DMCA
Systems with Relevance Feedback
Cached
Download Links
BibTeX
@MISC{Kojić_systemswith,
author = {Nenad S. Kojić and Slobodan K. Čabarkapa and Goran J. Zajić and Branimir D. Reljin and Senior Member},
title = {Systems with Relevance Feedback},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstract — A content-based image retrieval system where an active learning strategy is used to gain relevance feedback (RF) is described. In this way retrieving process may be highly accelerated without significant degradation of accuracy Searching procedure was performed through the two basic steps: an objective one, based on the Euclidean distances and a subjective one based on the user’s relevance feedback. Images recognized from user as the best matched to a query are labeled and used for updating the query feature vector through a RBF (radial basis function) neural network. In this process user change feature vector which became more refined and appropriate for future search. In practice, several iterative steps are sufficient, as confirmed by intensive simulations. Index Terms — Content-based image retrieval, low level image descriptors, neural network decision, relevance feedback, subjective perception of images.
Keyphrases
relevance feedback intensive simulation euclidean distance significant degradation future search neural network decision subjective perception accuracy searching procedure content-based image retrieval system basic step index term content-based image retrieval query feature vector low level image descriptor process user change feature vector user relevance feedback active learning strategy radial basis function neural network several iterative step
