Elevator Door Automatic Stopper for Covid-19 Physical Distancing
Dian Andriana · 2021
Fast spreading covid-19 demand people to quickly cut the transmission over humans. Physical distancing is one of the efforts by keeping safe distance with at least 1 meter between people. This work is to find ways and implement computer vision methods to monitor, detect, and provide warnings of violation of physical distancing in public areas such as in elevators (lifts) which are equipped with monitoring cameras. Yolov3 still has problem of detecting half or full body proportion of human. Finally in this work, human proportional face over body detection has been checked whether it is full body or half body bounding boxes of human object detection to increase accuracy by 12.5 percent of distance measurement. Uncertainty of changing state of microcontroller has also been solved by collecting consecutive states which contain a number of warning/no warning condition of physical distancing violation.