| 引用本文: | 童凯南,林友芳,刘军,等.面向时空交通栅格流量预测的3D通道注意力网络.[J].国防科技大学学报,2022,44(3):41-49.[点击复制] |
| TONG Kainan,LIN Youfang,LIU Jun,et al.3D channel-wise attention network for spatio-temporal traffic raster flow prediction[J].Journal of National University of Defense Technology,2022,44(3):41-49[点击复制] |
|
| |
|
|
| 本文已被：浏览 5241次 下载 3732次 |
| 面向时空交通栅格流量预测的3D通道注意力网络 |
| 童凯南1,林友芳1,刘军2,郭晟楠1,万怀宇1 |
|
(1. 北京交通大学 计算机与信息技术学院, 北京 100044;2. 中国民航信息网络股份有限公司, 北京 101318)
|
| 摘要: |
| 城市交通流量预测对交通管理和公共安全具有重要意义。然而,交通栅格流量数据中的规律在时刻变化,在城市中存在全局范围的时空间关系,并且不同特征通道在每个城市区域上有不同的重要性。为解决这些挑战并做出更准确的预测,设计了一种新颖的时空神经网络模型——3D通道注意力网络(three-dimensional channel-wise attention networks,3D-CANet)。提出一个3D通道内注意力(three-dimensional inner channel attention,3D-InnerCA)单元来动态捕获各个通道中不同的全局时空相关性,同时设计通道间注意力(inter channel attention,InterCA)单元来自适应地重校准每个区域上不同特征通道的贡献。在3个真实交通栅格流量数据集上的实验结果表明,3D-CANet模型的预测能力优于其他对比方法,证明了模型的有效性。 |
| 关键词: 时空数据 交通栅格流量 3D通道注意力 通道内注意力 通道间注意力 |
| DOI：10.11887/j.cn.202203006 |
| 投稿日期：2021-06-16 |
| 基金项目:中国博士后科学基金资助项目(2021M700365) |
|
| 3D channel-wise attention network for spatio-temporal traffic raster flow prediction |
| TONG Kainan1, LIN Youfang1, LIU Jun2, GUO Shengnan1, WAN Huaiyu1 |
|
(1. School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China;2. TravelSky Technology Limited, Beijing 101318, China)
|
| Abstract: |
| Urban traffic flow forecasting is of great significance for traffic management and public safety. However, the correlations of traffic raster flow change with time. There are global spatio-temporal correlations in the city, and the contributions of channel-wise features vary on each city region. To tackle these challenges and make more accurate prediction, a novel spatio-temporal neural network model, named 3D-CANet (three-dimensional channel-wise attention network), was designed. A 3D-InnerCA (three-dimensional inner-channel attention) unit was proposed to dynamically capture the global spatio-temporal correlations for different channel-wise features. Meanwhile, an InterCA (inter-channel attention) unit was designed to adaptively recalibrate the contributions of different channel-wise features on each region. The experimental results on three real-world traffic raster flow datasets demonstrate that the predictive performance of the 3D-CANet model was better than the others,which proved the validity of the model proposed. |
| Keywords: spatio-temporal data traffic raster flow 3D channel-wise attention inner channel attention inter channel attention |
|
|
|
|
|
