Presentation 2019-03-06
Complementary Color Reconstruction by Autoencoders
Akihiro Suzuki, Hakaru Tamukoh,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This study proposes a novel training method for autoencoders (AEs) that gives the AEs complementary color images as target data. Using the proposed method, AEs are trained to reconstruct complementary color images from input images. The trained AEs are applied to anomaly detection. In this case, correct data is correctly reconstructed, and anomalous data is complementary color reconstructed, by the trained AEs. Therefore, the AEs are available for anomaly detection using reconstruction error to distinguish between correct data and anomalous data. This paper employs HSV image instead of RGB image and reports anomaly detection with CIFAR-10 dataset using an AE trained by the proposed method. In addition, features which the AE obtain via proposed method were confirmed by visualizing value of a hidden layer of the AE.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Deep Learning / Autoencoders / Anomaly Detection
Paper # SIS2018-41
Date of Issue 2019-02-27 (SIS)

Conference Information
Committee SIS
Conference Date 2019/3/6(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Tokyo Univ. Science, Katsushika Campus
Topics (in Japanese) (See Japanese page)
Topics (in English) Soft Computing, etc.
Chair Takayuki Nakachi(NTT)
Vice Chair Noriaki Suetake(Yamaguchi Univ.) / Tomoaki Kimura(Kanagawa Inst. of Tech.)
Secretary Noriaki Suetake(Kyushu Inst. of Tech.) / Tomoaki Kimura(Tokyo Metropolitan Univ.)
Assistant Takanori Koga(National Inst. of Tech. Tokuyama College) / Hideaki Misawa(National Inst. of Tech., Ube College)

Paper Information
Registration To Technical Committee on Smart Info-Media Systems
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Complementary Color Reconstruction by Autoencoders
Sub Title (in English)
Keyword(1) Deep Learning
Keyword(2) Autoencoders
Keyword(3) Anomaly Detection
1st Author's Name Akihiro Suzuki
1st Author's Affiliation Kyushu Institute of Technology(Kyutech)
2nd Author's Name Hakaru Tamukoh
2nd Author's Affiliation Kyushu Institute of Technology(Kyutech)
Date 2019-03-06
Paper # SIS2018-41
Volume (vol) vol.118
Number (no) SIS-473
Page pp.pp.23-28(SIS),
#Pages 6
Date of Issue 2019-02-27 (SIS)