EPCBIR: An efficient and privacy-preserving content-based image retrieval scheme in cloud computing

详细信息    查看全文

• 作者：Zhihua Xia^a ; ^{xia_zhihua@163.com} ; Neal N. Xiong ; ^b ; ^{xiongnaixue@gmail.com} ; ^{neal.xiong@swosu.edu} ; Athanasios V. Vasilakos^c ; ^{vasilako@ath.forthnet.gr} ; Xingming Sun^a ; ^{sunnudt@163.com}
• 关键词：Searchable encryption ; Content-based image retrieval ; Secure k-nearest neighbors algorithm ; Locality-sensitive hashing
• 刊名：Information Sciences
• 出版年：2017
• 出版时间：May 2017
• 年：2017
• 卷：387
• 期：Complete
• 页码：195-204
• 全文大小：1187 K
• 卷排序：387

文摘

The content-based image retrieval (CBIR) has been widely studied along with the increasing importance of images in our daily life. Compared with the text documents, images consume much more storage and thus are very suitable to be stored on the cloud servers. The outsourcing of CBIR to the cloud servers can be a very typical service in cloud computing. For the privacy-preserving purposes, sensitive images, such as medical and personal images, need to be encrypted before being outsourced, which will cause the CBIR technologies in plaintext domain unusable. In this paper, we propose a scheme that supports CBIR over the encrypted images without revealing the sensitive information to the cloud server. Firstly, the feature vectors are extracted to represent the corresponding images. Then, the pre-filter tables are constructed with the locality-sensitive hashing to increase the search efficiency. Next, the feature vectors are protected by the secure k-nearest neighbor (kNN) algorithm. The security analysis and experiments show the security and efficiency of the proposed scheme.