax University, Sfax, Tunisia), Nadra Ben Romdhane (MIRACL, Sfax University, Sfax, Tunisia) and Hanene Ben-Abdallah (MIRACL, Sfax University, Sfax, Tunisia)
Copyright: © 2013 |Pages: 15
DOI: 10.4018/ijcvip.2013070101
OnDemand PDF Download:
No Current Special Offers


Lane detection and tracking are very crucial treatments in lane departure warning systems as they help the vehicle-mounted system to keep its lane. In this context, the authors’ work aims to develop vision-based lane detection and tracking method to detect and track lane limits in highways and main roads. The authors’ contribution focuses on the detection step. By exploiting the fact that, in an image, the road can be formed by linear and curvilinear portions, the authors propose two types of appropriate treatments to detect the lane limits. The authors’ method offers high precision rates independently of the painted lane marking’s characteristics, of the time of acquisition and in different weather conditions. Besides the challenges it overcomes, the authors’ method has the advantage of operating with a timing complexity that is reasonable for real-time applications. As shown experimentally, compared to three leading methods from the literature, the authors’ method has a higher efficiency.
Article Preview

The aim of lane detection and tracking based on image processing is, first, to locate in a sequence frame the lane limits of the road in which the vehicle is engaged and, then, to track these limits in the remaining frames.

The acquired images are, first, pre-processed in order to reduce the noise always caused by the sensors (Min, Liu, & Xu, 2006; Wennan, Qiang, & Hong, 2006; Ming-Dar, et al., 2008), to reduce the execution time based on the processing of a region of interest (Min, Liu, & Xu, 2006; Lim, Seng, Ngo, & Ang, 2009), or also to reduce the perspective effect of the lane by applying the inverse perspective mapping (Juan, Hilario, de la Escalera, & Armingol, 2005; Zu, 2006). The second step extracts the approximate pixels of LM either by detecting their edges (Min, Liu, & Xu, 2006; Ming-Dar, et al., 2008; Chih-Hsein & Chen, 2006; Haiping, Ko, Shil, Kim, & Kim, 2007), or by detecting their regions throw image segmentation (Lim, Seng, Ngo, & Ang, 2009; Serge, Michel, & Xuan, 1989; Chao, Mei, & D, 2010). The third step, the lane detection step, determines effective limits on an acquired image, based on the approximate pixels retained in the second step. Various lane detection methods are proposed in the literature and can be grouped in two categories: model-based approach (Min, Liu, & Xu, 2006; Wennan, Qiang, & Hong, 2006; Ming-Dar, et al., 2008; Romuald, Roland, & Frédéric, 2000) and feature-based approach (Ming-Dar, et al., 2008; Zu, 2006; Craig & Zou, 2007; Jan Pablo & Ozguner, 2000) (see Table 1). Finally, the last step, the lane tracking step allows the continuous detection within all the video sequence by updating, in every frame, the last detected limits. This step is required to minimize the noises and the execution time. Different methods were proposed to carry out this tracking step, they can be broadly classified in two approaches: deterministic-based approach (Liu, Dai, Song, He, & Zhang, 2011; Min, Liu, & Hu, 2006; Yue, Teoha, & Shenb, 2004; Muhammad, Arshad, Irfan, & Yahya, 2007) and stochastic-based approach (Ming-Dar, et al., 2008; M, A, & N, 2007) (see Table 2).

Complete Article List

Search this Journal:
Volume 12: 4 Issues (2022): 1 Released, 3 Forthcoming
Volume 11: 4 Issues (2021)
Volume 10: 4 Issues (2020)
Volume 9: 4 Issues (2019)
Volume 8: 4 Issues (2018)
Volume 7: 4 Issues (2017)
Volume 6: 2 Issues (2016)
Volume 5: 2 Issues (2015)
Volume 4: 2 Issues (2014)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing