A robust image representation and learning method using svm for image retrieval

Abstract

In the past recent years, the faster growth in the public network ie Internet has rapidly increased availability of the large number of collection of image . The accumulation of these image collections is attracting more and more users in various professional fields [Rui Y et al., 1999]. Conventional methodologies are been employed to pop up the desired actual images by manually matching the framed keywords which are been assigned to every image of any format . Practical analysis depicts that, due to the collection of the uncountable images in number , manual analysis for processing image retrieval has become next to impossible. As since users are not likely to imagine or make out all the keywords of the query that are been employed in a finding similar images in the process of matching and above all keywords assignment to every individual image is practically not possible in turn this approach increases the effectiveness of the standard retrieval of image to limited capacity. newline