Office Room Smart Lighting Control with Camera and SSD MobileNet Object Localization

Abstract

There have been many studies that offer smart lighting control for the office. However, there is still room for improvement to enhance the performance of previous studies. This study proposes using a camera and SSD MobileNet object localization to detect the location of people in a room and control the lights according to the person’s position. The smart lighting system is built based on the Internet-of-things (IoT) architecture by implementing the Raspberry Pi (Raspi) and Raspi Camera for image capture and localization. The Raspi communicates with the smart lighting controller consisting of led lights, relays, and NodeMCU. The test bed of this research is an office room, where we create five zones inside it. One light controller controls the illumination of each zone. The test scenario is to ask a person to sit in each zone and sample the light controller 100 times. Then is to gather another 100 samples per zone with no one in the zone. True positive rate (TPR) and true negative rate (TNR) are the metrics to measure the performance of the proposed system. The test results show that the performance of smart lighting control in each zone is TPR=0.942 and TNR=0.956. Our research contributes to smart lighting control in the office room using a camera and object localization with better performance than existing research.