基于空洞-稠密网络的交通拥堵预测模型

doi:10.16183/j.cnki.jsjtu.2020.99.009

上海交通大学学报 ›› 2021, Vol. 55 ›› Issue (2): 124-130.doi: 10.16183/j.cnki.jsjtu.2020.99.009

所属专题：《上海交通大学学报》2021年12期专题汇总专辑；《上海交通大学学报》2021年“自动化技术、计算机技术”专题

基于空洞-稠密网络的交通拥堵预测模型

石敏, 蔡少委(), 易清明

暨南大学　信息科学技术学院，广州　510632

收稿日期:2019-10-29 出版日期:2021-02-01 发布日期:2021-03-03
通讯作者: 蔡少委 E-mail:caishaowei@stu2017.jnu.edu.cn
作者简介:石敏(1984-)，女，湖北省襄阳市人，副教授，主要研究方向为信号处理、图像及图像处理研究．
基金资助:
国家青年科学基金(61603153);广州市产业技术重大攻关技术项目(201802010028);广州市“羊城创新创业领军人才支持计划”(2019019)

A Traffic Congestion Prediction Model Based on Dilated-Dense Network

SHI Min, CAI Shaowei(), YI Qingming

School of Information Science and Technology, Jinan University, Guangzhou 510632, China

Received:2019-10-29 Online:2021-02-01 Published:2021-03-03
Contact: CAI Shaowei E-mail:caishaowei@stu2017.jnu.edu.cn

摘要/Abstract

摘要：

在利用卷积神经网络模型对短时交通拥堵情况等预测场景进行预测时，由于模型的卷积池化操作过程会丢失部分数据，使得目标位置的信息出现丢失及特征的分辨率持续下降，导致模型的预测能力降低．针对此，本文提出一种空洞-稠密神经网络模型．首先，利用空洞卷积用较少的网络参数获取更大感受野的特点，充分提取出复杂多变的数据时空特征．其次，通过下采样及稠密网络的等值映射，解决参数在神经网络层数增加过程出现退化的问题．最后，取实际的城市道路平均车速数据块对网络结构的有效性进行验证．结果表明：同卷积神经网络模型相比，该网络结构预测平均绝对误差降低3%～23%.

关键词: 空洞-稠密网络, 时空特征, 卷积神经网络, 短时交通拥堵预测

Abstract:

When using the convolutional neural network (CNN) model to predict short-term traffic congestion, due to the convolution pooling operation of the model, part of the data for the information of the target position will be lost, resulting in the decline of the resolution of the output features and the decrease in the predictive ability of the model. To solve this problem, this paper proposes a dilated-dense neural network model. First, it uses dilated convolution to obtain the characteristics of a larger receptive field with fewer network parameters, and fully extracts complex and variable data spatio-temporal characteristics. Then, through down-sampling and equivalent mapping of dense network, it solves the problem of parameter degradation in the process of increasing layers of neural network. Finally, it uses the actual urban road average speed data blocks to verify the validity of the model. The results show that compared with the convolutional neural network model, the average absolute error of the network structure prediction is reduced by 3% to 23%.

Key words: dilated-dense network, spatio-temporal characteristics, convolutional neural networks, short-term traffic congestion prediction

中图分类号:

TP391
U495

石敏, 蔡少委, 易清明. 基于空洞-稠密网络的交通拥堵预测模型[J]. 上海交通大学学报, 2021, 55(2): 124-130.

SHI Min, CAI Shaowei, YI Qingming. A Traffic Congestion Prediction Model Based on Dilated-Dense Network[J]. Journal of Shanghai Jiao Tong University, 2021, 55(2): 124-130.

图/表 7

图1

图2

图3

表1

图4

图5

表2

参考文献 17

[1]	CHEN P, DING C, LU G, et al. Short-term traffic states forecasting considering spatial-temporal impact on an urban expressway[J]. Transportation Research Record, 2016, 2594(1): 61-72.
[2]	ZHANG W, YU Y, QI Y. Short-term traffic flow prediction based on spatio-temporal analysis and CNN deep learning[J]. Transportation Research Record, 2019, 15 (2): 1688-1711.
[3]	陈琛，吴青，高嵩．基于时空因素的水路短期货运量预测[J]. 上海交通大学学报，2019, 53 (5): 556-562.
	CHEN Chen, WU Qing, GAO Song. Short-term shipping freight volume prediction based on temporal-spatial features [J]. Journal of Shanghai Jiao Tong University, 2019, 53(5): 556-562.
[4]	FENG X, LING X, ZHENG H, et al. Adaptive multi-kernel SVM with spatial-temporal correlation for short-term traffic flow prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2018, 20(6): 2001-2013.
[5]	罗文慧，董宝田，王泽胜．基于CNN-SVR混合深度学习模型的短时交通流预测[J]. 交通运输系统工程与信息，2017, 17(5): 68-74.
	LUO Wenhui, DONG Baotian, WANG Zesheng. Short-term traffic flow prediction based on CNN-SVR hybrid deep learning model [J]. Journal of Transportation Systems Engineering and Information Technology, 2017, 17(5): 68-74.
[6]	DENG S, JIA S, CHEN J. Exploring spatial-temporal relations via deep convolutional neural networks for traffic flow prediction with incomplete data[J]. Applied Soft Computing, 2019, 78(3): 712-721.
[7]	LIN F, XU Y, YANG Y, et al. A Spatial-temporal hybrid model for short-term traffic prediction[J]. Mathematical Problems in Engineering, 2019, 29(1): 1-12.
[8]	KANG D, LV Y, CHEN Y. Short-term traffic flow prediction with LSTM recurrent neural network[C]∥2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC). Piscataway, NJ, USA: IEEE, 2017: 1-6.
[9]	AN J, FU L, HU M, et al. A novel fuzzy-based convolutional neural network method to traffic flow prediction with uncertain traffic accident information[J]. IEEE Access, 2019, 21(1): 20708-20722.
[10]	WANG P, LI L, JIN Y, et al. Detection of unwanted traffic congestion based on existing surveillance system using in freeway via a CNN-architecture tra-fficNet [C]∥2018 13th IEEE Conference on Industrial Electronics and Applications (ICIEA). Piscataway, NJ, USA: IEEE，2018: 1134-1139.
[11]	张伟斌，余英豪，戚湧，等. 基于时空分析和CNN的交通流量短时预测方法[C]∥第十三届中国智能交通年会大会．北京：电子工业出版社，2018: 349-361.
	ZHANG Weibin, YU Yinghao, QI Yong, et al. Short-term traffic flow prediction based on spatio temporal analysis and CNN deep learning[C]∥The 13th China Intelligent Transportation Annual Meeting. Beijing: Publishing House of Electronics Industry, 2018: 349-361.
[12]	LIU Q, WANG B, ZHU Y, et al. Short-term traffic speed forecasting based on attention convolutional neural network for arterials[J]. Computer-Aided Civil and Infrastructure Engineering, 2018, 33(11): 999-1016.
[13]	YU F, KOLTUN V. Multi-scale context aggregation by dilated convolutions[C]∥International Conference on Learning Representations (ICLR). Montevideo, Uruguay: IEEE, 2016: 1-13.
[14]	HUANG G, LIU Z, VAN DER, et al. Densely connected convolutional networks[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, NJ, USA: IEEE, 2017: 2261-2269.
[15]	DOLZ J, GOPINATH K, YUAN J, et al. Hyperdense-Net: A hyper-densely connected CNN for multi-modal image segmentation[J]. IEEE Transactions on Medical Imaging, 2018, 38(5): 1-13.
[16]	ZHANG C, ZHANG H, YUAN D, et al. Citywide cellular traffic prediction based on densely connected convolutional neural networks[J]. IEEE Communications Letters, 2018, 22(8): 1656-1659.
[17]	KINGMA D K, BA J L. Adam: A method for stochastic optimization[C]∥International Conference for Learning Representations (ICLR). Montevideo, Uruguay: IEEE, 2018: 6980-6995.

道路数	Loss
	采样间隔
	1、2、3	1、2、4	1、2、5
28	0.002 3	－	－
32	0.002 6	0.002 6	－
36	0.002 9	0.003 1	0.002 7
38	0.003 4	0.003 2	0.003 0
40	0.003 7	0.003 3	0.003 3
42	0.003 9	0.003 6	0.003 5

模型	MAE			RMSE
模型	假日	正常日	测试集	假日	正常日	测试集
LeNet	3.16	3.29	3.17	4.93	5.01	4.67
空洞卷积	2.61	2.65	2.72	3.73	3.79	3.86
空洞-残差结构	2.44	2.46	2.51	3.41	3.52	3.50
空洞-稠密结构	2.35	2.43	2.44	3.32	3.49	3.43

基于空洞-稠密网络的交通拥堵预测模型

A Traffic Congestion Prediction Model Based on Dilated-Dense Network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献 17

相关文章 13

编辑推荐

Metrics

本文评价

[1]	曾国治, 魏子清, 岳宝, 丁云霄, 郑春元, 翟晓强. 基于CNN-RNN组合模型的办公建筑能耗预测[J]. 上海交通大学学报, 2022, 56(9): 1256-1261.
[2]	吴庶宸, 戚宗锋, 李建勋. 基于深度学习的智能全局灵敏度分析[J]. 上海交通大学学报, 2022, 56(7): 840-849.
[3]	全大英, 陈赟, 唐泽雨, 李世通, 汪晓锋, 金小萍. 基于双通道卷积神经网络的雷达信号识别[J]. 上海交通大学学报, 2022, 56(7): 877-885.
[4]	赵勇, 苏丹. 基于4种长短时记忆神经网络组合模型的畸形波预报[J]. 上海交通大学学报, 2022, 56(4): 516-522.
[5]	陶海红, 闫莹菲. 一种基于GA-CNN的网络化雷达节点遴选算法[J]. 空天防御, 2022, 5(1): 1-5.
[6]	武光利, 郭振洲, 李雷霆, 王成祥. 融合FCN和LSTM的视频异常事件检测[J]. 上海交通大学学报, 2021, 55(5): 607-614.
[7]	邱忠宇, 赵文龙, 高文, 潘洪涛, 史冉东. 动态视觉传感器的目标检测算法对比分析[J]. 空天防御, 2021, 4(4): 101-106.
[8]	祁生勇, 臧月进, 吕国云, 杜明. 基于生成对抗网络的空中目标图像生成算法研究[J]. 空天防御, 2021, 4(2): 67-.
[9]	薛蓉蓉, 王志武, 颜国正, 庄浩宇. 肠道机器人获取的肠道图像降噪处理方法[J]. 上海交通大学学报, 2021, 55(10): 1303-1309.
[10]	李辰, 李建勋. 卷积神经网络的正交性特征提取方法及其应用[J]. 上海交通大学学报, 2021, 55(10): 1320-1329.
[11]	章云港,杨剑锋,易本顺. 低剂量CT图像去噪的改进型残差编解码网络[J]. 上海交通大学学报, 2019, 53(8): 983-989.
[12]	颜波，张磊，褚学宁. 基于卷积神经网络的用户感知评估建模[J]. 上海交通大学学报, 2019, 53(7): 844-851.
[13]	李洋洋，史历程，万卫兵，赵群飞. 基于卷积神经网络的三维物体检测方法[J]. 上海交通大学学报（自然版）, 2018, 52(1): 7-12.