Personal and Ubiquitous Computing
| Hi-index | 0.00 |
A method of detecting mastication by using a microwave Doppler sensor is proposed in this article. For solitary elderly people, it is important to check whether they have taken meals regularly for health care reasons. In this article, a method for detecting whether meals have been taken regularly that implements privacy protection and offers hands-free operation is proposed. We have adopted a couple of microwave Doppler sensor. Microwave Doppler sensor detects a moving object from the difference between the sent frequency and the received frequency. The up-and-down motion of the jaw is detected by this sensor. A person's mastication is known to have constant periodicity. [1] Based on the above assumption, mastication is detected from the periodicity of the up-and-down motion of the jaw in this article. A person often moves the body and face in addition to the jaw during the act of mastication. Therefore, it is necessary to extract the signal caused by the movement of the jaw from the Doppler signal obtained by the sensors. Concretely, only the signal caused by the up-and-down motion of the jaw is extracted from the Doppler signal by using the Wavelet transformation, and the periodicity is detected by using the auto-correlation function. And mastication is distinguished from other movement. In the results of an experiment using two micro wave Doppler sensors of 10.5GHz, the accuracy was from 70 to 90%.