Feature based adaptive tolerance tree fatt an efficient indexing technique for content based image retrieval using 2d wavelet transform

Abstract

The area of Content Based Image Retrieval CBIR has been an newlineactive area of research for the past two decades This thesis proposes new newlinestorage and access techniques for effective representation of images in image newlinedatabases and retrieval of relevant images based on tree structures newlineEarlier attempts in the area of text based indexing and information newlineretrieval systems have had a number of limitations In multimedia and newlineimaging databases the retrieval complexity increases with increase in the newlinenumber of images stored in the database The main goal of CBIR is to newlineincrease the efficiency in image indexing and retrieval thereby reducing the newlinehuman intervention in the indexing process The major challenges of the newlineCBIR systems are that of retrieving the exact or similar images as per the newlineusers request as well as improvement in the speed of retrieval irrespective of newlinethe database size newline newline