A Novel Image Classification System Based on Evidence Probabilistic Transformation

Abstract

This paper uses the evidence probabilistic transformation (EPT) for unsupervised image retrieval framework. The main advantages with EPT are substantially resolves the "take-them-or-leave-them" problem, gives a firmer epistemological basis for acquiring and for using in decisions. The proposed framework makes use of gray level co-occurrence matrix (GLCM) for images feature extraction. These features are used to provide a new axiomatic analysis and interpretation for images by using Dempster-Shafer belief functions. The mass functions for images information are combined under the normalized Dempster’s combination rule. The proposed model is characterized to extraction a set of accurate rules for images and their corresponding cluster based on probabilistic information in different format. Experiments show that the proposed model achieves a very good performance in terms of the precision, recall and F-measurement.

Authors and Affiliations

A. E. Amin

Keywords

Related Articles

An Improved Performance of MANET using AODV Protocol for Black Hole Detection

Routing attacks have been identified for single path routing in wireless ad hoc networks. The effects of routing attacks on multipath routing have not been addressed so far. In this paper, an approach have been propo...

Scale Down Speculation and Exalt Network Performance

In mobile Ad hoc Networks are having dynamic nature of its network infrastructure and it is vulnerable to all types of attacks. Among these attacks the routing attacks getting more attention because its changing the...

Client-Merchant Online Payment System Exploiting Visual Cryptography

This paper exhibits another methodology for giving restricted data just that is important for asset exchange amid web shopping along these lines defending client information and expanding client certainty and avoidin...

Public Key Encryption Using Key Aggregate Cryptosystem

Data contribution is an imperative functionality in cloud storage. For illustration bloggers can let their friends view a separation of their private pictures. An endeavour may contribution her employee’s contact to...

AN EFFECTIVE QUALITY SPEECH ENHANCEMENT USING PARTICLE FILTERS

Speech Enhancement aims to improve speech quality. The objective of enhancement is improvement in intelligibility and overall perceptual quality of degraded speech signal using audio signal processing techniques. Th...

Download PDF file
  • EP ID EP28142
  • DOI -
  • Views 236
  • Downloads 1

How To Cite

A. E. Amin (2015). A Novel Image Classification System Based on Evidence Probabilistic Transformation. International Journal of Research in Computer and Communication Technology, 4(2), -. https://europub.co.uk/articles/-A-28142