A Research On Some Key Technology Of Iris Recognition

With the quick development of Internet and information technology,and the increasing human's consciousness on security, the trandationalsecurity technology doesn't meet the requirement of security quality of thepeople at present. Then the people turn to biology recognition technologyand make it into a fast developing phase. Iris recognition technology withits huge potent is relatively new among different kinds of biometricrecognition technologies for its advantages, such as the unique andreliability characteristics of an iris all through one's life, and it can't forgeand infringe.This paper is divided into two parts, former part elucidates theprinciples and the typical structure of Iris-Identification System, givebriefe introduction to the background of this dissertation and currentsituation research on the iris recognition technique. The latter mainlystudy some continuous key iris recognition arithmetic on the base ofexisiting algorithms and give the experiment result finally. In the processing of iris recognition, the individual difference isobvious especially the interference of eyelash and eyelids, the existingmethod can not distill the pure iris area accurately. Another problem is theabout the feature extraction process, it needs either excessive work or toolittle eigenvalue.When comes to solution of above problems, this dissertation mainlydo following works:1) Study characteristic of the eyelid and eyelash, also thedifferece from iris pixels, a fast eyelash and eyelids segmentationalgorithm is presented based on former investigation results.2) Aim at the different interference situation of eyelash andeyelids, the algorithms of iris adaptive recognition is put forward. ForThen, adaptively locate the iris area with a fast method consequently, itkeeps the most iris information with little infection of non-iris area.3) In the feature extraction processing, the theory oftree-structured wavelet transforms is put forward based on the analysis ofwavelet, wavelet package and multiresolution analysis theory. Finally thealgorithm of matching based on Hamming Distance is applied to matchingthe iris eigenvalue.The experimental results are shown and analyzed, it shows that thisadaptive recognition algorithm is faster and more accurate, this encodingmethod is reasonable and effective. At the end of this dissertation someexisting problems are pointed out, also the good future of iris recognition technology is predicted.