The Advancement of Thermal Imaging Cameras for Human Body Temperature Measurement During 2019-nCoV

The Advancement of Thermal Imaging Cameras for Human Body Temperature Measurement During 2019-nCoV

   The prevalence of new crown virus pneumonia in early 2020 has put forward higher requirements for timely screening of suspected infections. In scenes such as airports, stations, docks, hospitals, shopping malls, etc., where there is a large flow of people, thermal imaging cameras have played a huge role. Infrared thermal imaging cameras are fast, safe and effective in measuring human body temperature. The non-contact human body temperature measurement method largely avoids the probability of contact infection. The human body is a source of infrared radiation, and the skin has a high emissivity. By detecting and collecting the wavelength reflected by the skin, the skin surface temperature can be calculated.

   The normal human body temperature is 36°C to 37°C, and the temperature of fever patients is generally higher than 37.3°C. It can be seen that the minimum temperature difference between low-grade fever patients and normal people is 0.3°C, which requires the human body temperature measurement accuracy of infrared thermal imaging cameras to be less than 0.3°C. Generally, infrared thermal imaging cameras purchased in public places are small devices, but have higher requirements for temperature, which poses a greater technical challenge. At present, algorithms compensation, image processing optimization and other technologies are used, so that the temperature measurement accuracy of some infrared thermal imaging cameras can reach ±0.3℃, which can basically meet the requirements of preliminary screening of fever patients.

Figure 1. Infrared thermal imaging camera is measuring human body temperature.

   The greater the distance between the infrared thermal imager and the shooting object, the more the radiation energy attenuates from the source to the camera's transmission path, which has a greater impact on the temperature detection results. Therefore, the best temperature measurement distance of most infrared cameras at home and abroad is 1.5~3.5 meters. This is the result obtained after combining multiple sub-module parameters such as lens, infrared detector, image processing, etc., which can ensure the accuracy of the test to the greatest extent.

   Infrared electromagnetic waves propagating in the air will be absorbed by dust, carbon dioxide, etc., causing energy attenuation, resulting in temperature measurement errors. Therefore, infrared thermal imaging cameras are susceptible to environmental interference in temperature measurement accuracy. Some domestic manufacturers have introduced blackbody technology to optimize such problems. Blackbody is generally used for calibration and verification of radiation thermometers to thermal imaging, etc. The blackbody calibration temperature is compared with the blackbody temperature detected by thermal imaging, so as to calibrate the temperature measurement results, thereby reducing the environmental impact on thermal imaging The interference of human body temperature measurement accuracy improves the temperature measurement accuracy. At present, most domestic infrared thermal imaging cameras that use blackbody for calibration can have a temperature measurement accuracy of less than 0.3°C in a complex environment, which can meet the primary screening requirements for epidemic prevention.

   Although infrared thermal imaging technology can measure and display human body temperature, it cannot effectively identify people. During the epidemic, people wear masks, and the traditional facial recognition rate has decreased. However, after deep optimization of the face recognition algorithm of the visible light camera, the public can complete face recognition without removing the mask. This provides access detection for people in airports, stations, terminals, hospitals, schools, large shopping malls and other areas. This has greatly facilitated and avoided the spread of the epidemic caused by the dense flow of people. At the same time, relying on big data technology, the infrared thermal imager's human body temperature measurement and the visible light camera's face recognition function are effectively combined, which can assist in tracking the large number, difficult, and costly suspected objects in the path of disease control infection, and further confirm their identity.