CONTROLE DE ACESSO A EVENTOS UTILIZANDO RECONHECIMENTO FACIAL
Carregando...
Data
2023-11
Autores
Orientador
Pretto, Fabrício
Banca
Diemer, Mouriac Halen
Wolf, Alexandre Stürmer
Título do periódico
ISSN
Título do Volume
Editor
Resumo
O objetivo deste trabalho foi desenvolver uma aplicação de reconhecimento facial para dispositivos móveis para controle de acesso e reconhecimento automatizado de pessoas. A justificativa do desenvolvimento do sistema se deve à comum existência de filas em eventos sociais. Soma-se isso, a possibilidade de falsificação de documentos e a perda dos métodos garantidores de acesso. A autenticação é a base para o desenvolvimento da pesquisa e é feita por biometria, neste caso identificação facial. O estudo, conduzido de forma exploratória, se direciona à procura de técnicas para a implementação de uma ferramenta acessível e de código aberto. Focou-se na área de reconhecimento facial para controle de acesso. Foi criada uma solução utilizando as linguagens de programação Python, para o processamento, a qual apoia se na biblioteca face_recognition para realizar o reconhecimento. Para o mobile, foi utilizado React Native com auxílio do framework Expo para detecção de faces e estruturação do aplicativo. Com o desenvolvimento completo, realizaram-se testagens piloto e de campo, as quais apontaram as falhas iniciais e posteriormente, corrigidas, os resultados. Esses, por sua vez, são muito satisfatórios, o algoritmo reconheceu corretamente cerca de 90% das vezes e em tempo hábil, de aproximadamente cinco segundos.
The objective of this work was to develop a facial recognition application for mobile devices for access control and automated people recognition. The justification for the development of the system is due to the common existence of queues at social events. Additionally, there is the possibility of document forgery and the loss of access assurance methods. Authentication is the foundation for the research development and is done through biometrics, in this case, facial identification. The study, conducted in an exploratory manner, focuses on finding techniques for the implementation of an accessible and open-source tool. It focused on the area of facial recognition for access control. A solution was created using the Python programming language for processing, which relies on the face_recognition library for recognition. For the mobile aspect, React Native was used with the support of the Expo framework for face detection and application structuring. With the complete development, pilot and field testing were conducted, which identified initial flaws that were later corrected, yielding very satisfactory results. The algorithm correctly recognized faces about 90% of the time and in a timely manner, approximately five seconds.
The objective of this work was to develop a facial recognition application for mobile devices for access control and automated people recognition. The justification for the development of the system is due to the common existence of queues at social events. Additionally, there is the possibility of document forgery and the loss of access assurance methods. Authentication is the foundation for the research development and is done through biometrics, in this case, facial identification. The study, conducted in an exploratory manner, focuses on finding techniques for the implementation of an accessible and open-source tool. It focused on the area of facial recognition for access control. A solution was created using the Python programming language for processing, which relies on the face_recognition library for recognition. For the mobile aspect, React Native was used with the support of the Expo framework for face detection and application structuring. With the complete development, pilot and field testing were conducted, which identified initial flaws that were later corrected, yielding very satisfactory results. The algorithm correctly recognized faces about 90% of the time and in a timely manner, approximately five seconds.
Descrição
Palavras-chave
Autenticação; Mobile; Eventos; Authentication; Events
Citação
AUDIBERT, João Pedro Basso. CONTROLE DE ACESSO A EVENTOS UTILIZANDO RECONHECIMENTO FACIAL. 2023. Monografia (Graduação em Engenharia da Computação) – Universidade do Vale do Taquari - Univates, Lajeado, 13 dez. 2023. Disponível em: http://hdl.handle.net/10737/4233.