A New Partitioning Method in Frequency Analysis of the Retinal Images for Human Identification

Retinal image is one of the robust and accurate biometrics methods to recognize a person. In this article we present a new biometric identification system based on Fourier transform and angular partitioning of the spectrum. In this method, at first, the optical disc is localized using template matching technique and used for rotating the retinal image into the reference position. It compensates the rotation effects which might occur during the scanning process. Fourier transform coefficient and angular partitioning of these coefficients are used for the purpose of feature definition in our method. The extract features are rotation invariant and robust against noise. Finally we employ Euclidean distance for feature matching. The proposed algorithm was tested using 40 images from DRIVE database and experimental results showed the efficiency of the proposed algorithm for the identification of retinal images with noise and rotation.