| Presentation | 2014-01-23 Introducing Machine-Learning to Individual Recognition Using Doppler Sensors Shunsuke MASHIMA, Hiroaki MORINO, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In recent years, individual recognition techniques have been extensively studied focusing on the natural behavior of humans, in which image recognition and voice recognition are the mainstream for techniques. However, these techniques have limitation of applicable situations and cases. For example, image recognition cannot be applied to the dark environments and it could infringe privacy of the users. Voice recognition could put psychological burden on the users. To solve these problems, we focus on the use of Microwave Doppler sensors. Microwave Doppler sensors can be used even in the dark situations. Also, it does not incur privacy problems nor let the users feel any mental burden as for being sensed since they do not need to acquire images by lenses. In the previous literature, we have proposed an individual recognition scheme that utilizes a Doppler sensor to detect the natural human motions and have shown through experiments that it was able to recognize three subjects by using the frequency spectrum data of the signal received by the Doppler sensor. This paper presents a novel learning method based on back-propagation neural networks that can be used in the recognizer. The paper shows that the proposed method enables recognition of up to five users by performing the parameter adjustment and changing the number of the input nodes in the back propagation neural networks. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Individual recognition / Doppler sensor / FFT / the feature quantity / neural networks |
| Paper # | ASN2013-127 |
| Date of Issue |
| Conference Information | |
| Committee | ASN |
|---|---|
| Conference Date | 2014/1/16(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Ambient intelligence and Sensor Networks(ASN) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Introducing Machine-Learning to Individual Recognition Using Doppler Sensors |
| Sub Title (in English) | |
| Keyword(1) | Individual recognition |
| Keyword(2) | Doppler sensor |
| Keyword(3) | FFT |
| Keyword(4) | the feature quantity |
| Keyword(5) | neural networks |
| 1st Author's Name | Shunsuke MASHIMA |
| 1st Author's Affiliation | Graduate School of Engineering and Science, Shibaura Institute of Technology() |
| 2nd Author's Name | Hiroaki MORINO |
| 2nd Author's Affiliation | Graduate School of Engineering and Science, Shibaura Institute of Technology |
| Date | 2014-01-23 |
| Paper # | ASN2013-127 |
| Volume (vol) | vol.113 |
| Number (no) | 399 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |