Abstract

Biometrics provide an easy and accurate solution for identity authentication which links identity authentication to individuals unique physical attributes. So, these methods work based on what you are not what you know (password), or what you have (identification card). Iris recognition is one of the safest and most reliable biometrics. In general, iris recognition is based on individual identification system, including iris imaging, estimating iris image quality, detecting fake iris, and recognizing iris. This study aimed to analyze iris recognition in identity authentication. In this theoretical research, data were examined descriptively and analytically. The results showed that the first step in iris recognition is to isolate the actual iris area in the digital image of an eye. In the second step, i.e. normalization process, iris area with identical stable dimensions is generated, so that two shots of the same iris in different situations will have the same characteristics in identical space. In the next step, to provide accurate identity authentication, most of the discriminatory information available in iris pattern is extracted. In the fourth step, the attribute extracted using a one-dimensional Gabor logarithmic filter, FFT, and IFFT is entered as an input into phase quantification process to generate iris pattern with binary values of 0 and 1. This result is known as iris code. Finally, for adaptation, the hamming distance is selected as a measure of identification, hence a bit comparison is required. Hamming distance algorithm uses only the main bits to calculate a hamming distance between two iris codes.