Iris Recognition based on Histogram Equalization and Discrete Cosine Transform

One of the most efficient and evolving methods in biometric identification is the iris recognition as the human iris has a unique texture, which represents a verity of details. This paper presents a technique based on histogram equalization and Discrete Cosine Transform (DCT) to capture the discriminative features of iris image. In order to investigate the performance of the proposed technique, Discrete Wavelet Transform (DWT) and Local Binary Pattern (LBP) have been implemented and compared their effectiveness to capture the features with Discrete Cosine Transform. This approach is applied on CASIA interval-v4 database. The results indicate significant achievement in iris recognition accuracies by using DCT compared with DWT and LBP.