作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2022, Vol. 48 ›› Issue (2): 25-33,39. doi: 10.19678/j.issn.1000-3428.0061419

• 热点与综述 • 上一篇    下一篇

一种基于区块链可验证的加密图像检索方案

彭红艳1,2, 李杰1, 石贞奎1,2, 李先贤1,2   

  1. 1. 广西师范大学 计算机科学与信息工程学院, 广西 桂林 541004;
    2. 广西多源信息挖掘与安全重点实验室, 广西 桂林 541004
  • 收稿日期:2021-04-25 修回日期:2021-08-12 发布日期:2022-02-14
  • 作者简介:彭红艳(1973-),女,副教授,主研方向为信息安全、区块链技术;李杰,硕士研究生;石贞奎,讲师、博士;李先贤,教授、博士。
  • 基金资助:
    国家自然科学基金地区科学基金项目(61862008,61662008);广西自然科学基金面上项目(2020JJA170023)。

A Blockchain-based Verifiable Encrypted Image Retrieval Scheme

PENG Hongyan1,2, LI Jie1, SHI Zhenkui1,2, LI Xianxian1,2   

  1. 1. College of Computer Science and Information Engineering, Guangxi Normal University, Guilin, Guangxi 541004, China;
    2. Guangxi Key Lab of Multi-Source Information Mining and Security, Guilin, Guangxi 541004, China
  • Received:2021-04-25 Revised:2021-08-12 Published:2022-02-14

摘要: 由于难以构造通用的认证结构对图像类型数据的相似度计算过程进行验证,因此对加密图像检索结果的验证面临很大挑战。同时,现有多数加密图像检索方案没有考虑恶意云服务器的问题,可能返回不正确或不完整的检索结果。利用区块链技术的去中心化、不可篡改等特性,提出一种基于区块链可验证的加密图像检索方案BVEIR,确保搜索结果的可靠性与搜索过程的透明性。将加密索引存储在区块链(以太坊)上,通过区块链的共识机制保证在智能合约上完成搜索的功能,确保搜索结果完备性,同时将相应的加密图像数据外包到云服务器以降低存储成本,并在相似图片索引过程中使用基于视觉词袋模型和simhash的双层索引结构,进一步提高检索效率和精度。实验结果表明,BVEIR具有良好的隐私保护效果,其建立索引时间较SEIR方案更少,并且可实现对图像数据的细粒度访问控制,在提高索引效率的同时保证图像检索的精确率。

关键词: 加密图像检索, 区块链, 可搜索加密, 局部敏感哈希, 属性加密

Abstract: It is difficult to construct a general authentication structure to verify the similarity calculation process of image type data, so the verification of encrypted image retrieval results faces great challenges.At the same time, most existing encrypted image retrieval schemes do not consider the problem of malicious cloud servers, and may return incorrect or incomplete retrieval results.Taking advantage of the decentralized and tamper proof characteristics of blockchain technology, a Blockchain-based Verifiable Encrypted Image Retrieval(BVEIR) scheme is proposed to ensure the reliability of search results and the transparency of search process.With the encrypted index stored in blockchain(Ethereum), the search function is implemented by using the consensus mechanism of blockchain on the smart contract to ensure the completeness of search results.At the same time, the corresponding encrypted image data is outsourced to the cloud server to reduce storage cost.In the process of similarity image indexing, the double-layer index structure adopting visual word bag model and simhash is used to improve the retrieval efficiency and accuracy.Experimental results show that BVEIR displays good privacy protection effect.Its indexing time is less than that of the SEIR scheme, and can realize fine-grained access control of image data, so it can improve the indexing efficiency and ensure the accuracy of image retrieval.

Key words: encrypted image retrieval, blockchain, searchable encryption, locality-sensitive hashing, attribute-based encryption

中图分类号: