Santos, Jaine de Senna2023-07-242023-07-242023-07-19SANTOS, Jaine de Senna. Reconhecimento facial para a identificação de pessoas em ambientes restritos: um estudo avaliativo. 2023. 66 f. Trabalho de Conclusão de Curso (Graduação em Sistemas de Informação) - Departamento de Computação e Tecnologia, Centro de Ensino Superior do Seridó, Universidade Federal do Rio Grande do Norte, Caicó, 2023.https://repositorio.ufrn.br/handle/123456789/54046This paper presents an evaluative study that aims to investigate the feasibility and effectiveness of using facial recognition systems to identify and monitor individuals in restricted environments. The study utilizes the Haar Cascade algorithm from Viola-Jones for facial detection and conducts a comparative study between the LBPH and SVM facial recognition algorithms from the scikit-learn (sklearn) library. The specific objective is to determine which of these algorithms provides the most efficient and reliable solution for identifying individuals in controlled environments. The experimental results demonstrate that the proposed model is capable of accurately identifying individuals in closed environments. The comparative study revealed that both algorithms are effective in facial recognition, but SVM showed slightly superior performance in terms of accuracy. Based on the obtained results, it can be concluded that the computer vision-based facial recognition model presented in this work can significantly contribute to the security and management of restricted environments, especially in the university context.Attribution 3.0 Brazilhttp://creativecommons.org/licenses/by/3.0/br/Visão computacionalReconhecimento facialOpenCVLBPHSVMOpen Source Computer Vision Library (OpenCV)Local Binary Patterns Histograms (LBPH)Support Vector Machine (SVM)bachelorThesis