Content Based Image Retrieval For Web Searching Application

Abstract

Retrieving the images from web pages is a trend in modern society. In order to retrieve an image; colour, texture and shape are three visual features that could help. Image retrieval systems attempt to search through a database to search images that are content fully same as that of query image. various methods are reported in literature for retrieval of an images. Previously text-based approach was used, but it was found tedious and difficult. A significant and increasingly popular approach that helps in the retrieval of image data from a huge collection is Content Based Image Retrieval (CBIR). Here, the CBIR approach with support vector machine (SVM) is proposed to be used for retrieval of an image. CBIR describes the approach of retrieving desired images from large collection of data on the basis of features. It is proposed to implement an efficient image retrieval technique which uses dynamic dominant colour, texture and shape features of an image.