ITB Lecturer Team Develops Non-Contact Body Temperature Thermal Cameras

April 30, 2020 - ( update : 20-05-2020 )

By Adi Permana

Editor Adi Permana

Initial Test

To measure body temperature in a non-contact manner in an effort to reduce the spread of COVID-19 to the current standard procedure, because according to WHO guidelines one of the indications since infection with COVID-19 started in 5-14 days will cause body temperatures greater than 38oC or known as fever symptoms. The use of the infrared gun thermometer is a technique to measure the temperature of people entering a certain room.

The use of a thermal camera with a resolution of 300x200 is equivalent to a temperature measurement using a 60 thousand infrared gun thermometer, so it is possible to measure the temperature of the body quickly and precisely in a queue of people , for example at the train station.

Thermal cameras measure body temperature based on thermal radiation from the body, so that many parameters of the object, including emissivity, distance of the focus area and scanning procedures, must be calculated in the system. Internationally, references used for the use of thermal cameras to obtain maximum accuracy are given guidance by ISO / TR 13154: 2017 on "deployment, implementation and operational guidelines for the identification of febrile humans using a screening thermograph."

In order to maintain the consistency of body temperature measurements and reduce the cost of installing a thermal scanner, ITB has developed a reference prototype with specific emissivity on a plate that is maintained at 37.5 ± 0.2oC. The ITB team consists of Dr. Suprijanto, Dr. Vebi Nadhira and Naila Zahra, ST MT, KK Instrumentation and Control, Faculty of Industrial Technology, ITB, and Tri Untoro ST MT, alumni of master degree in Instrumentation and Control, ITB.

“In accordance with ISO / TR 13154: 2017, the recommended body temperature measurement is in the face of the head area. The use of computer vision technology that combines automatic face detection that uses a machine learning from a registered colour image with a thermal scanner image is used to be able to perform non-contact body initial temperature screening at fast and accurate result,” said Dr. Supri.