Análise de algoritmos de biometria facial para controle de pessoas em espaços privativos

Abstract Identifying people, whether through facial or biometric recognition, is associated with services that are beginning to be deployed around the world, whether in a public place, a large or small business, or even a simple condominium, for greater control. and organization in the active environment. However, the process to accomplish this type of task requires the involvement of new technologies, skilled people and adequate financial investment. Regarding costs, you can have a variation according to the purpose of the application and its place of operation, from a small bakery, to a large football stadium, for example. The main objective of this work is to analyze different types of facial detection and recognition algorithms for Facial Biometrics activity, aiming to show the interaction and differences of these algorithms. Some facial recognition algorithms have been used for comparisons and their advantages, comparative results are combined with other facial recognition methods, such as Haar Cascade with KNN or even using SVM, or decision tree, etc. Python and its associated OpenCV recognition libraries and the Dlib and Face Recognition facial recognition library, which has a sharper, more market-oriented accuracy, will be available as part of this work. . To generate the results, images recorded by the author and public databases available on the internet were used.