Research On Iris Recognition Preprocessing Algorithm Based On Information Variation

With the rapid development of network information technology,how to protect the security of user information has become a hot topic.The identity authentication system is a comprehensive system used to protect the security of user information.Currently,there are three authentication methods in the identity authentication system,namely the authentication of trusted objects,the authentication of biometrics and the authentication of information secrets.Of identity authentication based on biological characteristics of technology,more and more get the attention of the scientific research scholars in many biological characteristics,the iris identification authentication technology due to the uniqueness of the iris itself,stability,reliability and high security advantages favored by the majority of scientific research workers,so the iris recognition technology of identity authentication products are applied in areas such as access control,security system and confidential security.Iris quality evaluation is an important module in iris recognition system.Iris image quality evaluation can improve the performance of iris recognition.Iris image quality evaluation plays an increasingly important role in image acquisition,human-computer interaction,recognition performance prediction and adaptive iris recognition algorithm design.In recent years,with the development of the iris recognition system development,iris quality evaluation has been widely studied,especially with the advent of the era of big data,large populations will become the research hotspot of iris recognition,therefore the iris image quality evaluation research to improve the processing capacity of the iris recognition system is of great significance.Iris image quality evaluation is one of the key steps in the pre-processing of iris recognition.As the iris image collection environment is complex and changeable,the image collected under the ideal environment is called constrained iris.On the contrary,it is unconstrained iris.In addition,in the actual collection of iris images,what is collected is not an image but an iris video stream,also known as a sequential iris.In order to select the iris image with good quality and can be recognized from a large number of iris images,a constrained state sequence iris quality evaluation algorithm based on decision reasoning is proposed in this paper,which is based on the iris taken in the constrained ideal state as the research object and improves the efficiency of iris quality evaluation.In the case of other indicators set idealized,constraint condition were collected for the infrared camera,the iris image,from the definition,the effective iris area and the degree of deviation Angle set a variety of iris quality evaluation index,based on the statistical characteristics of the way,according to the feedback of the evaluation results,the various index choice evaluation index with good effect,and adjust target size,achieve the goal of static index selection.According to the causal relationship between the quality indicators,the order of the evaluation of the indicators should be adjusted to form the reasoning process of quality decision.Feature extraction and recognition of iris.Two-dimensional Gabor filter was used to extract feature texture information,and Hamming distance was used to judge the category of test iris in iris recognition.The experimental data were collected from the iris library independently collected by the biometrics and information security technology laboratory of jilin university.The correct recognition rate(CRR),equal error rate(EER)and ROC curve were used as the experimental evaluation indexes,and the efficiency of the algorithm was compared with other methods.Experimental results show that the algorithm can effectively improve the survival rate of available iris in the iris sequence and play a positive role in improving the iris recognition accuracy.