| 谢贵才,段磊,蒋为鹏,肖珊,徐一凡.面向校园公共区域人流量预测的多尺度时序卷积网络设计.软件学报,2021,32(3):14-0 |
| 面向校园公共区域人流量预测的多尺度时序卷积网络设计 |
| Multi-scale Temporal Patterns Convolution Networks for Predicting Pedestrian Volume in Campus Public Area |
| 投稿时间：2020-07-21 修订日期：2020-11-06 |
| DOI：10.13328/j.cnki.jos.006183 |
| 中文关键词: 公共区域人流量预测 多尺度时序模式 卷积神经网络 多组件融合 |
| 英文关键词:pedestrian volume prediction in public area multi-scale temporal patterns convolution network multi-component fusion |
| 基金项目:国家自然科学基金项目（61972268） |
|
| 摘要点击次数: 133 |
| 全文下载次数: 108 |
| 中文摘要: |
| 校园公共区域人流量预测对于维护校园安全，提升校园管理水平有重大意义.尤其在疫情防控下，高校复学对公共区域的人流量预测和控制提出了更高的要求.以高校食堂为例，通过预测就餐人数，有助于食堂防疫人员合理调度和安排，既降低了人群聚集的潜藏风险，也可以针对食堂人流量分布情况提供分时分批服务.然而，由于校园管理需求，如节假日和教学安排等因素，使得校园公共区域人流量预测问题颇具挑战性.为此提出一种基于深度学习的多尺度时序卷积网络MSCNN （multi-scale temporal patterns convolution neural networks），实现人流量时序数据中短时依赖、长时周期模式的获取和多尺度时序模式特征的重标定，以对任一时段人流量进行预测.通过在真实校园环境数据集以及公开数据集上的实验，验证了MSCNN模型的有效性和执行效率. |
| 英文摘要: |
| Predicting pedestrian volume in campus public area is of significance for maintaining campus safety and improving campus management level. In particular, due to the outbreak of epidemic, the resumption of college education has put forward higher requirements for the prediction and control of the pedestrian volume in public area. Taking college canteens as an example, predicting the pedestrian volume in canteen is helpful with canteen epidemic prevention worker to make scheduling and arrangement, which not only reduces the risk of crowd gathering, but also provides more considerate service according to the distribution of the pedestrian volume in canteen. Considering the requirements of campus management, e. g., holiday, course arrangement, pedestrian volume prediction in the campus public area is challenging. In this paper, we propose a multi-scale temporal patterns convolution neural networks (MSCNN) based on deep learning to obtain the short-term dependencies as well as long-term periodicities, and reweight the multi-scale temporal pattern characteristics to predict the pedestrian volume at any given time. The effectiveness and efficiency of the MSCNN model are verified by experiments on real-world datasets. |
|
HTML 下载PDF全文 查看/发表评论 下载PDF阅读器 |
