A novel algorithm of human iris recognition

This paper propose a new collection of partial iris images from full iris images instead of the traditional iris normalization. Because of the distortion and eyelid covering appearing in the traditional iris normalization, the following iris-coding algorithm will appear distortion accordingly, and the matching will appear error even within the inner iris image class. This collection of partial iris image method is presented using an isometric iris image-collecting model. Moreover, we describe and analyses an iris-coding technique based on Gabor filter. At last, the performance of the proposed algorithm is analyzed using CASIA database2.0. The original images are non-ideal and thus require a sequence of pre-processing steps prior to application of encoding method. The results show that our algorithm is feasible.