Open Access
ARTICLE
GACNet: A Generative Adversarial Capsule Network for Regional Epitaxial Traffic Flow Prediction
Jinyuan Li1, Hao Li1, Guorong Cui1, Yan Kang1, *, Yang Hu1, Yingnan Zhou2
1 National Pilot School of Software, Yunnan University, Kunming, 650091, China.
2 School of Engineering and Applied Science, The George Washington University, Washington DC, 20052, USA.
* Corresponding Author: Yan Kang. Email: .
Computers, Materials & Continua 2020, 64(2), 925-940. https://doi.org/10.32604/cmc.2020.09903
Received 25 January 2020; Accepted 28 February 2020; Issue published 10 June 2020
Abstract
With continuous urbanization, cities are undergoing a sharp expansion within
the regional space. Due to the high cost, the prediction of regional traffic flow is more
difficult to extend to entire urban areas. To address this challenging problem, we present
a new deep learning architecture for regional epitaxial traffic flow prediction called
GACNet, which predicts traffic flow of surrounding areas based on inflow and outflow
information in central area. The method is data-driven, and the spatial relationship of
traffic flow is characterized by dynamically transforming traffic information into images
through a two-dimensional matrix. We introduce adversarial training to improve
performance of prediction and enhance the robustness. The generator mainly consists of
two parts: abstract traffic feature extraction in the central region and traffic prediction in
the extended region. In particular, the feature extraction part captures nonlinear spatial
dependence using gated convolution, and replaces the maximum pooling operation with
dynamic routing, finally aggregates multidimensional information in capsule form. The
effectiveness of the method is evaluated using traffic flow datasets for two real traffic
networks: Beijing and New York. Experiments on highly challenging datasets show that
our method performs well for this task.
Keywords
Cite This Article
APA Style
Li, J., Li, H., Cui, G., Kang, Y., Hu, Y. et al. (2020). Gacnet: A generative adversarial capsule network for regional epitaxial traffic flow prediction. Computers, Materials & Continua, 64(2), 925-940. https://doi.org/10.32604/cmc.2020.09903
Vancouver Style
Li J, Li H, Cui G, Kang Y, Hu Y, Zhou Y. Gacnet: A generative adversarial capsule network for regional epitaxial traffic flow prediction. Comput Mater Contin. 2020;64(2):925-940 https://doi.org/10.32604/cmc.2020.09903
IEEE Style
J. Li, H. Li, G. Cui, Y. Kang, Y. Hu, and Y. Zhou "GACNet: A Generative Adversarial Capsule Network for Regional Epitaxial Traffic Flow Prediction," Comput. Mater. Contin., vol. 64, no. 2, pp. 925-940. 2020. https://doi.org/10.32604/cmc.2020.09903
Citations