Traffic Lane Line Classification System by Real-time Image Processing

Huang Chingting, Hu Zhuqi, Shigeyuki Tateno

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The traffic safety has been a major concern in recent years. One of the effective approaches to prevent the traffic accident is to develop advanced driver assistance systems which can alarm driver in dangerous situation. In fact, changing lane or overtaking another vehicle is one of the most dangerous driving behaviors. Therefore, it is important for drivers to recognize current lane line types to take proper actions. However, classification systems proposed so far can only distinguish up to five types of lane lines, such as dashed and solid. Hence, the existing road classification systems are not suitable if there are more types of lane lines on the road. In this paper, an improved method is proposed to classify more lane line types by real-time image processing. In order to increase the detection accuracy of lane line types, the image stitching method is applied to reduce the misjudgment caused by blocked lane lines. A set of features about pixel distribution is utilized in the classifier to distinguish more than five lane line types. Furthermore, the results of experiments which are carried out in real road driving show high accuracy of the proposed classification method under the various situations.

Original languageEnglish
Title of host publication2018 International Automatic Control Conference, CACS 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538662786
DOIs
Publication statusPublished - 2019 Jan 9
Event2018 International Automatic Control Conference, CACS 2018 - Taoyuan, Taiwan, Province of China
Duration: 2018 Nov 42018 Nov 7

Publication series

Name2018 International Automatic Control Conference, CACS 2018

Conference

Conference2018 International Automatic Control Conference, CACS 2018
CountryTaiwan, Province of China
CityTaoyuan
Period18/11/418/11/7

ASJC Scopus subject areas

  • Control and Optimization
  • Modelling and Simulation

Fingerprint Dive into the research topics of 'Traffic Lane Line Classification System by Real-time Image Processing'. Together they form a unique fingerprint.

  • Cite this

    Chingting, H., Zhuqi, H., & Tateno, S. (2019). Traffic Lane Line Classification System by Real-time Image Processing. In 2018 International Automatic Control Conference, CACS 2018 [8606775] (2018 International Automatic Control Conference, CACS 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CACS.2018.8606775