CBIR Representation In Terms of Rotation Invariant Texture using LBP Variance
Rushikesh T. Bankar1, Rudra Prasad Patra2, Arjun Choudhari3, Gaurav Jasutkar4

1Rushikesh T. Bankar, Department of Electronics & Telecommunication Engineering, G. H. Raisoni College of Engineering, Nagpur, India.
2Rudra Prasad Patra, Student Department of Electronics & Telecommunication Engineering, G. H. Raisoni College of Engineering, Nagpur, India.
3Arjun Choudhari, Student Department of Electronics & Telecommunication Engineering, G. H. Raisoni College of Engineering, Nagpur, India.
4Gaurav Jasutkar, Student Department of Electronics & Telecommunication Engineering, G. H. Raisoni College of Engineering, Nagpur, India.

Manuscript received on March 11, 2014. | Revised Manuscript received on March 15, 2014. | Manuscript published on March 25, 2014. | PP:55-57 | Volume-2 Issue-5, March 2014. | Retrieval Number: E0699032514/2014©BEIESP

Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Local rotation invariant feature extraction has been widely used in texture classification.. This paper proposes an alternative hybrid scheme, using LBP distribution, we first estimate the principal orientations of the texture image and then use them to align LBP histograms. Then the aligned LBP histograms were in turn used to measure different images from the database. A new texture descriptor, LBP variance (LBPV), is proposed to characterize the local contrast information into the one-dimensional LBP histogram. For more accurate result we propose a method to reduce feature dimensions using Euclidian Distance measurement. The experimental results of the databases show that the proposed LBPV operator can achieve significant Improvement, sometimes more than 10% in terms of classification point of view, over traditional locally rotation invariant LBP method.
Keywords: The experimental results of the databases show that the proposed LBPV