[1]蔡英凤,张田田,王海,等.基于实例分割和自适应透视变换算法的多车道线检测[J].东南大学学报(自然科学版),2020,50(4):775-781.[doi:10.3969/j.issn.1001-0505.2020.04.023]
 Cai Yingfeng,Zhang Tiantian,Wang Hai,et al.Multi-lane detection based on instance segmentation and adaptive perspective transformation[J].Journal of Southeast University (Natural Science Edition),2020,50(4):775-781.[doi:10.3969/j.issn.1001-0505.2020.04.023]
点击复制

基于实例分割和自适应透视变换算法的多车道线检测()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
50
期数:
2020年第4期
页码:
775-781
栏目:
交通运输工程
出版日期:
2020-07-20

文章信息/Info

Title:
Multi-lane detection based on instance segmentation and adaptive perspective transformation
作者:
蔡英凤1张田田1王海2李祎承1孙晓强1陈龙1
1江苏大学汽车与交通工程学院, 镇江 212013; 2江苏大学汽车工程研究院, 镇江 212013
Author(s):
Cai Yingfeng1 Zhang Tiantian1 Wang Hai2 Li Yicheng1 Sun Xiaoqiang1 Chen Long1
1Institute of Automotive Engineering, Jiangsu University, Zhenjiang 212013, China
2School of Automotive and Traffic Engineering, Jiangsu University, Zhenjiang 212013, China
关键词:
车道线 深度学习 实例分割 自适应透视变换
Keywords:
lane line deep learning instance segmentation adaptive perspective transformation
分类号:
U142.6
DOI:
10.3969/j.issn.1001-0505.2020.04.023
摘要:
为了解决传统的车道线检测算法对光照变化、阴影遮挡等环境干扰较为敏感而导致鲁棒性不足的问题,提出了一种基于实例分割和自适应透视变换算法的多车道线检测方法.该算法首先通过设计的多分支实例分割网络实现多车道线分割,该多分支实例分割网络包括车道线语义分割分支和车道线Id分支;再应用自适应透视变换模型获得鸟瞰图视角下的实例分割后的车道线像素点集合;最后利用最小二乘法二阶多项式完成车道线像素点的拟合.基于Culane车道线数据集进行训练及验证,验证表明,每帧图片检测用时约28 ms,车道线检测准确率达91.4%.将车道线检测模型集成到实车ROS平台进行测试,测试表明,所提算法能够实现各类复杂交通场景下的多车道线实时检测.
Abstract:
To solve the problem that the traditional lane detection algorithm is sensitive to the environment interference such as illumination and occlusion, which leads to insufficient robustness, a multi-lane detection method based on instance segmentation and the adaptive transformation algorithm is proposed. Firstly, the multi-branch instance segmentation network is designed to achieve multi-lane line segmentation, and the multi-branch instance segmentation network includes lane line semantic segmentation branch and lane line Id branch. Then, an adaptive perspective transformation model is employed to obtain the set of lane pixel points from the bird’s eye view. Finally, the lane line fitting is completed by using the least square method. Using the Culane dataset for training and verification, each frame takes about 28 ms, and the detection rate is 91.4%. Integrating the lane detection model into the ROS(Robot Operating Systems)platform, the experimental results show that the proposed algorithm can achieve the real-time detection of multiple lane lines in various complex traffic scenarios.

参考文献/References:

[1] Chiu K Y, Lin S F. Lane detection using color-based segmentation[C]//IEEE Intelligent Vehicles Symposium. Las Vegas, USA, 2005: 706-711. DOI:10.1109/IVS.2005.1505186.
[2] Guo Y T, Zhang Y J, Liu S, et al. Robust and real-time lane marking detection for embedded system[J]. Lecture Notes in Computer Science, 2015: 223-235.DOI:10.1007/978-3-319-21969-1_20.
[3] Liu G, Li S, Liu W. Lane detection algorithm based on local feature extraction[C]//Chinese Automation Congress. Changsha, China, 2013: 59-64. DOI:10.1109/CAC.2013.6775702.
[4] Wang Y, Teoh E K, Shen D G. Lane detection and tracking using B-snake[J]. Image and Vision Computing, 2004, 22(4): 269-280. DOI:10.1016/j.imavis.2003.10.003.
[5] John V, Liu Z, Guo C, et al. Real-time lane estimation using deep features and extra trees regression[J]. Image and Video Technology, 2015, 9431:721-733. DOI:10.1007/978-3-319-29451-3_57.
[6] He B, Ai R, Yan Y, et al. Accurate and robust lane detection based on dual-view convolutional neutral network[C]//IEEE Intelligent Vehicles Symposium. Gothenburg, Sweden, 2016: 1041-1046.
[7] Denison D G T, Mallick B K, Smith A F M. Automatic Bayesian curve fitting[J].Journal of the Royal Statistical Society: Series B(Statistical Methodology), 1998, 60(2): 333-350. DOI:10.1111/1467-9868.00128.
[8] Zheng W N, Bo P B, Liu Y, et al. Fast B-spline curve fitting by L-BFGS[J].Computer Aided Geometric Design, 2012, 29(7): 448-462. DOI:10.1016/j.cagd.2012.03.004.
[9] 权开波,贾宁,杜培寿.基于最小二乘法的曲线拟合[J].商,2015(3):296.
[10] Mallot H A, Bülthoff H H, Little J J, et al. Inverse perspective mapping simplifies optical flow computation and obstacle detection[J].Biological Cybernetics, 1991, 64(3): 177-185. DOI:10.1007/BF00201978.
[11] Aly M. Real time detection of lane markers in urban streets[J]. Computer Science, 2014, 65(8): 928-947.
[12] Guo C, Meguro J I, Kojima Y, et al. Automatic lane-level map generation for advanced driver assistance systems using low-cost sensors[C]//IEEE International Conference on Robotics and Automation. Hong Kong, China, 2014: 3975-3982. DOI:10.1109/ICRA.2014.6907436.
[13] Jeong J, Kim A. Adaptive inverse perspective mapping for lane map generation with SLAM[C]//International Conference on Ubiquitous Robots and Ambient Intelligence. Xi’an, China, 2016: 38-41. DOI:10.1109/URAI.2016.7734016.
[14] Dupourque V. A robot operating system[C]//IEEE International Conference on Robotics and Automation. Taipei, China, 2003: 342-348. DOI:10.1109/ROBOT.1984.1087185.
[15] Farahnak-Ghazani F, Baghshah M S. Multi-label classification with feature-aware implicit encoding and generalized cross-entropy loss[C]//2016 24th Iranian Conference on Electrical Engineering(ICEE). Shiraz, Iran, 2016: 1574-1579. DOI:10.1109/IranianCEE.2016.7585772.
[16] Paszke A, Chaurasia A, Kim S, et al. ENet: A deep neural network architecture for real-time semantic segmentation[EB/OL].(2016-06-07)[2020-01-10]. https://arxiv.org/abs/1606.02147.
[17] Tran T N, Drab K, Daszykowski M. Revised DBSCAN algorithm to cluster data with dense adjacent clusters[J].Chemometrics and Intelligent Laboratory Systems, 2013, 120: 92-96. DOI:10.1016/j.chemolab.2012.11.006.
[18] Pan X G, Shi J P, Luo P, et al. Spatial as deep: Spatial CNN for traffic scene understanding[EB/OL].(2017-12-17)[2020-01-10]. https://arxiv.org/abs/1712.06080.
[19] Mongkonyong P, Nuthong C, Siddhichai S, et al. Lane detection using randomized Hough transform[J].IOP Conference Series: Materials Science and Engineering, 2018, 297: 012050. DOI:10.1088/1757-899x/297/1/012050.
[20] Saad A, Liljenquist J. A multi-robot testbed for robotics programming education and research[C]//Proceedings of the 2014 ACM Southeast Regional Conference. Kennesaw, Georgia, USA, 2014: 1-4. DOI:10.1145/2638404.2675737.

相似文献/References:

[1]程旭,张毅锋,刘袁,等.基于深度特征的目标跟踪算法[J].东南大学学报(自然科学版),2017,47(1):1.[doi:10.3969/j.issn.1001-0505.2017.01.001]
 Cheng Xu,Zhang Yifeng,Liu Yuan,et al.Object tracking algorithm based on deep feature[J].Journal of Southeast University (Natural Science Edition),2017,47(4):1.[doi:10.3969/j.issn.1001-0505.2017.01.001]
[2]李林超,曲栩,张健,等.基于特征级融合的高速公路异质交通流数据修复方法[J].东南大学学报(自然科学版),2018,48(5):972.[doi:10.3969/j.issn.1001-0505.2018.05.029]
 Li Linchao,Qu Xu,Zhang Jian,et al.Missing value imputation method for heterogeneous traffic flow data based on feature fusion[J].Journal of Southeast University (Natural Science Edition),2018,48(4):972.[doi:10.3969/j.issn.1001-0505.2018.05.029]
[3]周正东,李剑波,辛润超,等.基于带孔U-net神经网络的肺癌危及器官并行分割方法[J].东南大学学报(自然科学版),2019,49(2):231.[doi:10.3969/j.issn.1001-0505.2019.02.005]
 Zhou Zhengdong,Li Jianbo,Xin Runchao,et al.Parallel segmentation method for organs at risk in lung cancer based on dilated U-net neural network[J].Journal of Southeast University (Natural Science Edition),2019,49(4):231.[doi:10.3969/j.issn.1001-0505.2019.02.005]
[4]彭博,蔡晓禹,唐聚,等.基于改进Faster R-CNN的无人机视频车辆自动检测[J].东南大学学报(自然科学版),2019,49(6):1199.[doi:10.3969/j.issn.1001-0505.2019.06.025]
 Peng Bo,Cai Xiaoyu,Tang Ju,et al.Automatic vehicle detection with UAV videos based on modified Faster R-CNN[J].Journal of Southeast University (Natural Science Edition),2019,49(4):1199.[doi:10.3969/j.issn.1001-0505.2019.06.025]

备注/Memo

备注/Memo:
收稿日期: 2020-02-16.
作者简介: 蔡英凤(1985—),女,博士,教授,博士生导师,caicaixiao0304@126.com.
基金项目: 国家重点研发计划资助项目(2017YFB0102603, 2018YFB0105003)、国家自然科学基金资助项目(51875255, 61601203, 61773184, U1564201, U1664258, U1764257, U1762264)、江苏省自然科学基金资助项目(BK20180100)、江苏省战略性新兴产业发展重大专项资助项目(苏发改高技发(2016)1094号)、镇江市重点研发计划资助项目(GY2017006).
引用本文: 蔡英凤,张田田,王海,等.基于实例分割和自适应透视变换算法的多车道线检测[J].东南大学学报(自然科学版),2020,50(4):775-781. DOI:10.3969/j.issn.1001-0505.2020.04.023.
更新日期/Last Update: 2020-07-20