| 366 | 8 | 174 |
| 下载次数 | 被引频次 | 阅读次数 |
提出了一种基于多模态融合的人群流量预测算法(multimodal fusion for crowd flow prediction, MFCFP)。首先使用图卷积算子来探索区域之间的相关性以建立多模态,不同的模态可以捕捉不同的影响因素。然后进行多模态融合,并将带有注意力机制的基于图卷积神经网络应用于本文模型,以更好地建立相关区域关联。在真实数据集的实验证明了所提模型可以准确地预测人群活动流量。
Abstract：A multimodal fusion based crowd flow prediction algorithm called multimodal fusion for crowd flow prediction(MFCFP) was proposed. Firstly, the graph convolution operator was used to explore the correlation between regions to establish multi-modality. Different modalities could capture different influencing factors. Then, multi-modal fusion was performed, and the graph-based convolutional neural network with attention mechanism was applied to the model to better establish the correlation of related regions. Finally, the experiments on the real Shanghai data set proved that the proposed model could accurately predict the flow of crowd activities.
[1] LIU J M,SUN L L,LI Q,et al.Functional zone based hierarchical demand prediction for bike system expansion[C]//Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York:ACM Press,2017:957-966.
[2] 肖立志，张争.基于递归卷积神经网络的行人检测方法[J].信阳师范学院学报(自然科学版),2021,34(4):655-660.XIAO L Z,ZHANG Z.Pedestrian detection method based on recurrent convolutional neural networks[J].Journal of Xinyang normal university (natural science edition),2021,34(4):655-660.
[3] 黄宗超，王思为，祝恩，等.基于子空间融合的多视图聚类算法[J].郑州大学学报(理学版),2021,53(1):68-73.HUANG Z C,WANG S W,ZHU E,et al.Multi-view clustering algorithm based on subspace fusion[J].Journal of Zhengzhou university (natural science edition),2021,53(1):68-73.
[4] ASIF M T,DAUWELS J,GOH C Y,et al.Spatiotemporal patterns in large-scale traffic speed prediction[J].IEEE transactions on intelligent transportation systems,2014,15(2):794-804.
[5] CHEN L B,ZHANG D Q,WANG L Y,et al.Dynamic cluster-based over-demand prediction in bike sharing systems[C]//Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing.New York:ACM Press,2016:841-852.
[6] KWON J,VARAIYA P.Effectiveness of California′s high occupancy vehicle (HOV) system[J].Transportation research part C:emerging technologies,2008,16(1):98-115.
[7] BRUNA J,ZAREMBA W,SZLAM A,et al.Spectral networks and locally connected networks on graphs[C]//International Conference on Learning Representations.San Diego:ICLR Press,2014:1-14.
[8] CHAI D,WANG L Y,YANG Q.Bike flow prediction with multi-graph convolutional networks[C]//Proceedings of the 26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems.New York:ACM Press,2018:397-400.
[9] LI Y G,YU R,SHAHABI C,et al.Diffusion convolutional recurrent neural network:data-driven traffic forecasting[C]//International Conference on Learning Representations.San Diego:ICLR Press,2018:1-16.
[10] SUN J K,ZHANG J B,LI Q F,et al.Predicting citywide crowd flows in irregular regions using multi-view graph convolutional networks[EB/OL].(2020-07-13)[2021-01-10].DOI:10.1109/TKDE.2020.3081717.
[11] KUMAR S V,VANAJAKSHI L.Short-term traffic flow prediction using seasonal ARIMA model with limited input data[J].European transport research review,2015,7(3):1-9.
基本信息:
DOI：10.13705/j.issn.1671-6841.2021081
中图分类号:TP311.13;TP18
引用信息:
[1]刘玉强,顾晶晶,孙明,等.基于城市区域多模态融合的人群流量预测[J],2022,54(01):25-31.DOI:10.13705/j.issn.1671-6841.2021081.
基金信息:
国家自然科学基金项目(62072235);; 南京航空航天大学2020研究生创新基地开放基金项目(Kfjj20191605)
2021-03-07
2021
2021-03-11
2021
1