UCSY's Research Repository

Combination of Keyword and Visual Feature based Image Retrieval System

Show simple item record

dc.contributor.author Kyaing, Htwe Htwe
dc.contributor.author Nge, Mi Mi
dc.date.accessioned 2019-07-12T05:04:28Z
dc.date.available 2019-07-12T05:04:28Z
dc.date.issued 2010-12-16
dc.identifier.uri http://onlineresource.ucsy.edu.mm/handle/123456789/858
dc.description.abstract Keyword-based image retrieval systems have become popular for many image database applications. To improve the performance of keyword-based web image queries, combination of keyword and visual feature based image retrieval system is presented in this paper. Firstly, DOM (Document Object Model) trees are constructed from collected web pages. And several text blocks are segmented based on text cohesion. Then, visual features are extracted from color images in RGB (Red, Green and Blue) color space by using color histogram. When user query is entered, text blocks which contain web images are taken as the associated texts of corresponding images and TF*IDF values are used to index web images. Finally, keyword and visual features are combined by using Gaussian Mixture Model to produce the relevance images. en_US
dc.language.iso en en_US
dc.publisher Fifth Local Conference on Parallel and Soft Computing en_US
dc.subject DOM tree en_US
dc.subject text cohesion en_US
dc.subject keyword en_US
dc.subject visual feature en_US
dc.subject Gaussian mixture model en_US
dc.title Combination of Keyword and Visual Feature based Image Retrieval System en_US
dc.type Article en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search Repository



Browse

My Account