Створення та дослідження системи безпеки з функцією локалізації на платформі Raspberry PI

Abstract

The paper substantiates the relevance of introducing object recognition algorithms in security systems. A security system has been developed based on Raspberry PI4 with built-in YOLOv10n architecture for human recognition. A study was carried out of recognition performance, the probabilities of correct and false recognition at different illumination levels. The efficiency of the designed system is proven and practical recommendations for choosing the input image resolution are given.