Study On Key Issue In The Stability Of Iris Feature Extraction

With the development of information technology,information security has become an important issue.As an important means of information security,personal identification has been widely used in various fields of society.Biometric identification technology due to its important characteristics,has been widely concerned and applied,the iris recognition technology because of its own characteristics(high accuracy,uniqueness,stability,security and non-contact)and has been considered as a high research value and application prospect of biological recognition technology.However,in practical applications,the iris recognition technology still has many defects,such as iris image interference,low-quality iris positioning is not accurate and the improper selection of iris feature and other factors will lead to poor stability of iris feature extraction,which reduces the performance of the recognition system.Therefore,it is necessary to study the key technologies that affect the stability of iris features.Firstly,this paper summarizes and analyzes the domestic and foreign research status of iris recognition technology,iris recognition of several current iris recognition algorithm is mainly studied,and the related algorithms are compared;And then,from the whole detailed introduction of the iris recognition system,the contents of each part,and the use of the algorithm;finally,detailed introduces the innovation and improvement of the key technology of iris recognition on the experimental analysis of the improved algorithm in the Chinese Academy of Sciences CASIA_V1.0 database and self-built database on BA.The main contents of this paper are as follows:1.In order to eliminate low-quality iris images(different acquisition equipment and acquisition environment)effect on the performance of the system,this paper adopts the iris image quality assessment algorithm based on Fourier transform,low quality image acquisition removal,reduce the error rate of iris recognition system.2.The outer circle of the iris localization is affected by the quality of the image,the search range is time-consuming,studied a kind of integral and differential positioning based on the statistical knowledge of the outer circle of the iris localization algorithm,the positioning accuracy of the narrow circle,improve the positioning speed.3.Study of the local details of the iris features,a feature extraction method isproposed based on multi-channel Gabor filter to quantify the local details of the iris block,according to the phase information combined into a feature vector,in order to iris feature encoding.Experiment on the sample image database.The experimental results show that the iris image quality evaluation Fourier transform algorithm,the outer circle of the iris localization algorithm and iris feature extraction algorithm based on feature vector key points based on iris feature extraction,which has high stability,and improve the recognition rate and recognition system the correct rate,meet the requirements of the development technology of the existing iris recognition system.