الفهرس 
introductionOnly 14 pages are availabe for public view
 |  | from | 124 |  |  |
| | | from | 124 | | |
Abstract
Iris recognition is considered as one of the best biometric methods used for human identification and verification, this is because of its unique features that are different from one person to another, and its importance in the security field.
There are many techniques used for iris recognition differ from each others in the way of identifying and verifying human irises, merging two or more of these methods will produce good results. This thesis proposes two algorithms for iris recognition using hybrid systems based on merging two or more approaches together to build a hybrid system which is able to recognize and classify input iris images.
Firstly, Iris Recognition System will pass a series of steps starting with the preprocessing step. The localization and segmentation techniques are presented using both Canny edge detection and Hough Circular Transform in order to isolate an iris from the whole eye image and for noise detection.
Secondly a system based on Haar wavelet and Hidden Markov model is introduced in order to extract the iris features, and matching iris images respectively. Another system will be presented using the Local Binary Pattern (LBP) and histogram properties as a statistical approaches for feature extraction, and Combined Learning Vector Quantization (LVQ) Classifier as Neural Network approach for classification, in order to build a hybrid model depends on both features.
Feature vectors extracted from LBP is applied to a Combined LVQ classifier with different classes to determine the minimum acceptable performance, and the results are based on majority voting among several LVQ classifiers. Different iris datasets CASIA, MMU1, MMU2 and LEI with different image size formats are presented.