The spread of Covid-19 virus occurs quickly, one of which is through sneezing and coughing (drops or saliva). This becomes an infectious material for dentists in carrying out dental care during the pandemic. Extra-oral suction (EOS) is a device for sucking the patient's Aerosol when performing surgery or dental treatment, but the nozzle of the suction device is still manually moved, therefore the position of the EOS nozzle is right above the patient's mouth. This allows for some aerosol particles that have not been inhaled by the device when the patient turns or changes his head position. In this paper, Visual Servoing (VS) is needed, which is an approach to guide a robot using visual information. Image processing (face and mouth detection), and controls are combined to be able to move or change the position of the nozzle automatically according to the position and direction of the patient's mouth. The human face and mouth openness pose detection can be done using the Haar-cascade method and the Adaptive Boosting (AdaBoost) algorithm. This system is expected to optimize performance in dentist operations and minimize the transmission of the Covid-19 virus.