•   Home
  • Proceedings
  • 2012 International Conference on Computational Intelligence ...

2012 International Conference on Computational Intelligence and Software Engineering (CiSE 2012)(E-BOOK)

Wuhan,China,2012-12-142012-12-162012

ISBN: 978-1-61896-036-8 Scientific Research Publishing

E-Book 275pp Pub. Date: December 2012

Category: Computer Science & Communications

Price: $100

Title: Enhancing SVM Active Learning for Image Retrieval Using Unlabeled Data
Source: 2012 International Conference on Computational Intelligence and Software Engineering (CiSE 2012)(E-BOOK) (pp 131-134)
Author(s): Guizhi Li, Computer Center, Beijing Information and Science and Technology University,Beijing, China
Abstract: Relevance feedback (RF) schemes based on support vector machine (SVM) have been widely used in content-based image retrieval to bridge the semantic gap between low-level visual features and high-level human perception. However, the Conventional SVM based RF uses only the labeled images for learning which gives rise to the small sample problem. When the training data is insufficient, the performance of SVM may drop dramatically. In this paper, we proposed a method to alleviate the small sample problem in SVM based RF by using semisupervised active learning algorithm which uses a large amount of unlabeled data together with labeled data to build better models. In relevance feedback, active learning is often used to alleviate the burden of labeling by selecting only the most informative data. In addition, a semi-supervised approach has been developed which employs Bayes Classifier to label the unlabeled data with a certain degree of uncertainty in its class information. Using these automatically labeled samples, Fuzzy support vector machine (FSVM) which takes into account the fuzzy nature of some training samples during its training is trained. We compared our method with standard active SVM based RF on a database of 10,000 images, the experiment results show that our method has a better performance and prove that it is an effective algorithm for CBIR.
Proceedings by Subjects
Resources
Products
Contact us
book@scirp.org
WhatsApp +86 18163351462(WhatsApp)
Click here to send a message to me 1243940697
Book Publishing WeChat
Follow SCIRP
Free SCIRP Newsletters
Copyright © 2006-2024 Scientific Research Publishing Inc. All Rights Reserved.
Top