Fuzzy Dynamic Parameter Adaptation for Gray Wolf Optimization of Modular Granular Neural Networks Applied to Human Recognition Using the Iris Biometric Measure

Abstract:

In this chapter, a gray wolf optimizer with dynamic parameter adaptation based on fuzzy theory for modular granular neural network (MGNN) has been proposed. These architectures have several parameters, such as the number of sub-granules, the percentage of data for the training phase, the learning algorithm, the goal error, the number of hidden layers and their number of neurons, and the optimization that seeks to find them. The effectiveness of this optimization with its fuzzy dynamic parameters adaptation is proved using a database of iris biometric measures. The advantage of the proposed fuzzy dynamic parameters adaptation is focused on determining the gray wolf optimizer parameters depending on the population behavior, i.e., depending on current results, the parameters are adjusted to improve results. Human recognition is an important area that can offer security to areas or information, especially if …

Año de publicación:

2022

Keywords:

Fuente:

google