Paper 2024/460

Encrypted Image Classification with Low Memory Footprint using Fully Homomorphic Encryption

Lorenzo Rovida, University of Milano-Bicocca
Alberto Leporati, University of Milano-Bicocca
Abstract

Classifying images has become a straightforward and accessible task, thanks to the advent of Deep Neural Networks. Nevertheless, not much attention is given to the privacy concerns associated with sensitive data contained in images. In this study, we propose a solution to this issue by exploring an intersection between Machine Learning and cryptography. In particular, Fully Homomorphic Encryption (FHE) emerges as a promising solution, as it enables computations to be performed on encrypted data. We, therefore, propose a Residual Network implementation based on FHE which allows the classification of encrypted images, ensuring that only the user can see the result. We suggest a circuit which reduces the memory requirements by more than 85% compared to the most recent works, while maintaining a high level of accuracy and a short computational time. We implement the circuit using the well-known CKKS scheme, which enables approximate encrypted computations. We evaluate the results from three perspectives: memory requirements, computational time and calculations precision. We demonstrate that it is possible to evaluate an encrypted ResNet20 in less than five minutes on a laptop using approximately 15GB of memory, achieving an accuracy of 91.67% on the CIFAR-10 dataset, which is almost equivalent to the accuracy of the plain model (92.60%).

Metadata
Available format(s)
PDF
Category
Applications
Publication info
Published elsewhere. Minor revision. International Journal of Neural Systems
DOI
10.1142/S0129065724500254
Keywords
Homomorphic EncryptionPrivacy-Preserving Machine LearningEncrypted Neural Networks
Contact author(s)
lorenzo rovida @ unimib it
alberto leporati @ unimib it
History
2024-03-22: approved
2024-03-18: received
See all versions
Short URL
https://ia.cr/2024/460
License
Creative Commons Attribution
CC BY

BibTeX

@misc{cryptoeprint:2024/460,
      author = {Lorenzo Rovida and Alberto Leporati},
      title = {Encrypted Image Classification with Low Memory Footprint using Fully Homomorphic Encryption},
      howpublished = {Cryptology {ePrint} Archive, Paper 2024/460},
      year = {2024},
      doi = {10.1142/S0129065724500254},
      url = {https://eprint.iacr.org/2024/460}
}
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.