Iris Recognition Algorithm Based On Wavelet Transform

With the rapid development of the information technology and the increasing requirement for security, automated personal identification based on biometrics has been receiving extensive attention over the past decade. The biometric recognition technology has very high reliability, because it bases on some personal unique characteristics which don't change for life. In addition the technology has not the hidden troubles, such as password loss.In all kinds of personal biometric characteristics, such as fingerprint, gait, voice, iris, etc., the personal identification based on iris has more advantages than others. Currently, it has been agreed that the iris recognition technology is most promising one of the biometric recognition technology.The process of the iris recognition includes iris imaging, pretreatment, characteristic obtaining, and pattern matching, and our research work mainly aims at pretreatment, characteristics obtaining, and pattern matching.In the paper we will introduce the birth, development and current situation of research on the iris recognition technology and analyze some famous algorithms, and then we will propose our own algorithms. In order to the accuracy of locating iris exterior edge and the speed of iris localization, we will propose an improved algorithm basing on histogram equalization and coarse-to-fine localization. In the aspect of obtaining characteristics and matching, we utilized wavelet transform to detect the signal mutations, and then we matched the signal base on the comparability of the mutations so as to obtain the result of matching. Through experimentations, the algorithm proposed in this paper is proved to be valid and efficient.At the end of some problems existing in this paper will be pointed out, and then the good future of the iris recognition technology will be predicted.