Using Watermark-Based Protocol, Increase Robustness and Privacy of Content-Based Image Retrieval in Cloud Computing Environment
Sonal H. Kunte, Shivaji R. Lahane · 2017
Content-based image retrieval (CBIR) [1] are easy to manipulate and edit due to availability of powerful image processing and cloud computing environment. Searchable encryption (SE) scheme [11] allows image users to search over encrypted data collection. Nowadays, it is possible to do retrieval with preserving privacy [2] by encrypting it before outsourcing [5] and with copy-deterrence [3] so that unauthorized image users will not get access to it. Also tried to increase robustness of CBIR. Watermark certification authority (WCA) is a trusted agency who generate watermark-based protocol [4] for copy deterrence to avoid illegal distribution. Generating a unique watermark and directly Embedding it into the encrypted images by the cloud server before images are sent to the query user is one of the critical challenge WCA. The image integrity verification as well as generating trapdoor for retrieval in this environment is now days becoming the challenging for image users. Secure kNN algorithm [12] is used to protect feature vectors [2] (SCD, CSD, CLD, EHD) which are used to represent corresponding images. Recently hashing algorithm, auditing and alert generation are used at image users’ side for verification purpose.Firstly, Zhangs’ algorithm was used for encryption, decryption and generating watermark-based images. But now-a-days, watermark-based protocol is used, which improve robustness.