Font Size: a A A

The Research On Iris Recognition Based On EMD

Posted on:2012-01-26Degree:MasterType:Thesis
Country:ChinaCandidate:Z J YuFull Text:PDF
GTID:2218330368487437Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
With the development of information technology in contemporary society, the demands for security are becoming more and more important. The technology of biometrics identification has an important effect on improving the information security, can be widely used in many fields. Iris recognition is an important component of biometric identification. Compared with other kinds of biological recognition technology, Iris Recognition has the following advantages: stability, uniqueness, collection and non-invasion etc.This thesis first gives detailed analysis and conclusion to main component of an iris recognition system, then studies some crucial algorithm deeply. The thesis also proposes some new algorithms based on above work. The main contributions of this work are as follows:Firstly, the improved Empirical Mode Decomposition Algorithm was introduced in this paper. The algorithm used piecewise power function to generate envelope curves and extending their end points. Experiments showed that the improved EMD can effectively solve overshoot phenomena and endpoint wing phenomenon in the decomposing process, make the EMD more precisely.Secondly A novel iris image enhancement algorithm based on BEMD is proposed. Experiments showed that this method can not only enhance an iris texture image's details but can also inhibited the affect of noise effectively.Thirdly In this paper, an iris feature extraction method based on EMD was proposed, in which iris signal is decomposed into several Intrinsic Mode Functions by EMD firstly. And then several IMFs suitable represent the most important information of iris images. The experiments show that the chosen IMFs can reduce noise interference and extract the iris features effectively and the proposed algorithm can achieve an excellent performance.
Keywords/Search Tags:Biometrics, Iris Recognition, Iris Localization, Feature Extraction, Iris Match
PDF Full Text Request
Related items