Sistema de detecção e rastreamento de pessoas por visão computacional

Abstract

The advancement of computer vision techniques has enabled the development of intelligent systems for real-time monitoring and security. In this context, this work proposes the development of a system capable of detecting and tracking human presence in environments such as classrooms and laboratories at UTFPR, Toledo campus, using only a webcam and a conventional computer. The proposal aims to optimize access control and occupancy monitoring of these spaces, offering an automated and low-cost alternative. For this purpose, the YOLOv8 object detection algorithm was implemented, known for its high accuracy and real-time performance in identifying people under different visual conditions. In the tracking stage, the SORT algorithm was employed, allowing the system to follow the movement of detected individuals over time. The methodology involved tests in controlled environments with variations in spatial arrangements to validate the system’s performance. The results demonstrated that the proposed approach is feasible, showing good performance in people counting and tracking. It is concluded that the application of computer vision techniques can significantly contribute to the improvement of monitoring and access control systems.