Secure video retrieval using image query on an untrusted cloud

Abstract

Video retrieval has been widely used in many applications such as video surveillance, object detection. Recently, the security of video retrieval systems has attracted much attention of researchers. However, the most of the previous works focus on secure image search, which usually utilize homomorphic encryption (HE) or order-preserving encryption (OPE) to achieve privacy-preserving. Specially, few works pay attention to secure video retrieval. In this paper, we propose a novel secure video retrieval (SVR, for simplicity) scheme by utilizing comparable encryption with a vector extending mechanism. The distance between images could be presented as range query. Security analysis demonstrate that the proposed scheme can preserve the privacy of data. The experimental results show that the ciphertext search results are consistent with the plaintext form.