Paper Details: Downloads: 1835
Serial Number: P1151536403
Title: Heterogeneous Features using S-Transform and Local Binary patterns for Non-ideal Iris Recognition
Authors: P.V.L. Suvarchala and S. Srinivas Kumar and B. Chandra Mohan
Abstract: Iris Recognition under non ideal imaging conditions like eyelash occlusions, rotation of eye and CCD noise etc., is a challenging problem that sought the attention of researchers. In the proposed method heterogeneous features are extracted using 2D Discrete Orthonormal Stockwell Transform (DOST) and rotation invariant Local Binary Patterns (LBP). The DOST provides frequency information where as the LBP provides the spatial textural information. The feature set size is reduced based on the entropy of the features, which are used to train and test the Support Vector Machines (SVM) for checking the classification accuracy. The verification performance of the proposed scheme is validated using the benchmark databases and the \% Correct Recognition Rate (CRR) is found to be more than (99\%) under non ideal imaging conditions also.
Keywords: Biometrics, Iris recognition, S Transform, LBP, Statistical Moments, SVMs.
Journal/Conference: International Journal of Graphics, Vision and Image Processing
Volume: 15
Issue: 2
Submission Date: 9/1/2015 12:00:00 AM
Review Date: 9/11/2015 2:37:38 PM
Publishing Date: 9/26/2015 3:10:45 PM
Article Downloads: 1835
Download:

Facebook