Distance based Similarity Search with Distinct Encrypted Image Storage on Cloud

Abstract

Secure picture recovery just on cloud has become increasingly popular in recent years as a result of the introduction of cloud computing. Both businesses and people frequently use the internet to manage and store their sensitive information, including such old photos and personal health records, due to its high level of convenience and financial savings. The typical method would be to encrypt the information prior to it being exported to using cloud to guarantee the secrecy of the information. The retrieval of information from encrypted data, for instance, may be impossible due to conventional encryption. The difficult issue in the cipher - text scenario is how to accomplish an effective retrieval while maintaining client anonymity. This study suggests a cutting-edge image retrieval method that achieves excellent efficiency and confidentiality over encrypted cloud data. An indexing tree is frequently used in the information extraction strategy to increase search performance. The security of private cloud data, including outsourced photos, the search tree, and query requests, is also a major concern. First, use an algorithm for extracting features from integrated picture features, which are made up of fundamental aspects like colour and shape. In particular, proposed technique can reach logarithmic search time thanks to the use of a balancing index tree. Second, the picture and query feature are encrypted using the secure inner product. Additionally, put in place a method to detect duplicate image material. Finally, extract the data owner details based on uploaded user-submitted image attributes that have been linked with distance measurements.