APSIPA Transactions on Signal and Information Processing > Vol 11 > Issue 1

An Overview of Compressible and Learnable Image Transformation with Secret Key and its Applications

Hitoshi Kiya, Tokyo Metropolitan University, Japan, kiya@tmu.ac.jp , April Pyone Maung Maung, Tokyo Metropolitan University, Japan, Yuma Kinoshita, Tokyo Metropolitan University, Japan, Shoko Imaizumi, Chiba University, Japan, Sayaka Shiota, Tokyo Metropolitan University, Japan
Suggested Citation
Hitoshi Kiya, April Pyone Maung Maung, Yuma Kinoshita, Shoko Imaizumi and Sayaka Shiota (2022), "An Overview of Compressible and Learnable Image Transformation with Secret Key and its Applications", APSIPA Transactions on Signal and Information Processing: Vol. 11: No. 1, e11. http://dx.doi.org/10.1561/116.00000048

Publication Date: 09 May 2022
© 2022 H. Kiya, A. P. M. Maung, Y. Kinoshita, S. Imaizumi and S. Shiota
Compressible encryptionlearnable encryptionencryption-then-compressionprivacy-preserving machine learningadversarial defensemodel protection


Open Access

This is published under the terms of CC BY-NC.

Downloaded: 2167 times

In this article:
Image Transformation with Key and its Applications 
Compressible Image Encryption for EtC systems 
Learnable Image Transformation for Traditional Machine Learning 
Learnable Image Transformation for DNN 
Image Transformation for Adversarially Robust Defense 
Model Protection with Image Transformation 


This article presents an overview of image transformation with a secret key and its applications. Image transformation with a secret key enables us not only to protect visual information on plain images but also to embed unique features controlled with a key into images. In addition, numerous encryption methods can generate encrypted images that are compressible and learnable for machine learning. Various applications of such transformation have been developed by using these properties. In this paper, we focus on a class of image transformation referred to as learnable image encryption, which is applicable to privacy-preserving machine learning and adversarially robust defense. Detailed descriptions of both transformation algorithms and performances are provided. Moreover, we discuss robustness against various attacks.