Modification of Daugman's Integrodifferential Operator Using Bresenham's Algorithm for Iris Localization

In this paper, we have analyzed Daugman's integrodifferential operator which is used for iris localization. The several problems meeting at operator's implementation stage are considered in detail. One of these problems is selecting a step size for the numerical integration. We have presen...

Full description

Saved in:
Bibliographic Details
Published in2018 XIV International Scientific-Technical Conference on Actual Problems of Electronics Instrument Engineering (APEIE) pp. 183 - 187
Main Authors Minakova, Natalia N., Petrov, Ivan V.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2018
Subjects
Approximation algorithms
Bresenham's algorithm
Brightness
Computational efficiency
Daugman's integral-differential operator
IndexTerms -Iris localization
Iris recognition
Mathematical model
Reflection
Signal processing algorithms
Online AccessGet full text

Cover

More Information
Summary:In this paper, we have analyzed Daugman's integrodifferential operator which is used for iris localization. The several problems meeting at operator's implementation stage are considered in detail. One of these problems is selecting a step size for the numerical integration. We have presented examples of cases when using large step size leads to localization error. The detailed analysis of such examples is provided. The issues with a computational efficiency appear when using too short step. To increase the efficiency, we have proposed the method based on Bresenham's algorithm for computation of operator's integral part. Adaptation of Bresenham's algorithm allowing to compute operator in separate arcs has been proposed. The algorithm uses only integer-based arithmetic which improves the speed and accuracy. A high speed of the proposed method in comparisons with the conventional one has been confirmed.
ISSN:2473-8573
DOI:10.1109/APEIE.2018.8545151