深度学习的车道线检测方法

doi:10.16638/j.cnki.1671-7988.2023.012.010

汽车实用技术 ›› 2023, Vol. 48 ›› Issue (12): 48-52.DOI: 10.16638/j.cnki.1671-7988.2023.012.010

• 智能网联汽车 • 上一篇

深度学习的车道线检测方法

张林

西安汽车职业大学智能制造工程学院

出版日期:2023-06-30 发布日期:2023-06-30
通讯作者: 张林
作者简介:张林（1995－），女，硕士，助教，研究方向为汽车机械，E-mail:zhanglin4424@163.com。

Lane Line Detection Method Based on Deep Learning

ZHANG Lin

School of Intelligent Manufacturing Engineering, Xi'an Automotive Vocational University

Online:2023-06-30 Published:2023-06-30
Contact: ZHANG Lin

摘要/Abstract

摘要： 随着人工智能技术的不断发展，无人驾驶技术已经成为当下社会发展的热门，车道线检测是无人驾驶技术的关键一环，但传统的基于视觉的车道线检测方法处理时间较长、过程繁琐、需要人为干预。基于深度学习的车道线检测可大大减少此类问题。文章设计了一个完成双任务的 Enet 网络，分别解决目标区域分割问题和不同车道区分问题，以实现端到端的检测。与主流车道线检测网络模型 deeplab v3 和 YOLO v3 进行对比。实验验证表明，网络模型的最终训练准确率为 99.8%，相对于 deeplab v3 和 YOLO v3 网络分别提升 1.2%和 0.8%。

关键词: 车道线检测；深度学习；端到端；实例分割

Abstract: With the continuous development of artificial intelligence technology, driverless technology has become a hot topic in the current social development. Lane line detection is a key link of driverless technology, but the traditional vision-based lane line detection method takes a long time to process, and the process is tedious, requiring human intervention. The lane line detection based on deep learning can greatly reduce such problems. In this paper, an Enet network is designed to complete dual tasks, which solves the problem of target area segmentation and different lane differentiation respectively to achieve end-to-end detection.Compared with mainstream lane line detection network models deeplab v3 and YOLO v3, the experimental verification shows that the final training accuracy of the network model used in this experiment is 99.8%, which is 1.2% and 0.8% higher than deeplab v3 and YOLO v3, respectively.

Key words: Lane line detection; Deep learning; End-to-end; Instance segmentation

张林. 深度学习的车道线检测方法[J]. 汽车实用技术, 2023, 48(12): 48-52.

ZHANG Lin. Lane Line Detection Method Based on Deep Learning[J]. Automobile Applied Technology, 2023, 48(12): 48-52.

参考文献

[ 1 ] ALY M.Real Time Detection of Lane Markers in Urban Streets[J].IEEE,2014:4621152. [ 2 ] YANIV A,ANDRAS F,AMNON S S.Off-road Path Following Using Region Classification and Geometric Projection Constraints[J]. IEEE, 2006(1):689-696. [ 3 ] LOOKINGBILL A,ROGERS J,LIEB D,et al.Reverse Optical Flow for Self-supervised Adaptive Autonomous Robot Navigation[J].International Journal of Computer Vision,2007,74(3):287-302. [ 4 ] WANG C,HU Z,UCHIMURA K.A Novel Lane Detection Approach Fusion by Vehicle Localization[C]// Intelligent Control & Automation. Piscataway:IEEE, 2011:1218-1223. [ 5 ] 石林军,余粟.基于多约束条件下的霍夫变换车道线检测方法[J].计算机测量与控制,2018,26(9): 9-12,38. [ 6 ] LI J, XIANG J A, HAN G He. Lane Detection Based on Visual Attention[C]//In Image and Graphics (ICIG), 2011 Sixth International Conference on.Piscataway: IEEE,2011:570-575. [ 7 ] GUO K Y, LI N,ZHANG M.Lane Detection Based on the Random Sample Consensus[C]//In Information Technology,Computer Engineering and Management Sciences (ICM),2011 International Conference on. Piscataway: IEEE,2011(3):38-41. [ 8 ] KAZEMI F M,SAMADI S,POORREZA H R,et al. Vechicle Recognition Using Curvelet Transform and SVM[C] //Proc of the 4th International Conference on Information Technology.Piscataway:IEEE.2007:516- 521. [ 9 ] HUANG L,ZHANGY S.Identification of Traffic Signs Using Deep Convolutional Neural Network[J].Modern Electronic Technology,2015,25(1):101-106. [10] 段建民,李岳,庄博阳.基于改进 SIS算法和顺序 RANSAC 的车道线检测方法研究[J].计算机测量与控制, 2018,26(8):280-284,289. [11] NIU S F,LI L.Application of Neural Network and Bayesian Filter in the Prediction of Change of Channel [J].Science and Technology and Engineering,2016 (14):212-216.

深度学习的车道线检测方法

Lane Line Detection Method Based on Deep Learning

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics