Presentation 2022-03-08
[Poster Presentation] Influence of Data Hiding in Etc-Image Classification with SVM
Kosuke Abe, Shoko Imaizumi, Hitoshi Kiya,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Classification accuracy for Encryption-then-Comperssion (EtC) images by support vector machine(SVM) is comparable to that for original images. In this paper, we evaluate the effect of a reversible embedding process for EtC images on the SVM-based image classification accuracy. In the embedding process, a client embeds a classification label into each EtC image. We assume that the classification label does not consist of words that directly represent the image content but a symbol that the client preliminarily assigns to the words. In this way, the client can send EtC images with classification labels to a third party such as a service provider, and allow the third party to learn the EtC images without disclosing the image content. Simulation results show that the classification accuracy of the marked EtC images is the same as the accuracies of the original and EtC images. We proved that not only the EtC process but also the embedding process does not affect the image classification accuracy.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) SVM / Encryption-then-Compression / Reversible data hiding
Paper # EMM2021-112
Date of Issue 2022-02-28 (EMM)

Conference Information
Committee EMM
Conference Date 2022/3/7(2days)
Place (in Japanese) (See Japanese page)
Place (in English) (Primary: Online, Secondary: On-site)
Topics (in Japanese) (See Japanese page)
Topics (in English) Image and Sound Quality, Metrics for Perception and Recognition, Human Auditory and Visual System, etc.
Chair Ryoichi Nishimura(NICT)
Vice Chair Masaaki Fujiyoshi(Tokyo Metropolitan Univ.) / Masatsugu Ichino(Univ. of Electro-Comm.)
Secretary Masaaki Fujiyoshi(Utsunomiya Univ.) / Masatsugu Ichino(NICT)
Assistant Shoko Imaizumi(Chiba Univ.) / Youichi Takashima(Kaishi Professional Univ.)

Paper Information
Registration To Technical Committee on Enriched MultiMedia
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) [Poster Presentation] Influence of Data Hiding in Etc-Image Classification with SVM
Sub Title (in English)
Keyword(1) SVM
Keyword(2) Encryption-then-Compression
Keyword(3) Reversible data hiding
1st Author's Name Kosuke Abe
1st Author's Affiliation Chiba University(Chiba Univ.)
2nd Author's Name Shoko Imaizumi
2nd Author's Affiliation Chiba University(Chiba Univ.)
3rd Author's Name Hitoshi Kiya
3rd Author's Affiliation Tokyo Metropolitan University(Tokyo Metropolitan Univ.)
Date 2022-03-08
Paper # EMM2021-112
Volume (vol) vol.121
Number (no) EMM-417
Page pp.pp.107-111(EMM),
#Pages 5
Date of Issue 2022-02-28 (EMM)