AAJS: An Anti-Malicious Attack Graphic Similarity Judgment System in Cloud Computing Environments

• Main Authors: Chen, X. (Author), Liu, X. (Author), Luo, D. (Author), Xiong, N. (Author), Xu, G. (Author)
• Format: Article
• Language: English
• Published: MDPI 2023
• Subjects: graphic similarity matching; graphics search in cloud computing; malicious attack; real/ideal model paradigm; secure multi-party computation
• Online Access: View Fulltext in Publisher; View in Scopus

 With the rapid development of cloud computing and other modern technologies, collaborative computing between data is increasing, and privacy protection and secure multi-party computation are also attracting more attention. The emergence of cloud computing provides new options for data holders to perform complex computing problems and to store images; however, data privacy issues cannot be ignored. If a graphic is encrypted and stored in the cloud, the cloud server will perform confidential similar matching when the user searches. At present, most research on searchable encryption is focused on text search, with few schemes researched on how to finish the graphic search. To solve this problem, this paper proposes a secure search protocol based on graph shape under the semi-honest model. Using the cut-choose method and zero-knowledge proof, further designs of the anti-malicious attack graphic similarity judgment system (AAJS) based on the Paillier encryption algorithm, can achieve the secure search and matching of the graph while resisting malicious adversary attacks. The proposed protocol’s security is proved by the real/ideal model paradigm. This paper conducts performance analysis and experimental simulation on the existing scheme and the experiments demonstrate that the system achieves high execution efficiency. © 2023 by the authors.