The development of the Internet of Things today is very rapid. one form of this development is a Smart-Building or smart building. Smart buildings require a form of system that can monitor the building in real time and efficiently. in this study designed a prototype form of a smart building system by adding a detector for the number of people using the ESP32cam camera sensor. The number of people object detection adapts the careful YOLOv3 Algorithm to 5 different lighting conditions, dimmed 50% and 75%, increased 50% and 75% and normal conditions. The results of this implementation show that the mAP value exceeds 90% in all of these lighting conditions. The YOLOv3 algorithm applied to the smart building prototype system has succeeded in detecting and counting the number of people objects in an image captured by the camera sensor. Keywords— Internet of Things; Object Detection; Smart Building; YOLO
Copyrights © 2023