Construcción e implementación de un prototipo para reconocimiento e identificación de clientes bancarios usando técnicas de recognition y deep learning

Abstract

In a globalized world, the recognition and identification of people becomes relevant and more so in a financial environment in which there are various identification strategies, for this project identification by facial biometry is proposed for which the specification of the project is generated which has It was carried out in the Banco De Bogotá company, detailing its scope, limitations and objectives, whose main theme is the experimental study of how to recognize and identify people inside the Bogotá bank using Recognition and Deep Learning techniques. taking into account the costs. The prototype generated consists of identifying a person through the biometric metadata of their face, for which the capture of the frames is carried out through a device capable of capturing and processing images. Once the person is recognized, they will proceed to look for their identification number in the Banco de Bogotá database, there they will be identified by their names and surnames, followed by a greeting if they are an active client of Banco de Bogotá and otherwise. Giving you the option of obtaining a product if you are not a client or if you are an inactive client, it should be added that the generated prototype is based on a cloud architecture which will allow multiple implementations of the same prototype in different locations only having a stable internet connection. The prototype will offer benefits both to the bank with a proprietary, sustainable and extensible solution complying with the code quality guidelines and continuous integration, on the other hand, it will also benefit the user by offering another means of identification.