On the Security of Pixel-Based Image Encryption for Privacy-Preserving Deep Neural Networks
This addresses security concerns for privacy-preserving deep learning applications, but it is incremental as it focuses on evaluating and attacking a specific existing method.
The paper tackles the security of pixel-based image encryption for privacy-preserving deep neural networks by evaluating its robustness against ciphertext-only attacks and proposing a novel DNN-based attack. The results show that the attack can recover visual information when images use the same encryption key, but the encryption method remains robust otherwise.
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.