Visual analysis method for abnormal passenger flow on urban metro network
- PDF / 3,197,257 Bytes
- 18 Pages / 595.276 x 790.866 pts Page_size
- 96 Downloads / 205 Views
R E G UL A R P A P E R
Yong Zhang • He Shi • Feifei Zhou
•
Yongli Hu • Baocai Yin
Visual analysis method for abnormal passenger flow on urban metro network
Received: 15 August 2019 / Revised: 2 April 2020 / Accepted: 5 April 2020 The Visualization Society of Japan 2020
Keywords Metro data visualization Anomaly verification Visualization analytics
1 Introduction Public transit, especially the metro system, has become one of the main facilities to carry mega passenger flows in big cities like Beijing and New York. However, with the increase in passenger flow, some unexpected situations pose negative effects on the metro system, such as the gatherings of passengers and bad weather. If anomalies cannot be effectively observed and controlled in time, they may spread rapidly with the expansion of negative influences. Therefore, the accurate detection of abnormal events in public transit, as well as the effective reasoning of potential causes can help relevant departments propose effective emergency measures to prevent abnormal events from occurring again. Several approaches have been developed to overcome the aforementioned issues. However, they mainly focus on detecting abnormal stations without exploring the reasons behind the abnormal passenger flow in subways. In addition, it is difficult to verify their results as well. Only through the complex analysis of the original traffic data may users be able to understand whether the test results are reasonable. These kinds of methods are not suitable for users without the background of computer science, and therefore they cannot satisfy the requirement of handling huge and complex traffic data efficiently. At the same time, the rapid development of social network encourages people to express their views and ideas on the social media platform such as Weibo. The massive social network data may contain clues of abnormal situations in metro systems. Manual queries of relevant social network data are labor-consuming and may lead to heavy workload with low efficiency. Thus, it is necessary to develop an intuitive and interactive visual analytic system to detect abnormal events and reason their causes with an auto-integration of semantic data and smart card data. In this paper, we try to display the original traffic data and verify the anomalies through 3D visualizationbased method. At the same time, we also provide a set of visualization views based on Weibo data to explore the causes of anomalies. The interactive visual analytic system interface we have designed is shown in Fig. 1. The main contributions of the paper are as follows:
Y. Zhang H. Shi F. Zhou (&) Y. Hu B. Yin Beijing Key Laboratory of Multimedia and Intelligent Software Technology, Faculty of Information Technology, Beijing Artificial Intelligence Institute, Beijing University of Technology, Beijing 100124, China E-mail: [email protected] Y. Zhang E-mail: [email protected]
Y. Zhang et al.
• A 3D hierarchical view is proposed to analyze spatio-temporal data of metro systems and each
Data Loading...
