Abstract:With the development of cloud computing, more and more resource constrained data owners tend to store their images in the cloud and rely on image retrieval services to get back images they want. However, this process raises huge privacy leakage risks. Privacy-preserving image retrieval (PPIR) can prevent privacy leakage during the retrieval process. Using convolutional neural networks for image feature extraction can significantly improve retrieval accuracy and has become a common trend in PPIR, but it also brings a new problem of network model parameter leakage, which is often ignored by most PPIR schemes. Besides, existing PPIR schemes often adopt centralized image retrieval service, which typically require frequent interactions between the cloud server and users during the retrieval process, resulting in excessive computational overhead for resource-limited users. Aiming to solve these problems, a distributed privacy-preserving image retrieval scheme is proposed in this paper. Three collaborative cloud servers utilize three neural network models split and outsourced by the data owner to extract features of the query image and perform subsequent retrieval tasks. A new secure Euclidean distance calculation protocol is designed for cloud servers to measure feature similarity, and secure sort pair protocol, secure sorting protocol, secure search protocol are utilized to achieve secure image retrieval. Additionally, a hierarchical index tree is employed to narrow down the search scope and improve search efficiency. This approach avoids the problem of network model parameter leakage caused by using neural networks to extract image features at the users' side, reduces the computational burden on users, and improves the accuracy and the efficiency of the retrieval process while ensuring retrieval security. Security analysis and experimental results demonstrate the security and retrieval performance of the proposed scheme.

Economic Costs of Multi-sever Private Information Retrieval in Cloud Computing.

Practical private information retrieval supporting keyword search in the cloud

An Efficient Scheme for Private Information Retrieval

VPIR: an efficient verifiable private information retrieval scheme resisting malicious cloud server

Lower-Cost epsilon-Private Information Retrieval

Efficient Ranked Multi-Keyword Retrieval with Privacy Protection for Multiple Data Owners in Cloud Computing

Towards Privacy-Preserving Content-Based Image Retrieval in Cloud Computing

Multi-Server Weakly-Private Information Retrieval

Inter-Cloud Data Integration System Considering Privacy And Cost

Dynamic Content-Based Cloud Data Integration System with Privacy and Cost Concern

An Upper-Bound Control Approach for Cost-Effective Privacy Protection of Intermediate Dataset Storage in Cloud

Private Information Retrieval with Private Noisy Side Information

Cloud-based Privacy-Preserving Collaborative Consumption for Sharing Economy

Privacy-preserving image retrieval based on additive secret sharing in cloud environment

Two-Server Private Information Retrieval with High Throughput and Logarithmic Bandwidth.

Efficient Privacy-Preservation Multi-factor Ranking with Approximate Search over Encrypted Big Cloud Data.

The Capacity of Private Information Retrieval from Decentralized Uncoded Caching Databases

Protecting Access Privacy in Ethereum Using Differentially Private Information Retrieval

Content-based multi-source encrypted image retrieval in clouds with privacy preservation

A Flexible and Verifiable Keyword PIR Scheme for Cloud-Edge-Terminal Collaboration in AIoT

Research on privacy-preserving keyword retrieval technology in cloud storage