An Improved Iris Recognition Method Based on Gaussian-Hermite Moments
Di Zhou · Neimenggu Shi-da xuebao. Zhexue shehui kexue hanwen ban · 2008
An iris recognition algorithm is proposed based on Gaussian-Hermite moments.In the iris location,a previously proposed coarse-to-fine technique is adopted.A located iris is changed to a certain number of one-dimensional signals for extracting local information using Gaussian-Hermite moments.The results are coded with 0 or 1 so that classifications are obtained by Hamming distance measure.Experiments on the CASIA iris database obtain 98.55% classification rate and approximate 0.5 second for a single recognition.