Fast iris recognition algorithm based on Gabor filtering
Xiaoning Li · Jisuanji gongcheng yu sheji · 2011
To improve the performance of iris recognition,an improved algorithm is proposed.On the stage of iris location,the real boundaries of the pupil are obtained through edge correcting for the pupil,and the morphology method is introduced to fill up the gaps among the outer edges that are detected by Canny operator.Then the direct least square fitting of ellipse and circle could be used for de-tecting the boundaries of the pupil and the iris separately.2D-Gabor filters with different frequencies and orientations filter the every sub-block in the iris normalized image,and the iris texture is encoded by plus or minus to the coefficient of the imaginary unit.Experiments are shown the applicability and effectively of the algorithm.And the proposed method can class the irises precisely and quickly.