Presentation | 2017-11-17 A Study on Fall Detection by Networked Sensors Daiki Miura, Yuta Fukushima, Takashi Hamatani, Hirozumi Yamaguchi, Teruo Higashino, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Human fall detection has been on immense interest for elderly care. In this tequnical report, we propose a novel system for indoor fall detection with use of multiple infrared sensors. Our key idea is to virtually deploy convolutional neural networks on 2-D array of connected sensor nodes. The system can automatically tune its parameter set with a deep learning algorithm carried out within the sensor network where each IoT node transfers data to each other, and operates convolution and pooling. We confirmed our method could successfully detect human fall with 92.2% precision and 79.2% recall through our dataset collected by 55 gait instances. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Fall Detection / Distributed Computing / Deep Learning / Convolutional Neural Network |
Paper # | MoNA2017-19 |
Date of Issue | 2017-11-09 (MoNA) |
Conference Information | |
Committee | CNR / IN / MoNA |
---|---|
Conference Date | 2017/11/16(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Tetsuo Ono(Hokkaido Univ.) / Katsunori Yamaoka(Tokyo Inst. of Tech.) / Ryoichi Shinkuma(Kyoto Univ.) |
Vice Chair | Masayuki Kanbara(NAIST) / Kazunori Takashio(Keio Univ.) / Takuji Kishida(NTT) / Shigeaki Tagashira(Kansai Univ.) / Gen Kitagata(Tohoku Univ.) |
Secretary | Masayuki Kanbara(Hokkaido Univ.) / Kazunori Takashio(Panasonic) / Takuji Kishida(NTT) / Shigeaki Tagashira(NTT) / Gen Kitagata(KDDI Research) |
Assistant | Wataru Mito(SECOM) / Daisuke Yamamoto(Toshiba) / Yoichi Takashima(NTT) / / Koichi Nihei(NEC) / Takayuki Nishio(Kyoto Univ.) / Takato Saito(NTT) |
Paper Information | |
Registration To | Technical Committee on Cloud Network Robotics / Technical Committee on Information Networks / Technical Committee on Mobile Network and Applications |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | A Study on Fall Detection by Networked Sensors |
Sub Title (in English) | |
Keyword(1) | Fall Detection |
Keyword(2) | Distributed Computing |
Keyword(3) | Deep Learning |
Keyword(4) | Convolutional Neural Network |
1st Author's Name | Daiki Miura |
1st Author's Affiliation | Osaka University(Osaka Univ.) |
2nd Author's Name | Yuta Fukushima |
2nd Author's Affiliation | Osaka University(Osaka Univ.) |
3rd Author's Name | Takashi Hamatani |
3rd Author's Affiliation | Osaka University(Osaka Univ.) |
4th Author's Name | Hirozumi Yamaguchi |
4th Author's Affiliation | Osaka University(Osaka Univ.) |
5th Author's Name | Teruo Higashino |
5th Author's Affiliation | Osaka University(Osaka Univ.) |
Date | 2017-11-17 |
Paper # | MoNA2017-19 |
Volume (vol) | vol.117 |
Number (no) | MoNA-308 |
Page | pp.pp.25-30(MoNA), |
#Pages | 6 |
Date of Issue | 2017-11-09 (MoNA) |