ZTE Communications ›› 2020, Vol. 18 ›› Issue (4): 69-77.DOI: 10.12142/ZTECOM.202004010

• Research Paper • Previous Articles     Next Articles

Robust Lane Detection and Tracking Based on Machine Vision

FAN Guotian1(), LI Bo2, HAN Qin2, JIAO Rihua2, QU Gang2   

  1. 1.ZTE Corporation, Shenzhen 518057, China
    2.Xidian University, Xi’an 710071, China
  • Received:2019-03-19 Online:2020-12-25 Published:2021-01-13
  • About author:FAN Guotian (fan.guotian@zte.com.cn) received M.Sc. degree in Network Systems from University of Sunderland, UK, in 2008 and is currently working as a senior product manager in ZTE Corporation. His research interests include big data mining, digital image processing, wireless network planning and optimization, and GIS application.|LI Bo received the B.Eng. degree in information security from Xidian University, China in 2007, and is currently working in ZTE Corporation. His current research interests include digital image processing, information management system and big data analysis.|HAN Qin received the B.Eng. degree in telecommunication engineering from Xi’an University of Architecture and Technology, China in 2017, and is currently pursuing the M.Sc. degree in electronics and communication engineering, at Xidian University,China. Her current research interests include intelligent transportation systems, engineering implementation, and software development.|JIAO Rihua received the B.Eng. degree in telecommunication engineering from Xidian University, China in 2018, and is currently working in Baidu Online Network Technology Co., Ltd., China. His current research interests include digital image processing, distributed systems and search engine architecture.|QU Gang received the B.Eng. degree in automobile engineering from Lanzhou Jiaotong University, China in 2014, and is currently pursuing the M.Sc. degree in electronics and communication engineering at Xidian University, China. His current research interests include intelligent transportation systems, engineering implementation, and software development.

Abstract:

Lane detection based on machine vision, a key application in intelligent transportation, is generally characterized by gradient information of lane edge and plays an important role in advanced driver assistance systems (ADAS). However, gradient information varies with illumination changes. In the complex scenes of urban roads, highlight and shadow have effects on the detection, and non-lane objects also lead to false positives. In order to improve the accuracy of detection and meet the robustness requirement, this paper proposes a method of using top-hat transformation to enhance the contrast and filter out the interference of non-lane objects. And then the threshold segmentation algorithm based on local statistical information and Hough transform algorithm with polar angle and distance constraint are used for lane fitting. Finally, Kalman filter is used to correct lane lines which are wrong detected or missed. The experimental results show that computation times meet the real-time requirements, and the overall detection rate of the proposed method is 95.63%.

Key words: ADAS, Hough transform, Kalman filter, polar angle and distance, top-hat