主办:陕西省汽车工程学会
ISSN 1671-7988  CN 61-1394/TH
创刊:1976年

汽车实用技术 ›› 2022, Vol. 47 ›› Issue (13): 21-25.DOI: 10.16638/j.cnki.1671-7988.2022.013.006

• 智能网联汽车 • 上一篇    

复杂光照环境的车道线检测方法

张彬腾,张 宁,袁 敏   

  1. 广州城市理工学院 汽车与交通工程学院
  • 出版日期:2022-07-15 发布日期:2022-07-15
  • 通讯作者: 张彬腾
  • 作者简介:张彬腾(1998-),男,研究方向为复杂光照环境下的图像处理与道路检测,E-mail:564006205@qq.com。
  • 基金资助:
    国家级大学生创新创业项目(51JY200517)。

Lane Line Detection Method in Complex Illumination Environment

ZHANG Binteng, ZHANG Ning, YUAN Min   

  1. School of Automotive and Traffic Engineering, Guangzhou City Institute of Technology
  • Online:2022-07-15 Published:2022-07-15
  • Contact: ZHANG Binteng

摘要: 为了解决高速公路和城市道路上复杂天气环境下车道线检测的问题,提出一种复杂光 照环境的车道线检测方法。该方法通过自动调节复杂光照环境下图像的对比度特征,从而更 好地提取道路上的车道线,实现黑暗及雾天环境下的车道线检测。通过实验验证,该方法在 黑暗环境下车道线检测识别率为 96.7%,雾天环境下的车道线检测识别率为 95.2%,具有较好 的算法鲁棒性、准确性和实时性。

关键词: 复杂光照环境;对比度特征;车道线检测

Abstract: In order to solve the problem of lane line detection in complex weather environment on expressway and urban roads, a lane line detection method in complex lighting environment is proposed. This method can better extract the lane line on the road by automatically adjusting the contrast feature of the image in complex lighting environment, and realize the lane line detection in dark and fog environment. Experiments show that the lane line detection and recognition rate of this method is 96.7% in dark environment and 95.2% in fog environment. It has good robustness, accuracy and real-time performance.

Key words: Complex lighting environment; Contrast characteristics; Lane line detection