In this paper, we present a deep learning based approach to performing the whole-day prediction of the traffic speed for the elevated highway. In order to learn the temporal features of traffic speed data in a hierarchical way, an improved convolutional neural network (CNN) with asymmetric kernels is proposed. Speed data are collected from loop detectors of Yan’an elevated highway of Shanghai. To test the performance of the presented method, we compare it with some conventional approaches of traffic speed estimation. Experimental results demonstrate that our method outperforms all of them.
CITATION STYLE
Zang, D., Ling, J., Cheng, J., Tang, K., & Li, X. (2017). Using convolutional neural network with asymmetrical kernels to predict speed of elevated highway. In IFIP Advances in Information and Communication Technology (Vol. 510, pp. 212–221). Springer New York LLC. https://doi.org/10.1007/978-3-319-68121-4_22
Mendeley helps you to discover research relevant for your work.