| Title | Real Time Traffic Flow Monitoring and Congestion Prediction Driven by Deep Learning |
| Creator | |
| Date Issued | 2023-12-08 |
| Source Publication | ACM International Conference Proceeding Series
 |
| Pages | 406-410 |
| Abstract | Spatiotemporal data, such that collected from road traffic monitoring and congestion prediction, exhibits temporal and geographical relationships. This investigation employs a two-pronged strategy, first investigating spatiotemporal characteristics and then creating a model for traffic flow monitoring and congestion prediction based on a deep neural network. With the use of a graph convolutional neural network and an attention mechanism, this research is the first to propose a method for learning spatial features of traffic flows. To improve our expression and the capacity to extract relevant spatial attributes, we introduce node adaptive learning and apply different weights to the degree of mutual influence across different nodes. Furthermore, we present a temporal convolutional network-based learning approach for temporal features of traffic flow, which uses causal convolution to guarantee that input and output data dimensions are consistent. Long-length spatiotemporal sequence data benefit greatly from the dilated convolution's ability to dynamically regulate the receptive field by adjusting the sampling interval. Using spatiotemporal graphs, a system is developed to monitor and foresee traffic congestion. In order to learn feature information, mode-specific parameter values, and overall model performance, this model combines a graph convolutional neural network with an attention mechanism. |
| Keyword | Congestion prediction Deep learning Neural network Traffic flow monitoring |
| DOI | 10.1145/3641343.3641426 |
| URL | View source |
| Language | 英语English |
| Scopus ID | 2-s2.0-85192548901 |
| Citation statistics | |
| Document Type | Conference paper |
| Identifier | http://repository.uic.edu.cn/handle/39GCC9TT/11554 |
| Collection | Beijing Normal-Hong Kong Baptist University |
| Corresponding Author | Wu，Yuqian |
| Affiliation | Beijing Normal University-Hong Kong Baptist University United International College,Zhuhai,Guangdong,519087,China |
| First Author Affilication | Beijing Normal-Hong Kong Baptist University |
| Corresponding Author Affilication | Beijing Normal-Hong Kong Baptist University |
| Recommended Citation GB/T 7714 | Wu，Yuqian. Real Time Traffic Flow Monitoring and Congestion Prediction Driven by Deep Learning[C], 2023: 406-410. |
| Files in This Item: | There are no files associated with this item. | |||||
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
BNBU 
Library
Edit Comment