Evaluation of Basic Noise Removal and Image Enhancement Techniques for Iris Database

Automated person identification is very important to improve the security level especially in highly sensitive areas. Human retina provides a reliable source for biometric based system which is almost impossible to forge. Here, we evaluate the performance of two basic stages of automated iris recognition, noise removal and image enhancement. Performance of various basic noise removal algorithms and image enhancement algorithms is evaluated on iris datasets. The performance of noise removal techniques is evaluated using PSNR, RMSE and correlation of coefficient. The performance of image enhancement is evaluated using Absolute mean difference, RMSE and Entropy. The work will provide a baseline for the researchers working in iris recognition for choosing the best noise removal and image enhancement technique.