Loading...
Please wait, while we are loading the content...
Similar Documents
Intruder detection in public space using suspicious behavior phenomena and wireless sensor networks
| Content Provider | ACM Digital Library |
|---|---|
| Author | Suthaharan, Shan Bandari, Snigdha |
| Abstract | Automatic detection of intruders in a public space is an important requirement for the safety of innocent citizens. Recently suspicious behavior based automatic detection models are proposed to address this problem. Wireless Sensor Networks (WSN) is adopted in the implementation and the evaluation of these models. In this paper we propose a WSN-based trajectory enhanced graphical model to detect the intruders in public space. In our model a supervised learning approach is used to estimate the threshold for speed parameter of the suspicious and unsuspicious tracks. The threat levels are increased when a person's trajectory deviates from an unsuspicious track, where it is closer to a suspicious track. Suspicious behavior of a person is detected using a threat level parameter that depends on the speed and direction of the subject. We show using the synthetic data that, the model is capable of detecting human intruders based on the suspicious behavior, and it is worth applying the model to real data sets and investigate its performance further. |
| Starting Page | 3 |
| Ending Page | 8 |
| Page Count | 6 |
| File Format | |
| ISBN | 9781450312882 |
| DOI | 10.1145/2248356.2248359 |
| Language | English |
| Publisher | Association for Computing Machinery (ACM) |
| Publisher Date | 2012-06-11 |
| Publisher Place | New York |
| Access Restriction | Subscribed |
| Subject Keyword | Intruder detection Wireless sensor networks Suspicious behavior Supervised learning |
| Content Type | Text |
| Resource Type | Article |
