Bio-Medical Image Retrieval Using SVM
Journal Title: International Journal of Advanced Research in Computer Engineering & Technology(IJARCET) - Year 2012, Vol 1, Issue 10
Abstract
This paper presents an image retrieval system in a computer system for browsing, searching and retrieving images from a large database of digital images. The support vector machine (SVM) is a concept based on learning methods that evaluate data and identify patterns, used for classification and regression analysis. In this work, the prediction of database and query image depends on the outputs of a multiclass support vectormachine (SVM) classifier. The irrelevant images are first to filter out, for similarity matching. The different low-level featuresmodalities are used in this framework. When compare to the query image, the main difficulty is to find unique feature. So an adaptive linear combination of similarity matching and user’s feedback information is used here. The predicted output of classifier may be inexact in some cases. Hence, the feature weights are calculated by considering both precision and rank order information based on users feedback. As a result, using SVM classifier with the retrieval performance clearly shows the advantageof searching images based on similarity fusion and filtering in terms of competent and success.
Authors and Affiliations
S. Nithya, , G. ShineLet,
Keywords
Related Articles
- EP ID EP146613
- DOI -
- Views 60
- Downloads 0
Particle Swarm Optimization in Cryptanalysis of DES
In this paper for the first time Particle Swarm Optimization is applied in the field of cryptanalysis of DES based on their ability to selectively explore the solution space of a given problem. PSO is a robust stoc...
Implementation of Citation Parser on the Basis of Knowledge Base Hierarchy
Use of the bibliographical information of publications available on the Internet is an important task in academic research. Accurate reference metadata extraction for publications is essential for the integration...
Wavelet Based Classification of Multispectral Satellite Image Using Fuzzy Incorporated Hierarchical Clustering With SVM Classifier
Multispectral satellite images are more efficient and a suitable method of obtaining information about land, because it can captures an image at specific frequency across the spectrum. This spectral image can allow...
Survey on Data Integrity in Cloud Computing
Data outsourcing to cloud storage has recently become a trend. Economic advantage is the significant reason. A large number of clients or data owners store their important data in remote servers in the cloud and it...
DIFFERENTIATION OF TERM AND PRETERM DELIVERY GROUPS BY EHG SIGNAL ANALYSIS
This article deals with the analysis of Uterine EMG (EHG) signal of a pregnant woman to reveal whether the delivery is going to be preterm or term. In the studied project, various linear and non-linear signal-processi...