Case study 1: image processing for medical application of preventing the spread of COVID-19

This study presented security system equipped with an ability to detect body temperature to reduce the spread of COVID-19. Generally, temperature screening was done using a thermometer gun. However, its usage was limited and cannot be used together as a security system. Thus, this study developed a smart lock door system with a contactless temperature screening based on Internet of Things (IoT). The built system consists of an ESP32, microcontroller, the MLX90640 sensor as temperature screening, solenoid door lock, the Haar cascade algorithm for face detection, the local binary pattern histogram algorithm for face recognition. The system may send temperature data to the database website, sending an alert if the object has a temperature above 38°C. The results of this study showed that the average temperature difference between thermos gun and the MLX90640 was 0.1°C and 0.09°C at a distance of 10 cm and 12 cm, respectively. Those distances were considered as the effective distance. The proposed face recognition succeeded in recognizing the object with an accuracy of 80% in dark condition and 100% in bright condition. Face recognition and temperature screening were successfully sent to the database with a time difference of sending data was 0.3 s. An alert can be sent to the Telegram in real-time manner when the temperature and face were recognized as a suspect of COVID-19. The security system cannot be opened when the recognized face was not in the database.