Multilane detection and tracking based on binocular vision stixel world estimation and IPM

Jing Li, Junzheng Wang, Guangtao Cui

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

Road detection is the basic component of many intelligent vehicle systems. In this paper, a robust multi-lane detection method based on binocular vision is proposed. First, a fast estimation technique of the Stixel World, which is the outside environment representation of stereo-vision, is developed. Taking advantage of the estimate for free space under the plane road hypothesis, the proposed method is robust against on-road obstacles when detecting lane markings. Then, the bird-view of the transitable area is obtained through Inverse Perspective Mapping (IPM) Transform; Steerable filter, parallel parabolas modal and RANdom SAmple Consensus (RANSAC) technique are introduced for multi-lane fitting. Experimental results in real urban driving situations indicate that our algorithm is robust under various conditions.

Original languageEnglish
Title of host publicationProceedings of the 38th Chinese Control Conference, CCC 2019
EditorsMinyue Fu, Jian Sun
PublisherIEEE Computer Society
Pages4602-4607
Number of pages6
ISBN (Electronic)9789881563972
DOIs
Publication statusPublished - Jul 2019
Event38th Chinese Control Conference, CCC 2019 - Guangzhou, China
Duration: 27 Jul 201930 Jul 2019

Publication series

NameChinese Control Conference, CCC
Volume2019-July
ISSN (Print)1934-1768
ISSN (Electronic)2161-2927

Conference

Conference38th Chinese Control Conference, CCC 2019
Country/TerritoryChina
CityGuangzhou
Period27/07/1930/07/19

Keywords

  • Binocular vision
  • Multilane detection and tracking
  • RANSAC
  • Stixel World

Fingerprint

Dive into the research topics of 'Multilane detection and tracking based on binocular vision stixel world estimation and IPM'. Together they form a unique fingerprint.

Cite this