Abstract：Non-motorized vehicle travel has gradually revived in China to form complex, mixed traffic flows in recent years that lead to traffic movement and safety problems. The characteristics of the mixed non-motorized vehicle speeds need to be accurately modeled to characterize this problem. Deep neural networks are used here for multi-target tracking in videos of the non-motorized vehicle lanes to measure the non-motorized vehicle speeds. A statistical analysis of the vehicle speeds is used to determine the number of components of the Gaussian mixture model for the speeds using information criteria with the maximum likelihood estimate of each parameter calculated using the expectation maximization algorithm. Then, the model parameters are related to the road operating conditions and statistical characteristics. This intelligent video recognition method accelerates data collection to obtain sufficient data and the component determination prior to other parameters improves the modeling efficiency. The fitting results show that the Gaussian mixture model more realistically describes the speed distribution of non-motorized vehicles than the single distribution model. The Gaussian mixture model results are divided into the various types of non-motorized vehicles when the vehicle flow is not obstructed and the parameters are related to the average speed, the standard deviation and the ratio of different vehicle types. The classifications according to flow states are consistent during peak periods with the mean speed of the faster component close to the free flow velocity of the vehicles.
刘贺子, 陈涛. 基于视频识别的混合非机动车速度分布模型[J]. 清华大学学报（自然科学版）, 2021, 61(2): 144-151.
LIU Hezi, CHEN Tao. Speed distribution model of mixed non-motorized vehicles based on video recognition. Journal of Tsinghua University(Science and Technology), 2021, 61(2): 144-151.
[1] 聂星. 中国城市非机动车交通发展现状研究[J]. 中国新技术新产品, 2016(21):126.NIE X. Research on the development of non-motorized vehicle transportation in Chinese cities[J]. New Technologies & New Products of China, 2016(21):126. (in Chinese) [2] 中华人民共和国统计局. 中国统计年鉴[M]. 北京:中国统计出版社, 2019.Statistics Bureau of the People's Republic of China. China statistical yearbook[M]. Beijing:China Statistics Press, 2019. (in Chinese) [3] YANG J, HU Y H, DU W, et al. Unsafe riding practice among electric bikers in Suzhou, China:An observational study[J]. BMJ Open, 2014, 4(1):e003902. [4] 陶思然. 基于自行车与电动自行车的二元混合交通流特性研究[D]. 西安:长安大学, 2015.TAO S R. Binary mixed traffic characteristics based on bicycle and electric bicycle[D]. Xi'an:Chang'an University, 2015. (in Chinese) [5] KATTI V K, RAGHAVACHARI S. Modeling of mixed traffic with speed data as inputs for the traffic simulation models[M]//Highway Research Bulletin 28. New Delhi, India:Indian Roads Congress, 1986:35-48. [6] ALLEN D P, ROUPHAIL N, HUMMER J E, et al. Operational analysis of uninterrupted bicycle facilities[J]. Transportation Research Record:Journal of the Transportation Research Board, 1998, 1636(1):29-36. [7] CHERRY C R. Electric two-wheelers in China:Analysis of environmental, safety and mobility impacts[R]. Berkeley, USA:University of California, 2007:101. [8] 陶志兴. 机非混行路段交通流特性研究[D]. 长春:吉林大学, 2007.TAO Z X. Research on characteristics of mixed vehicle-bicycle traffic flow on road section[D]. Changchun:Jilin University, 2007. (in Chinese) [9] LIN S, HE M, TAN Y L, et al. Comparison study on operating speeds of electric bicycles and bicycles:Experience from field investigation in Kunming[J]. Transportation Research Record:Journal of the Transportation Research Board, 2008, 2048(1):52-59. [10] 平萍. 城市路段非机动车流交通特性研究[D]. 镇江:江苏大学, 2018.PING P. Research on traffic characteristics of non-motorized vehicle flow in urban section[D]. Zhenjiang:Jiangsu University, 2018. (in Chinese) [11] 徐程, 曲昭伟, 王殿海, 等. 混合自行车交通流速度分布模型[J]. 浙江大学学报(工学版), 2017, 51(7):1331-1338.XU C, QU Z W, WANG D H, et al. Speed distribution model for heterogeneous bicycle traffic flow[J]. Journal of Zhejiang University (Engineering Science), 2017, 51(7):1331-1338. (in Chinese) [12] 梁春岩. 自行车交通流特性及其应用研究[D]. 长春:吉林大学, 2005.LIANG C Y. Study on characteristics and application of bicycle traffic flow[D]. Changchun:Jilin University, 2005. (in Chinese) [13] 王丹. 路段非机动车交通流特性研究[D]. 西安:长安大学, 2014.WANG D. Research on characteristics of non-motor vehicle traffic flow on road section[D]. Xi'an:Chang'an University, 2014. (in Chinese) [14] 李海航. 城市非机动车道通行能力研究[D]. 石家庄:石家庄铁道大学, 2015.LI H H. Research on the capacity of non-motorized vehicle lane of city[D]. Shijiazhuang:Shijiazhuang Tiedao University, 2015. (in Chinese) [15] 周旦. 城市基本路段混合非机动车交通流特性研究[D]. 杭州:浙江大学, 2016.ZHOU D. Study on characterisitcs of mixed bicycle traffic flow in basic sections of urban road[D]. Hangzhou:Zhejiang University, 2016. (in Chinese) [16] BEWLEY A, GE Z Y, OTT L, et al. Simple online and realtime tracking[C]//Proceedings of 2016 IEEE International Conference on Image Processing. Phoenix, USA:IEEE, 2016:3464-3468. [17] REDMON J, FARHADI A. YOLOv3:An incremental improvement[Z/OL]. arXiv:1804.02767, 2018. [18] ZAGORUYKO S, KOMODAKIS N. Wide residual networks[C]//Proceedings of the British Machine Vision Conference 2016. York, UK:BMVA Press, 2016:1-12. [19] WOJKE N, BEWLEY A. Deep cosine metric learning for person re-identification[C]//Proceedings of 2018 IEEE Winter Conference on Applications of Computer Vision. Lake Tahoe, USA, 2018:00087. [20] ZHENG L, SHEN L Y, TIAN L, et al. Scalable person re-identification:A benchmark[C]//Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago, Chile, 2015:1116-1124. [21] 何民. 混合交通流微观仿真关键技术研究[D]. 北京:北京工业大学, 2003.HE M. Research on key technologies of microscopic simulation of mixed traffic flow[D]. Beijing:Beijing Industry University, 2003. (in Chinese) [22] COIFMAN B. Revisiting the empirical fundamental relationship[J]. Transportation Research Part B:Methodological, 2014, 68:173-184. [23] WANG H Z, LI J, CHEN Q Y, et al. Logistic modeling of the equilibrium speed-density relationship[J]. Transportation Research Part A:Policy and Practice, 2011, 45(6):554-566.
2021, Vol. 61 
