MONITORING PHYSICAL DISTANCE UNTUK COVID-19 MENGGUNAKAN SINGLE SHOT DETECTOR (SSD)

Corona Virus Disease 2019 (Covid-19) is an infectious disease that has a high spread rate. One of preventing the spread of Covid-19 is physical distancing. This study proposes to detection physical distance in the classroom by using image processing. Application of physical distance to monitor the distance between people in the class during the exam. According to WHO, the minimum distance in physical distance is one meter. In this research, the SSD Mobilenet method is used to detect human objects and will combined with the Euclidean algorithm so that it can be developed to detect distance between humans. There are several stages in the research process such as data collection, pre-processing of data, training data, and detection of physical distance. The input of test is video in real time, it was found that the SSD method can detect objects with an accuracy of 78.69%. And detection of distance between humans has an accuracy of 81.08%. In real time testing, the average value of FPS obtained is 3.89