山东大学学报 (工学版) ›› 2024, Vol. 54 ›› Issue (5): 42-49.doi: 10.6040/j.issn.1672-3961.0.2023.314
王佳如1,吕斌1*,吴建清2,王志勇1
WANG Jiaru1, LÜ Bin1*, WU Jianqing2, WANG Zhiyong1
摘要： 针对交叉口排队长度实时感知的问题,提出一种结合交通数学模型与智能感知设备检测的排队长度感知方法。通过冲击波模型确定道路最大排队长度,将其作为以YOLOv5-DeepSORT为基础的视频感知模型的输入,实现交通数学模型与智能感知模型的单向耦合。为验证该方法的有效性和优越性,以兰州市某交叉口为例进行排队长度的实时感知,并在选定交叉口调查不同时间段的感知数据,模拟不同交叉口交通流量的差异对本研究方法感知精度的影响进行探究。研究结果表明,基于冲击波模型与YOLOv5-DeepSORT单向耦合的排队长度感知方法确定的排队长度检测区域在整体感知精度上优于对照组,平均绝对误差、均方根误差以及平均绝对百分比误差等均得到了有效降低,部分工况下精度提高40%以上。
中图分类号:
| [1] 吴建清, 宋修广. 智慧公路关键技术发展综述[J]. 山东大学学报(工学版), 2020, 50(4):52-69. WU Jianqing, SONG Xiuguang. Review on smart hghways critical technology[J]. Journal of Shandong University(Engineering Science), 2020, 50(4): 52-69. [2] 周晓昕, 廖祝华, 刘毅志, 等. 融合历史与当前交通流量的信号控制方法[J]. 山东大学学报(工学版), 2023, 53(4):48-55. ZHOU Xiaoxin, LIAO Zhuhua, LIU Yizhi, et al. Signal control method integrating history and current traffic flow[J]. Journal of Shandong University(Engineering Science), 2023, 53(4): 48-55. [3] LIU H X, WU X, MA W, et al. Realtime queue length estimation for congested signalized intersections[J].Transportation Research Part C: Emerging Technologies, 2009, 17(4): 412-427. [4] 施康, 杨晓光, 王一喆. 基于数据的交叉口车辆排队状态感知方法研究[J]. 公路交通科技, 2022, 39(1):114-119. SHI Kang, YANG Xiaoguang, WANG Yizhe. Study on vehicle queuing state perception method at intersection based on data[J]. Journal of Highway and Transportation Research and Development, 2022, 39(1):114-119. [5] YI P, TIAN Z Z, ZHAO Q. Consistency of input-ou-tput model and shockwave analysis in queue and delay estimations[J]. Journal of Transportation Systems Engineering and Information Technology, 2008, 6(8): 146-152. [6] VIGOS G, PAPAGEORGIOU M, WANG Y. Real-time estimation of vehicle-count within signalized links[J]. Transportation Research Part C: Emerging Technologies, 2008, 16(1): 18-35. [7] 羊钊, 刘攀, 朱仁伟, 等. 基于冲击波理论的信号交叉口最大广义排队长度计算方法[J].长安大学学报(自然科学版),2015, 35(增刊1):154-159. YANG Zhao, LIU Pan, ZHU Renwei, et al. Estimation of the maximum queue length at signalized intersections based on shockwave theory[J]. Journal of Chang'an University(Natural Science Edition), 2015, 35(Suppl.1): 154-159. [8] ZHAN X, LI R, UKKUSURI S V. Lane-based realtime queue length estimation using license plate recognition data[J]. Transportation Research Part C: Emerging Technologies, 2015, 57: 85-102. [9] PUDASAINI P, KARIMPOUR A, WU Y J. Realtime queue length estimation for signalized intersections using single-channel advance detector data[J]. Transportation Research Record, 2023, 2677(7): 144-156. [10] WU J, XU H, ZHENG Y, et al. A novel method of vehicle-pedestrian near-crash identification with roadside LiDAR data[J]. Accident Analysis & Prevention, 2018, 121: 238-249. [11] WU J, XU H, TIAN Y, et al. An automatic lane identification method for the roadside light detection and ranging sensor[J]. Journal of Intelligent Transportation Systems, 2020, 24(5): 467-479. [12] 连丽容, 罗文婷, 秦勇, 等. 双目机器视觉及Retina-Net模型的路侧行人感知定位[J]. 中国图象图形学报, 2021, 26(12):2941-2952. LIAN Lirong, LUO Wenting, QIN Yong, et al.Ro-adside pedestrian detection and location based onbi-nocular machine vision and RetinaNet[J]. Journal of Image and Graphics, 2021, 26(12):2941-2952. [13] TIAPRASERT K, ZHANG Y, WANG X B, et al. Queue length estimation using connected vehicle te-chnology for adaptive signal control[J]. IEEE Tran-sactions on Intelligent Transportation Systems, 2015, 16(4): 2129-2140. [14] 吴浩, 刘磊, 唐克双. 基于集成学习的信号控制交叉口排队长度估计[J]. 同济大学学报(自然科学版), 2023, 51(3):405-415. WU Hao, LIU Lei, TANG Keshuang.Queue length estimation at signalized intersection based on ense-mble Learning[J]. Journal of Tongji University(Natural Science), 2023, 51(3):405-415. [15] ZANIN M, MESSELODI S, MODENA C M. An efficient vehicle queue detection system based on image processing[C] //12th International Conference on Image Analysis and Processing(ICIAP). Mantova, Italy: IEEE, 2003: 232-237. [16] ZHU L, KHORAMSHAHI E, TURPPA T, et al. Traffic queue length measurement by using combined methods of Photogrammetry and digital image processing[C] //2015 Joint Urban Remote Sensing Event(JURSE). Lausanne, Switzerland: IEEE, 2015: 1-4. [17] 余志, 黄柳红, 李熙莹, 等. 基于视频的交叉口排队过程感知及预测[J]. 交通运输系统工程与信息, 2020, 20(1):33-39. YU Zhi, HUANG Liuhong, LI Xiying, et al. Queueing process sensing and prediction at intersection based on video[J]. Journal of Transportation Systems Engineering and Information Technology, 2020, 20(1):33-39. [18] 宋士奇, 朴燕, 蒋泽新. 基于改进YOLOv3的复杂场景车辆分类与跟踪[J]. 山东大学学报(工学版), 2020, 50(2):27-33. SONG Shiqi, PIAO Yan, JIANG Zexin.Vehicle classification and tracking for complex scenes based on improved YOLOv3[J]. Journal of Shandong University(Engineering Science), 2020, 50(2):27-33. [19] 毛昭勇, 王亦晨, 王鑫, 等. 面向高速公路的车辆视频监控分析系统[J]. 西安电子科技大学学报, 2021, 48(5):178-189. MAO Zhaoyong, WANG Yichen, WANG Xin, et al.Vehicle video surveillance and analysis system for the expressway[J]. Journal of Xidian University, 2021, 48(5):178-189. [20] UMAIR M, FAROOQ M U, RAZA R H, et al. Efficient video-based vehicle queue length estimation using computer vision and deep learning for an urban traffic scenario[J]. Processes, 2021, 9(10): 1786. [21] WOJKE N, BEWLEY A, PAULUS D. Simple onlyne and realtime tracking with a deep association metric[C] //2017 IEEE International Conference on Image Processing(ICIP). Beijing, China: IEEE, 2017: 3645-3649. [22] TAN C, YAO J, TANG K, et al. Cycle-based queue length estimation for signalized intersections using sparse vehicle trajectory data[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 22(1): 91-106. |
| [1] | 邹正标,刘毅志,廖祝华,赵肄江. 动态交通流量预测的时空注意力图卷积网络[J]. 山东大学学报 (工学版), 2024, 54(5): 50-61. |
| [2] | 仕小伟,朱文兴*,王青燕,邵士雨. 城市主干路交通溢流发生机理建模及其仿真[J]. 山东大学学报(工学版), 2013, 43(3): 43-48. |
| [3] | 林 洁,杨立才,吴晓晴,叶 杨 . 求解动态路径诱导K路最短问题的人工免疫优化方法[J]. 山东大学学报(工学版), 2007, 37(2): 103-108 . |
|
