[1]陈 诚,张新闻,李 强.基于CPAM-UFLD的车道线检测方法的研究[J].浙江科技大学学报,2024,(06):515-528.[doi:10.3969/j.issn.2097-5236.2024.06.006 ]
 CHEN Cheng,ZHANG Xinwen,LI Qiang.Research on lane line detection method based on CPAM-UFLD[J].,2024,(06):515-528.[doi:10.3969/j.issn.2097-5236.2024.06.006 ]
点击复制

基于CPAM-UFLD的车道线检测方法的研究(/HTML)
分享到:

《浙江科技大学学报》[ISSN:2097-5236/CN:33-1431/Z]

卷:
期数:
2024年06期
页码:
515-528
栏目:
出版日期:
2024-12-28

文章信息/Info

Title:
Research on lane line detection method based on CPAM-UFLD
文章编号:
2097-5236(2024)06-0515-14
作者:
陈 诚张新闻李 强
(浙江科技大学 机械与能源工程学院,杭州 310023)
Author(s):
CHEN Cheng ZHANG Xinwen LI Qiang
(School of Mechanical and Energy Engineering, Zhejiang University of Science and Technology, Hangzhou 310023, Zhejiang, China)
关键词:
车道线检测 损失函数 CPAM注意力机制 空洞金字塔池化
分类号:
TP391.41
DOI:
10.3969/j.issn.2097-5236.2024.06.006
文献标志码:
A
摘要:
【目的】为改进车道线检测算法的精度,优化自动驾驶系统中的车道保持,提出了一种新的车道线检测算法——基于空洞金字塔池化的通道和空间注意力机制的车道线检测算法(channel and position attention mechanism with atrous spatial pyramid pooling for ultra fast structure-aware deep lane detection, CPAM-UFLD)。【方法】首先,在超快速结构感知深度车道检测(ultra fast structure-aware deep lane detection,UFLD)算法的基础上融入了空洞金字塔池化模块(atrous spatial pyramid pooling,ASPP),以有效地捕捉车道线图像中不同尺度的特征; 使用了通道和空间注意力机制(channel and position attention mechanism,CPAM)以关注图像中的关键区域; 同时,为平衡类别权重和提高定位精度,使用了包括加权交叉熵损失函数等的四种损失函数; 其次,提出了一种亮度改善模块,该模块旨在提升输入图像的质量,从而增强车道线的识别度。【结果】本算法在TuSimple数据集上的检测精度由原来的95.86%提升至96.56%; 同时,在CULane数据集上,检测精度由原来的72.2%提升至73.7%。【结论】通过算法的改进,可以有效提高车道线检测的精度,这为智能网联汽车自动驾驶系统的环境感知提供了理论参考。

参考文献/References:

[1] ZHANG J, XU Y, NI B, et al. Geometric constrained joint lane segmentation and lane boundary detection[C]//Proceedings of the Computer Vision.Cham:Springer International Publishing, 2018.
[2] JAVEED M A, GHAFFAR M A, ASHRAF M A, et al. Lane line detection and object scene segmentation using Otsu thresholding and the fast hough transform for intelligent vehicles in complex road conditions[J]. Electronics, 2023, 12(5):1079.
[3] LEE S H, KWON H J, LEE S H. Enhancing lane-tracking performance in challenging driving environments through parameter optimization and a restriction system[J]. Applied Sciences-Basel, 2023, 13(16):9313.
[4] JO Y H, LEE D J. Real-time lane detection techniques using optimal estimator and deep learning-based lane segmentation for self-driving vehicles[J]. Journal of Institute of Control, Robotics and Systems, 2022, 28(4):353.
[5] 张勇,杜学峰,高越,等.基于传统图像处理算法的车道线检测[J]. 汽车实用技术,2022,47(2):20.
[6] 李志远,王光辉.基于传统边缘算子的车道线识别研究[J]. 现代电子技术,2024,47(7):61.
[7] HUR J, KANG S N, SEO S W. Multi-lane detection in urban driving environments using conditional random fields[C]//Proceedings of the 2013 IEEE Intelligent Vehicles Symposium(IV).Gold Coast:Institute of Electrical and Electronics Engineers,2013.
[8] ZHENG T, FANG H, ZHANG Y,et al.RESA:recurrent feature-shift aggregator for lane detection[EB/OL]. https://arxiv.org/abs/2008.13719.
[9] BORKAR A, HAYES M, SMITH M T. A novel lane detection system with efficient ground truth generation[J]. IEEE Transactions on Intelligent Transportation Systems, 2012, 13(1):365.
[10] LIN H Y, CHANG C K, TRAN V L. Lane detection networks based on deep neural networks and temporal information[J]. Alexandria Engineering Journal, 2024, 98:10.
[11] ZHANG J Q, DUAN H B, CHEN J L, et al. HoughLaneNet:lane detection with deep hough transform and dynamic convolution[J]. Computers & Graphics, 2023, 116:82.
[12] ZHENG T, FANG H, ZHANG Y, et al. RESA:recurrent feature-shift aggregator for lane detection[C]//Proceedings of the AAAI Conference on Artificial Intelligence. New York:Association for the Advancement of Artificial Intelligence(AAAI), 2020.
[13] PHILION J. FastDraw:addressing the long tail of lane detection by adapting a sequential prediction network[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Long Beach:Institute of Electrical and Electronics Engineers, 2019.
[14] LIU R, YUAN Z, LIU T, et al. End-to-end lane shape prediction with transformers[C]//Proceedings of the 2021 IEEE Winter Conference on Applications of Computer Vision(WACV).Piscataway:Institute of Electrical and Electronics Engineers, 2021.
[15] LIU X, DENG Z. Segmentation of drivable road using deep fully convolutional residual network with pyramid pooling[J]. Cognitive Computation, 2018, 10(2):272.
[16] 吕嘉璐,周力,巨永锋.基于YOLOP-L的多特征融合道路全景驾驶检测[J]. 计算机科学,2024,51(增刊1):45.
[17] FU J, LIU J, TIAN H, et al. Dual attention network for scene segmentation[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Long Beach:Institute of Electrical and Electronics Engineers, 2019.
[18] QIN Z, ZHANG P, LI X. Ultra fast deep lane detection with hybrid anchor driven ordinal classification[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024, 46(5):2555.
[19] CEHN L C, PANPANDREOU G, KOKKINOS I, et al. DeepLab:semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(4):834.
[20] DONG Y, SHEN X, JIANG Z, et al. Recognition of imbalanced underwater acoustic datasets with exponentially weighted cross-entropy loss[J]. Applied Acoustics, 2021, 174:107740.
[21] YANG Z M, WU Q L, ZHANG F, et al. A new semantic segmentation method for remote sensing images integrating coordinate attention and SPD-Conv[J]. Symmetry-Basel, 2023, 15(5):1037.
[22] 邓天民,王琳,杨其芝,等.基于改进SegNet算法的车道线检测方法[J]. 科学技术与工程,2020,20(36):14988.
[23] 郭心悦,韩星宇,习超,等.基于非对称卷积的多车道线检测方法[J]. 计算机工程与设计,2024,45(2):428.

备注/Memo

备注/Memo:
收稿日期:2024-07-06
基金项目:浙江省“尖兵”“领雁”研发攻关计划项目(2023C01254)
通信作者:李 强(1979— ),男,江苏省溧阳人,教授,博士,主要从事智能网联汽车自动驾驶技术研究。E-mail:liqiang@zust.edu.cn。
更新日期/Last Update: 2024-12-28