期刊名称：Oriental Journal of Computer Science and Technology
印刷版ISSN：0974-6471
出版年度：2010
卷号：3
期号：2
页码：273-282
出版社：Oriental Scientific Publishing Company
摘要：Considering multiple biometric templates per user account by biometric authentication systemsfor high acceptance rate leads to large storage space and computational overheads. Classification ofthese templates into significant sub groups will reduce the above overheads. Iris templates carry verydistinctive texture information such as brightness, shape, size, uniformity, directionality, regularity etc.Iris texture classification based on wavelet pattern analysis is one of the most effective existing methods.However using all frequency sub-bands in decomposition for classification may increase space andtime complexity of classification algorithms. In this paper sub-bands with high energy and entropy areonly considered for classification to reduce the overheads due to space and time. Fractal dimensionsare used to select significant sub-bands for decomposition at each level. Further statistical features ofthese significant sub-bands are used for classification. This paper describes iris texture classificationusing selective sub-bands of wavelets based on fractal dimensions and its results are compared withthe other classification methods using conventional features
