Privacy-Preserving Content-Based Image Retrieval in the Cloud: A Review

As digital imaging devices have increased in popularity and technology advanced, a large amount of image databases have been created. As a result of these image databases, it becomes increasingly necessary to develop efficient image retrieval search methods that meet user requirements. Numerous attempts have been made to improve content-based image retrieval methods. The semantic gap between the features of an image and how the human perceives it has been the focus of considerable attention. However, moving these databases to the cloud raises privacy concerns as the cloud service provider cannot be trusted. Hence researchers have proposed privacy-preserving content-based image retrieval schemes. Since research in this area has grown rapidly over the last two years, this paper analyses and compares twelve state-of-the-art articles from this field. This paper also provides an overview of Privacy-preserving content-based image retrieval (PPCBIR) system architecture, Datasets, encryption methods, feature extraction methods, similarity measurement, and performance measurement to inspire further research efforts.