Reliable automatic recognition of persons has long been an attractive goal of many researchers. Thus the recognition of an individual based on iris pattern is gaining more popularity due to the uniqueness of the pattern among the people which are highly stable starting from about one year past the date of birth, until death. The probability for the existence of two irises that are same has been theoretically estimated to be very high, i.e. one in 10 72 which counts for the unique characterization of the iris. Although many approaches for iris recognition have been proposed by many researchers in the last few years, in this paper a selective iris feature matching method for iris recognition based on optimized wavelet decomposition of normalized iris image has been proposed. Comparing the average normalised correlation of the wavelet coefficients of optimised level and its adjacent levels improved matching is obtained, thus performing uniqueness verification of a person.

No abstract

No abstract

No abstract

Human iris provides a unique structure suitable for non-invasive biometric assessment. So many researchers have been trying to improve the algorithm for reliable iris recognition. This paper presents a new iris matching method to control the misclassification error. Daubechies wavelet transform is used to extract the textural features. Combining the normalized correlation of the wavelet coefficients of optimized decomposition level with existing approach, efficient matching is obtained, thus performing person identification with much ease and evidenced.