A Privacy-Preserving and End-to-End-Based Encrypted Image Retrieval Scheme

TL;DR

This paper introduces a novel end-to-end encrypted image retrieval scheme that combines encryption during JPEG compression with a neural network trained on cipher images, achieving good retrieval accuracy without feature leakage.

Contribution

It proposes a new encrypted image retrieval method that integrates encryption with end-to-end deep learning, ensuring privacy and high performance.

Findings

Effective retrieval performance demonstrated

Ensures privacy with no feature leakage

Maintains compression efficiency

Abstract

Applying encryption technology to image retrieval can ensure the security and privacy of personal images. The related researches in this field have focused on the organic combination of encryption algorithm and artificial feature extraction. Many existing encrypted image retrieval schemes cannot prevent feature leakage and file size increase or cannot achieve satisfied retrieval performance. In this paper, A new end-to-end encrypted image retrieval scheme is presented. First, images are encrypted by using block rotation, new orthogonal transforms and block permutation during the JPEG compression process. Second, we combine the triplet loss and the cross entropy loss to train a network model, which contains gMLP modules, by end-to-end learning for extracting cipher-images' features. Compared with manual features extraction such as extracting color histogram, the end-to-end mechanism can economize on manpower. Experimental results show that our scheme has good retrieval performance, while can ensure compression friendly and no feature leakage.