On the Security of Pixel-Based Image Encryption for Privacy-Preserving Deep Neural Networks

TL;DR

This paper evaluates the robustness of pixel-based image encryption for privacy-preserving DNNs against ciphertext-only attacks and introduces a new attack method to reconstruct encrypted images.

Contribution

It provides an analysis of the security of pixel-based image encryption and proposes a novel ciphertext-only attack to assess its vulnerabilities.

Findings

Attack can recover images when encrypted with the same key.

Encryption is robust against ciphertext-only attacks with different keys.

Highlights security limitations of pixel-based encryption in certain scenarios.

Abstract

This paper aims to evaluate the safety of a pixel-based image encryption method, which has been proposed to apply images with no visual information to deep neural networks (DNN), in terms of robustness against ciphertext-only attacks (COA). In addition, we propose a novel DNN-based COA that aims to reconstruct the visual information of encrypted images. The effectiveness of the proposed attack is evaluated under two encryption key conditions: same encryption key, and different encryption keys. The results show that the proposed attack can recover the visual information of the encrypted images if images are encrypted under same encryption key. Otherwise, the pixel-based image encryption method has robustness against COA.