Presentation | 2019-03-17 Anomaly detection in hammering echoes using a domain-adapted DNN for unknown environment Fumito Ebuchi, Takanori Hasegawa, Masaya Iwata, Yuji Kasai, Masahiro Murakawa, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In this paper, we propose anomaly detection in hammering echoes using a domain-adapted deep neural network for an unknown environment. In this method, in order to acquire essential feature expression contributing to discrimination between normality and anomaly, while constructing a label classifier for hammering echoes acquired in the source domain, it trains a domain classifier so that it can not distinguish domains at the same time. This makes it possible to realize high precision discrimination without the training labels in the adaptation target domain. In order to verify the effectiveness, we evaluate the proposed method by using hammering echoes obtained from three different test concrete blocks. As a result, our proposed method improved the recognition rate by about 27% on average as compared to the conventional method. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Hammering inspection / Domain adaptation / Deep learning |
Paper # | BioX2018-43,PRMU2018-147 |
Date of Issue | 2019-03-10 (BioX, PRMU) |
Conference Information | |
Committee | PRMU / BioX |
---|---|
Conference Date | 2019/3/17(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Shinichi Sato(NII) / Kazuhiko Sumi(AGU) |
Vice Chair | Yoshihisa Ijiri(Omron) / Toru Tamaki(Hiroshima Univ.) / Hitoshi Imaoka(NEC) / Tetsushi Ohki(Shizuoka Univ.) |
Secretary | Yoshihisa Ijiri(NEC) / Toru Tamaki(Osaka Univ.) / Hitoshi Imaoka(Fujitsu Labs.) / Tetsushi Ohki(Univ. of Electro-Comm.) |
Assistant | Go Irie(NTT) / Yoshitaka Ushiku(Univ. of Tokyo) / Norihiro Okui(KDDI Research) / Daishi Watabe(Saitama Inst. of Tech.) |
Paper Information | |
Registration To | Technical Committee on Pattern Recognition and Media Understanding / Technical Committee on Biometrics |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Anomaly detection in hammering echoes using a domain-adapted DNN for unknown environment |
Sub Title (in English) | |
Keyword(1) | Hammering inspection |
Keyword(2) | Domain adaptation |
Keyword(3) | Deep learning |
1st Author's Name | Fumito Ebuchi |
1st Author's Affiliation | Tsukuba University/National Institute of Advanced Industrial Science and Technology(Tsukuba Univ./AIST) |
2nd Author's Name | Takanori Hasegawa |
2nd Author's Affiliation | Waseda University/National Institute of Advanced Industrial Science and Technology(Waseda Univ./AIST) |
3rd Author's Name | Masaya Iwata |
3rd Author's Affiliation | National Institute of Advanced Industrial Science and Technology(AIST) |
4th Author's Name | Yuji Kasai |
4th Author's Affiliation | National Institute of Advanced Industrial Science and Technology(AIST) |
5th Author's Name | Masahiro Murakawa |
5th Author's Affiliation | National Institute of Advanced Industrial Science and Technology/Tsukuba University(AIST/Tsukuba Univ.) |
Date | 2019-03-17 |
Paper # | BioX2018-43,PRMU2018-147 |
Volume (vol) | vol.118 |
Number (no) | BioX-512,PRMU-513 |
Page | pp.pp.85-89(BioX), pp.85-89(PRMU), |
#Pages | 5 |
Date of Issue | 2019-03-10 (BioX, PRMU) |