Incident detection based on mobile crowd sensing for smart city

作     者：Zhang, Peng Zhang, Zhenjiang Chao, Han-Chieh 

作者机构：Department of Electronic and Information Engineering Key Laboratory of Communication and Information Systems Beijing Municipal Commission of Education Beijing Jiaotong University Beijing100044 China Department of Software Engineering Key Laboratory of Communication and Information Systems Beijing Municipal Commission of Education Beijing Jiaotong University Beijing100044 China Department of Electrical Engineering National Dong Hwa University Hualien97401 Taiwan School of Information Science and Engineering Fujian University of Technology Fuzhou350118 China School of Mathematics and Computer Science Wuhan Polytechnic University Wuhan430023 China Department of Computer Science and Information Engineering National Ilan University Yilan County260 Taiwan 

出 版 物：《Journal of Computers (Taiwan)》 (J. Comput.)

年 卷 期：2019年第30卷第1期

页      面：96-104页

核心收录：

基　　金：This work has been supported by National Natural Science Foundation under Grant 61772064  the Academic Discipline and Postgraduate Education Project of Beijing Municipal Commission of Education and the Fundamental Research Funds for the Central Universities 2017YJS006 

主　　题：Environmental impact 

摘      要：In recent years, mobile devices have taken a significant role in improving the quality of people’s life. In order to enhance the usability of those devices, more and more sensors have been built in. Furthermore, the data storage and data processing capacity are becoming larger and larger. To capture more information about the city conditions and make full use of the computing resource, mobile crowd sensing has been proposed. Incident detection is an important component of the safety monitoring system, however, existing incident detection methods are mostly based on computer vision which is vulnerability to environmental impacts. This paper proposes an automatic incident detection model based on Mobile Crowd Sensing to deal with the problem above. The incident detection algorithm in the model is based on the human activity recognition and the crowd density. The simulation results show that this model is feasible and effective. 2019 © Computer Society of the Republic of China. All Rights Reserved.