Non-contact fever measurement with deep learning based face detection system

The common symptom of almost all pandemic diseases is high fever. Thus, in the simplest way, the detection of people with a fever of 38 °C and above, which is described as abnormal body temperature, will eliminate the risk of transmission, and ensure that the epidemic is brought under control. However, it is difficult to control these people in environments where hundreds of people enter and exit. This situation causes loss of time by requiring labor and cost. Therefore, in the advanced technology world we live in, it is necessary to automate this process by eliminating human influence. In this thesis, custom Single Shot Detection (SSD) was used to detect infected people faces because of Covid-19. It has been suggested to determine around the eyes area where the body temperature of the person most accurate by determining the facial landmarks with Ensemble of Regression Trees (ERT) model on these detected faces. Finally, the thermal value was measured from around the eyes area in a non-contact way using sensor fusion. As a result of the analyzes made, it was observed that the proposed system gave results close to the different measurement methods (Tuna & Akdemir, 2022).