MATLAB based Iris Pattern Recognition System

With the growing demand for secure identification systems, iris recognition has become a key biometric authentication method. This project focuses on implementing an iris recognition system inspired by the method developed by John Daugman, with some modifications in MATLAB The process begins with image preprocessing, where the iris region is extracted from the Visual Biometric Image. Next, the isolated iris is then adjusted to a standard format, and an Iris Code is created using 1D Gabor filters to capture its unique features. To verify identity, which quantifies the difference between them. The Hamming Distance method is used to compare two iris codes, measuring the differences between them to determine identity verification. The iris recognition algorithm developed by the author has been extensively tested in both laboratory and real-world environments, with millions of comparisons showing no false matches. This method works by analysing the unique phase structure of the iris, which is encoded using multi-scale quadrature wavelets. The system detects identity by checking for statistical independence between different iris patterns. Due to the high complexity and variability of iris features across individuals spanning approximately 249 degrees of freedom— the recognition process achieves exceptionally high accuracy. With the variation in iris patterns contains 3.2 bits of unique information per square millimetre, allowing for highly accurate identification in biometric systems, this system ensures reliable real-time identification. The high confidence level makes it possible to search through large databases in A one-to-many identification system can accurately compare a single input with multiple stored records without errors or mismatches. In contrast, other biometric methods, which lack such precision It can only be used for verifying a single identity at a time or for a small number of comparisons.