WSEAS Transactions on Information Science and Applications
Print ISSN: 1790-0832, E-ISSN: 2224-3402
Volume 14, 2017
Improving Physical Security with Machine Learning and Sensor-Based Human Activity Recognition
Authors: ,
Abstract: With current maturity and wide accessibility of low-cost sensor technologies, sensor-based human activity recognition is becoming more and more popular in various domains and novel innovative applications. Huge amount of research in this area is driven by smart-home assistive living applications, many of them mostly focused on the development of efficient methods and applications that can help in supporting independent living and provide assistance with everyday instrumental activities of daily living. On the other hand, in today’s world filled with uncertainty and ever increasing security risks, personal physical security is becoming more important than ever. In this paper we report on the identified need and present the current status and future steps towards developing a robust physical intrusion detection method aimed at improving people’s personal physical security. Proposed method relies on machine learning techniques and on sensor-based human activity recognition and will be validated on the application prototype for robust physical intrusion detection on home doors in real-life environment.
Search Articles
Keywords: activity recognition, machine learning, context-aware, real-time, dense-sensing, accelerometer, physical intrusion, physical security
Pages: 1-9
WSEAS Transactions on Information Science and Applications, ISSN / E-ISSN: 1790-0832 / 2224-3402, Volume 14, 2017, Art. #1