Black hole Entropic Fuzzy Clustering-based image indexing and Tversky index-feature matching for image retrieval in cloud computing environment

Abstract

Due to the expansion of social websites, the data owner accumulates multimedia data and stores in cloud server. The owner encrypts the images before uploading it in cloud server for security. However, the conventional encryption method failed to support feasible retrieval on the encrypted images. This paper proposes novel technique, namely Black Hole Entropic Fuzzy Clustering-based Tversky index for effective image retrieval. Here, the SLBP (Spatial Local Binary Pattern) features, semantic features, statistical features, and low image features are considered in the feature extraction. In addition, encryption of feature vector using Elliptic Curve Cryptography (ECC) is done for encrypting the images contained in cloud server. The Black Hole Entropic Fuzzy Clustering (BHEFC) is adapted for grouping the images. Whenever users request an input query then, the query image is fed to feature extraction, and encryption phase, wherein the Tversky index is applied for matching the similarity between the images for retrieval. The proposed BHEFC-based Tversky index outperformed other methods with maximal accuracy of 95.737%, maximal precision of 83.563%, maximal recall of 94.697%, and maximal F-measure of 83.014%.