Towards efficient privacy-preserving encrypted image search in cloud computing

With the rapid development of cloud computing and big data, it is an ever-increasing trend to outsource the huge amount of multimedia data to the cloud. Although data outsourcing can alleviate data maintenance cost of user side, it inevitably brings some new security challenges, of which privacy of image content is the most significant one. Thus, privacy-preserving image search becomes a new research interest in academic community. In this paper, we propose a novel efficient search scheme for encrypted image by incorporating K-means and secure modular hashing, which can simultaneously achieve privacy-preserving of image search and comparable search accuracy to plaintext image. Furthermore, we provide comprehensively security analysis on the proposed solution and detailed efficiency evaluation over some real-world datasets. The results show that our solution can achieve desired security goals while providing a comparable search efficiency and accuracy.