IRIS RECOGNITION OPTIMIZED BY ICA USING PARALLEL CAT SWARM OPTIMIZATION

Feature selection is an optimization technique used in Iris recognition technology. For producing the most accurate recognition of iris from the database, feature selection removes the unrelated, noisy and unwanted data. Parallel Cat Swarm Optimization Algorithm is one of the latest optimization algorithms in the nature league based algorithm. Its enhancement results are better than the PSO and CSO optimization algorithms. The proposal of applying the Parallel Cat Swarm algorithm is mainly used for feature selection in the process of Iris recognition. For human identification iris can be used as it is an integral part of the human body. Biometric iris recognition system compares the two iris images and produces a matching score to determine their degree of equality or inequality. Eyelid and eyelash are considered to be the unwanted parts of the eye apart from iris. By using Structure Tensor Analysis we can mask the unwanted parts of iris by taking the iris as region of interest. By using Independent Component Analysis, we can extract the texture feature in the iris from the eye. The best features are then selected using Parallel Cat Swarm algorithm from the extracted texture features. For identification purpose we need to compare the best feature with a number of features of various individuals in the database.