DESENVOLVIMENTO DE UM SISTEMA AUTÔNOMO DE DETECÇÃO HUMANA EM CÂMERAS DE SEGURANÇA

This study proposes the development of an autonomous system for detecting human presence in real-time video images from cameras, focusing on security cameras. Using Mercado Reghini, located in Itápolis-SP, as a case study, the system seeks to face the challenge of distinguishing human movements from other sources of movement, such as animals, thus motivating the creation of an efficient solution. To develop the system, Python, OpenCV, Deep Learning, CNNs (Convolutional Neural Networks), Pytorch, Control Logic and other technologies were used, thus offering a solid and robust foundation. It will work as an extension for any camera brands and models. The purpose of this research is to identify the presence of individuals in the captured images and, when a person is detected, the system will take automated actions based on the level of the situation that is occurring. These functionalities aim to increase the efficiency and security of video surveillance systems, enabling future implementations in mobile application format for greater accessibility and dynamism. The proposal stands out for its potential for application in different environments. The system will be developed with a focus on scalability and adaptability, using suitable technologies for real-time processing and automated decision making. As an expected result, we seek to demonstrate the effectiveness of the system in detecting people and responding immediately to security events, without the need for constant human intervention.