基于起始点引导多维特征细化的复杂场景车道线检测

朱松豪; 谭少涵

doi:10.12263/DZXB.20240816

您当前的位置：

首页 >

文章列表页 >

基于起始点引导多维特征细化的复杂场景车道线检测

更新时间：2025-03-06

- 基于起始点引导多维特征细化的复杂场景车道线检测
- Multi dimension feature refinement based on starting point guidance for lane line detection in complex scene.
- 电子学报 2025年页码：1-10
- 作者机构：
  
  南京邮电大学自动化学院、人工智能学院，江苏南京 210023
- 作者简介：
  
  [ "朱松豪男，1973年2月生，江苏如皋人.南京邮电大学自动化学院、人工智能学院副教授.主要研究方向为图像处理、人工智能. E-mail： zhush@njupt.edu.cn." ]
- 基金信息：
  
  国家自然科学基金(62001247)
- DOI：10.12263/DZXB.20240816
  中图分类号： TP391.41;
- 收稿：2024-09-03，
  
  修回：2024-11-28，
  
  网络出版：2025-03-06，
- 稿件说明：
移动端阅览
朱松豪, 谭少涵. 基于起始点引导多维特征细化的复杂场景车道线检测[J/OL]. 电子学报, 2025,1-10.

ZHU Song-hao, TAN Shao-han. Multi dimension feature refinement based on starting point guidance for lane line detection in complex scene.[J/OL]. ACTA ELECTRONICA SINICA, 2025, 1-10.
朱松豪, 谭少涵. 基于起始点引导多维特征细化的复杂场景车道线检测[J/OL]. 电子学报, 2025,1-10. DOI： 10.12263/DZXB.20240816.

ZHU Song-hao, TAN Shao-han. Multi dimension feature refinement based on starting point guidance for lane line detection in complex scene.[J/OL]. ACTA ELECTRONICA SINICA, 2025, 1-10. DOI： 10.12263/DZXB.20240816.

摘要

车道线检测作为智能驾驶系统的基石，在车道保持、自适应巡航等辅助驾驶过程中发挥着至关重要的作用.鉴于车道线检测在提升道路安全、促进智能驾驶及智慧交通发展中的关键作用，车道线检测技术的研究具有深远的学术价值与应用前景.然而，由于车道线的多样性、路况的复杂性及天气的多样性等因素，车道线检测面临诸多挑战.研究提出一种基于起始点引导的多维特征细化的复杂场景车道线检测方法，利用全局特征优化模块和车道线感知聚合模块对特征进行优化，捕捉足够的上下文信息，利用起始点坐标预测模块预测车道线的起始点坐标，生成高质量的锚点.为更好适应二维车道线检测，研究提出更为通用的惩罚车道线交并比损失函数评估车道线的预测结果.与目前车道线检测算法中精度最高的CLRNet-DLA34方法相比，所提方法在CULane数据集和CurveLanes数据集的

@50分别高出了0.62%和0.73%，达到81.39%和86.83%.实验结果表明，所提方法在复杂场景车道线检测任务中取得了良好的检测效果，且在现有方法中具有很强的竞争力.

Abstract

Lane detection

as the cornerstone of intelligent driving systems

plays a crucial role in assisting driving processes such as lane keeping and adaptive cruise control.Given the crucial role of lane detection in improving road safety

promoting intelligent driving and intelligent transportation development

the research on lane detection technology has profound academic value and application prospects.However

due to the diversity of lane categories

road conditions and weather environments

as well as the different aspect ratios of lane lines

lane detection algorithms face many challenges.This paper proposes a multi-dimensional feature refinement method for complex scene lane detection based on start point guidance.Firstly

a global feature optimization module is utilized to enhance the global feature representation capability and a lane line perception aggregation module is utilized to enhance the correlation of local features

which helps to improve the semantic understanding ability and prediction accuracy of the model.Secondly

a starting point coordinate prediction module is utilized to predict the starting point coordinates of lane line to generate flexible anchors under various complex scenarios.Finally

a more general penalty lane intersection to union ratio is selected as the loss function to represent lane lines with variable virtual widths

which helps to improve the detection accuracy of the model.Compared with the CLRNet-DLA34 method

which has the highest accuracy in current lane detection algorithms

the method proposed in this paper improves the detection accuracy in terms of F1@50 on the CULane and CurveLanes datasets by 0.62%and 0.73%respectively

reaching 81.39%and 86.83%The experimental results demonstrate that the proposed method achieves good detection performance in complex scene lane detection tasks and has strong competitiveness among existing methods.Experimental results demonstrate that the proposed method performs well in lane detection tasks in complex scenes and has strong competitiveness among existing methods.

关键词

Keywords

references

LIU G L , WÖRGÖTTER F , MARKELIĆ I . Combining statistical hough transform and particle filter for robust lane detection and tracking [C ] // 2010 IEEE Intelligent Vehicles Symposium . Piscataway : IEEE , 2010 : 993 - 997 .

ZHOU S Y , JIANG Y H , XI J Q , et al . A novel lane detection based on geometrical model and Gabor filter [C ] // 2010 IEEE Intelligent Vehicles Symposium . Piscataway : IEEE , 2010 : 59 - 64 .

徐频捷 , 陈逸杰 , 李之南 , 等 . 基于事件驱动的车道线识别算法研究 [J ] . 电子学报 , 2021 , 49 ( 7 ): 1379 - 1385 .

XU P J , CHEN Y J , LI Z N , et al . Research on event-driven lane recognition algorithms [J ] . Acta Electronica Sinica , 2021 , 49 ( 7 ): 1379 - 1385 . (in Chinese)

LIU L Z , CHEN X H , ZHU S Y , et al . CondLaneNet: A top-to-down lane detection framework based on conditional convolution [EB/OL ] .( 2021-05-11 )[ 2023-02-10 ] . https://arxiv.org/abs/2105.05003v3 https://arxiv.org/abs/2105.05003v3 .

TABELINI L , BERRIEL R , PAIXAO T M , et al . Keep your eyes on the lane: Real-time attention-guided lane detection [C ] // 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2021 : 294 - 302 .

ZHENG T , HUANG Y F , LIU Y , et al . CLRNet: Cross layer refinement network for lane detection [C ] // 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2022 : 888 - 897 .

DING X H , ZHANG X Y , HAN J G , et al . Scaling up your kernels to 31 × 31: Revisiting large kernel design in CNNs [C ] // 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2022 : 11953 - 11965 .

PAN X G , SHI J P , LUO P , et al . Spatial as deep: Spatial CNN for traffic scene understanding [J ] . Proceedings of the AAAI Conference on Artificial Intelligence , 2018 , 32 ( 1 ): 7276 - 7283 .

ZHENG T , FANG H , ZHANG Y , et al . RESA: Recurrent feature-shift aggregator for lane detection [J ] . Proceedings of the AAAI Conference on Artificial Intelligence , 2021 , 35 ( 4 ): 3547 - 3554 .

CHE Q H , NGUYEN D P , PHAM M Q , et al . TwinLiteNet: An efficient and lightweight model for driveable area and lane segmentation in self-driving cars [C ] // 2023 International Conference on Multimedia Analysis and Pattern Recognition (MAPR) . Piscataway : IEEE , 2023 : 1 - 6 .

KO Y , LEE Y , AZAM S , et al . Key points estimation and point instance segmentation approach for lane detection [J ] . IEEE Transactions on Intelligent Transportation Systems , 2022 , 23 ( 7 ): 8949 - 8958 .

QU Z , JIN H , ZHOU Y , et al . Focus on local: Detecting lane marker from bottom up via key point [C ] // 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2021 : 14122 - 14130 .

WANG J S , MA Y C , HUANG S F , et al . A keypoint-based global association network for lane detection [C ] // 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2022 : 1382 - 1391 .

TABELINI L , BERRIEL R , PAIXAO T M , et al . PolyLaneNet: Lane estimation via deep polynomial regression [C ] // 2020 25th International Conference on Pattern Recognition (ICPR) . Piscataway : IEEE , 2021 : 6150 - 6156 .

FENG Z Y , GUO S H , TAN X , et al . Rethinking efficient lane detection via curve modeling [C ] // 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2022 : 17041 - 17049 .

WU Y J , ZHAO L Q , LU J W , et al . Dense hybrid proposal modulation for lane detection [J ] . IEEE Transactions on Circuits and Systems for Video Technology , 2023 , 33 ( 11 ): 6845 - 6859 .

QIN Z Q , ZHANG P Y , 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 - 2568 .

HONDA H , UCHIDA Y . CLRerNet: Improving confidence of lane detection with LaneIoU [C ] // 2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) . Piscataway : IEEE , 2024 : 1165 - 1174 .

XIAO L Y , LI X , YANG S , et al . ADNet: Lane shape prediction via anchor decomposition [C ] // 2023 IEEE/CVF International Conference on Computer Vision (ICCV) . Piscataway : IEEE , 2023 : 6381 - 6390 .

GUO M H , LU C Z , HOU Q B , et al . Segnext:Rethinking convolutional attention design for semantic segmentation [C ] // Proceedings of the IEEE Conference on Neural Information Processing Systems . Piscataway : IEEE , 2022 : 1 - 17 .

WOO S , DEBNATH S , HU R H , et al . ConvNeXt V2: Co-designing and scaling ConvNets with masked autoencoders [C ] // 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2023 : 16133 - 16142 .

DUAN K W , BAI S , XIE L X , et al . Centernet:Key point triplets for object detection [C ] // Proceedings of the IEEE Conference on Computer Vision . Piscataway : IEEE , 2019 : 6569 - 6578 .

TuSimple . Tusimple benchmark [EB/OL ] .( 2020-10-7 )[ 2024-01-03 ] . https://github.com/TuSimple/tusimple-enchmark/ https://github.com/TuSimple/tusimple-enchmark/ .

XU H , WANG S J , CAI X Y , et al . CurveLane-NAS: Unifying lane-sensitive architecture search and adaptive point blending [C ] // Computer Vision-ECCV 2020 . Cham : Springer International Publishing , 2020 : 689 - 704 .

HE K M , ZHANG X Y , REN S Q , et al . Deep residual learning for image recognition [C ] // 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2016 : 770 - 778 .

YU F , WANG D Q , SHELHAMER E , et al . Deep layer aggregation [C ] // 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway : IEEE , 2018 : 2403 - 2412 .

PHILION J . FastDraw: Addressing the long tail of lane detection by adapting a sequential prediction network [C ] // 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2019 : 11582 - 11591 .

QIN Z Q , ZHANG P Y , 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 - 2568 .

SU J M , CHEN C , ZHANG K , et al . Structure guided lane detection [C ] // Proceedings of the International Joint Conference on Artificial Intelligence . Sacramento : IJCAI , 2021 : 997 - 1003 .

KONTENTE S , ORFAIG R , BOBROVSKY B Z . CLRmatchNet: Enhancing curved lane detection with deep matching process [EB/OL ] .( 2023-09-26 )[ 2024-03-31 ] . https://arxiv.org/abs/2309.15204v2 https://arxiv.org/abs/2309.15204v2 .

LI Q K , YU X W , CHEN J X , et al . PGA-net: Polynomial global attention network with mean curvature loss for lane detection [J ] . IEEE Transactions on Intelligent Transportation Systems , 2024 , 25 ( 1 ): 417 - 429 .

TONG G X , ZU C Y . AR-NET: Lane detection model with feature balance concerns for autonomous driving [J ] . Neural Computing and Applications , 2024 , 36 ( 8 ): 3997 - 4012 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据