Homomorphic Low Pass Filtering on Encrypted Multimedia over Cloud

Outsourcing the storage and processing of multimedia data to cloud data centers (CDCs) is becoming increasingly common. However, the use of third party CDCs raises security and privacy concerns. In this paper, we present a Shamir's Secret Sharing (SSS) based method to obtain enhanced digital images by performing low pass filtering (LPF)on the obfuscated (or encrypted) images over cloud. The proposed scheme is novel in performing averaging (division)operations involved in the spatial convolution of a mask overan image in the encrypted domain (ED) to get the same filtered resultant image as when the filtering operation is performed in the plain text domain (PD). Security and performance analyses corroborate that our scheme is information theoretically secure and transmission efficient (as compared to other schemes), respectively.