Feature Extraction Based Deep Indexing by Deep Fuzzy Clustering for Image Retrieval Using Jaro Winkler Distance

Abstract

Abstract Image retrieval aims to locate similar image based on given query from large-sized image dataset. In the recent decades, different methods are developed to retrieve similar images, but increasing the retrieval performance is a challenging task. Hence, an effective method named Lion Henry Gas Solubility Optimization-based Deep Fuzzy Clustering (LHGSO-based DFC) is developed to increase the retrieval efficiency of image retrieval. Based on the features extraction, images are indexed efficiently to the corresponding cluster using clustering model. The proposed method retrieves the relevant images with respect to query image based on the similarity. The similarity measure is computed using Jaro Winkler distance, in which the images corresponding to the cluster that have higher similarity measure is retrieved more effectively. The proposed method achieved higher performance in terms of the metrics, like F-measure, precision and recall with the values of 0.887, 0.891 and 0.882.