Automated identification of people based on biometric facial recognition processes has been increasingly utilized for different purposes and in a wide variety of applications over the past years. The use of Artificial Intelligence techniques, particularly Deep Learning, has led to significant improvements in the accuracy of correctly identifying individuals in different scenarios. However, challenges arise when facial recognition systems encounter images of individuals who were not included in the training phase, leading to potential misclassifications. This corresponds to an open-set face recognition scenario, where the ability to detect unknown identities introduces an additional challenge for the accuracy of facial recognition systems. In this context, this paper analyzes the impact of system scalability on recognition performance as the number of images and identities involved increases. Starting from a pretrained model, training it with images of people who must be recognized by the system, and applying techniques that optimize the identification of unknown individuals, this paper presents the most significant aspects observed in the tests conducted.