Hidden neighbor relations to tackle the uncertainness of sensors for an automatic human tracking
Tappei Yotsumoto, Masaru Shiozuka, Keníchi Takahashi, Takao Kawamura, Kazunori Sugahara · 2017 Second International Conference on Electrical, Computer and Communication Technologies (ICECCT) · 2017
A human tracking system based on mobile agent technologies has been proposed for achieving automatic human tracking function. In this system, a target person is tracked automatically by one mobile agent moving among sensors in which a person is detected. The current system utilizes an algorithm to predict which sensor detects the target person. However, in order to keep tracking without losing sight of the person, it should be necessary to deal with the uncertainness of sensors. In this paper, we propose a method which tackles the uncertainness of sensors by finding hidden neighbor relations based on the detection result of sensors.