Effect of Training Algorithms on the Accuracy in Iris Patterns Recognition using Neural Networks

TLDR: The results from the neural models trained by Levenberg- Marquardt algorithm is found to provide accuracy in recognition better than the methods presented in the literature.

Abstract: A biometric system provides automatic identification of an individual based on a unique feature or characteristic possessed by the individual. Iris recognition is regarded as the most reliable and accurate biometric identification system available. An approach for accurate Biometric Recognition and identification of Human Iris Patterns using Neural Network has been illustrated in (10). The same authors tried by reducing the size of the templates from 20 X 480 to 10 X 480 and concluded that this resulted in saving of computation effort with no loss in accuracy. In this paper, based on the accurate methodology (10(, we extend the work for optimization for Iris Patterns recognition using various neural training model algorithms. The results from the neural models trained by Levenberg- Marquardt algorithm is found to provide accuracy in recognition better than the methods presented in the literature.