Capturing People Mobility with Mobile Sensing Technology for Disaster Evacuation
Capturing People Mobility with Mobile Sensing Technology for Disaster Evacuation
复制标题
DOI：
10.1007/978-3-030-21935-2_15
复制
发表时间：
2019-07
期刊：
影响因子：
--
通讯作者：
Chenwei Song;Masaki Ito;K. Sezaki
中科院分区：
文献类型：
--
作者：
Chenwei Song;Masaki Ito;K. Sezaki
在本文中，我们提出了一种基于客户端-服务器服务的系统，该系统提供人群检测和移动捕获。人群检测是检测并计算指定区域内人群的密度。移动捕捉就是追踪人的方向。如果增加报警机制，系统可以在检测到的人数超过一定比例时发出警报，从而预防或疏散人群，避免公共场所发生事故。该技术在灾后也发挥了很大的作用，通过计算灾区人口密度相对较小的路径，可以将高密度人群引导至安全区域，避免拥挤造成的二次伤害。与相机等常见方法相比，我们提出的系统具有成本低和位置灵活的优点。该系统可以检测任何区域，无需预先部署，只要有足够数量的用户参与即可。在本文中，我们在真实环境中进行了多次实验，以确定系统是否能够准确捕获人群信息和路线跟踪。
In this paper, we propose a client-server-service-based system that provides crowd detection and mobility capture. Crowd detection is to detect and calculate the density of crowds within a specified area. Mobility capture is to track the direction of the people. If a warning mechanism is added, the system can prevent or dissolve the crowd to avoid accidents in public places by sending alerts when the detected number of people exceeds a certain set percentage. The technology also plays a big role after the disaster, by calculating the path with relatively small population density in the disaster area, the high-density crowd can be led to a safe area avoiding secondary injury by crowding. Compared to common methods, such as cameras, our proposed system has the advantages of low cost and location flexibility. The system can detect any area without pre-deployed, as long as there is a sufficient number of users involved. In this article, we conducted several experiments in real environments to determine if the system can accurately capture crowd information and route tracking.