Iris recognition system

Bakhita Salman, Tennessee State University


Biometrics systems give identification of humans by their characteristics or traits. It is also used to identify individuals that are under surveillance. Iris recognition system is one of the most reliable and accurate biometric identification system to date. Daugman's algorithm is the most accurate used now a day for iris recognition systems. Iris recognition as a biometric system has been an expanding field of research for the past several decades. Unfortunately, there is a lack of an open source code in Java. There is no fast, sufficient, accurate, and complete open source for iris recognition systems using Java. Therefore, there is a need for an open source for iris recognition system using Java. In this thesis we are developing an 'open-source' iris recognition system in order to verify the performance of iris as a biometric system. The database used in this research provided by the Chinese Academy of Science – Institute of Automation (CASIA). The database contains 1000 digitized grey scale eye images. Three steps must be completed in order for the system to function correctly, after image acquisition. These steps are: segmentation, normalization, and matching. To remove the noise from the image, we used the median filter. By using median filters we were able to smooth the whole iris image. The input to the system is an eye image, and the output is an iris template, which will provide a mathematical representation of the iris region. Circular Approximation technique was used to localize the circular iris and pupil region. Daugman's rubber sheet was used to normalize the extracted iris region into a rectangular block with equal dimensions. Finally, correlation coefficient was employed for classification of iris templates, and two templates were found to match if the correlation coefficient result is one or close to one. A prototype of the system was constructed and thoroughly tested. The prototype functions properly and meets all of the requirements of the system. The test results of the developed system are promising in providing reliable recognition. System performance and testing of the iris recognition software will be presented in this thesis.

Subject Area

Social research

Recommended Citation

Bakhita Salman, "Iris recognition system" (2012). ETD Collection for Tennessee State University. Paper AAI1533552.