Fusion of Text and Image in Multimedia Information Retrieval System

Abstract

Multimedia Information Retrieval is very useful for any application in our daily work. Most of the applications consist of Multimedia data that are images, text, audio and video. Multimedia information retrieval system is used to search an image. There are same meanings for different data which is also known as semantic gap. This problem is solved by fusion of text based image retrieval and content based image retrieval. Weighted Mean, OWA and WOWA are aggregation operators used in this system for the fusion of text and image numeric values. The Scale invariant feature transforms and speeded up robust feature are two algorithms for feature extraction. To increase the speed of system, the speeded up robust feature algorithm is used. Bag of Words and Bag of Visual Word approaches are used in this system for retrieving images.
DOI: 10.17762/ijritcc2321-8169.150662