|
An Individual Recognition Method based on Door Opening and Closing Motion using a Gyroscope
|
|
Akari Fukao1), Kyosuke Futami2), Kazuya Murao3)
|
1) College of Info. Sci. and Eng., Ritsumeikan University. 1-1-1 Nojihigashi, Kusatsu, Shiga 525-8577, Japan
2) College of Info. Sci. and Eng., Ritsumeikan University. 1-1-1 Nojihigashi, Kusatsu, Shiga 525-8577, Japan
3) College of Info. Sci. and Eng., Ritsumeikan University. 1-1-1 Nojihigashi, Kusatsu, Shiga 525-8577, Japan
|
|
Received: Oct. 1 2020; Revised: Dec. 12 2020; Accepted: Jan. 12 2021
|
|
Keywords: ubiquitous computing, angular velocity sensor, user recognition, door motion, doorknob motion
|
|
|
|
| Abstract | |
| There are many environments in our daily lives where multiple people enter and exit a room. Although many methods for individual recognition have been proposed so far, there are some problems for the recognition of individuals who enter and exit a room because of the psychological and physical burdens such as unnecessary action for users and the need to carry a device. In this paper, we propose a method to recognize a person entering or exiting a room by using the door opening/closing motion. The proposed method is based on the hypothesis that each user has their own characteristics in the opening and closing of the door. We implement the prototype system on a common lever-handle door, and conduct experiments on four subjects, recognizing their entry and exit motions. The F-measure of the entry and exit motions were recognized by 0.90 and 0.73, respectively.
|
|
|
|
| Copyright © 2019 The Academy of Human Informatics |
|
