Title: Sign LTP and magnitude LTP for image indexing and retrieval

Authors: K.N. Prakash; K. Satya Prasad

Addresses: Department of Electronics and Communication Engineering, Jawaharlal Nehru Technological University, Kakinada, Andhra Pradesh, India ' Department of Electronics and Communication Engineering, Jawaharlal Nehru Technological University, Kakinada, Andhra Pradesh, India

Abstract: In this paper, a new algorithm for image indexing and retrieval is proposed by using the combinations of sign and magnitude of local ternary patterns (LTPs), which are calculated from the Local Difference Operator (LDO). The LDO separates the local region of the image into two components (the sign and the magnitude). The sign LTP operator (SLTP) is a generalised LTP operator and the magnitude LTP operator (MLTP) is calculated using the magnitude LDO (MLDO), weighted by the mean of the MLDO. The retrieval results of the proposed method are tested on three different image databases, i.e., the Corel 1000 (DB1), the Brodatz database (DB2) and the MIT VisTex database (DB3). The results after investigation show a significant improvement in terms of average retrieval precision and average retrieval rate as compared to previously reported spatial and transform domain methods.

Keywords: image retrieval; image indexing; texture; local ternary patterns; sign LTP; magnitude LTP; local binary patterns

DOI: 10.1504/IJSISE.2013.056635

International Journal of Signal and Imaging Systems Engineering, 2013 Vol.6 No.4, pp.227 - 239

Received: 23 Aug 2011
Accepted: 12 Jan 2012
Published online: 30 Sep 2013 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article