AI Infrared Thermal Imaging

With the alleviation of travel restrictions, how to strictly prevent the import of overseas COVID-19 cases have become the most urgent issue, and the prevention and control of airport epidemic situations has entered a critical period.

When a passenger passes through the infrared temperature measuring device, the system will automatically detect the passenger’s body temperature and feed it back to the system.

If the passenger’s body temperature exceeds 37.3°C, the system will issue an alarm, and the staff will use a medical grade thermometer to conduct a second inspection, record the passenger information, and disinfect the area where the fever patient passes. 

Typically an airport needs to be equipped with several sets of AI infrared thermal imaging and temperature measurement equipment to measure the body temperature of all passengers coming and all personnel entering and leaving the terminal.

Consequently, these temperature measurement devices have been participating in the pandemic prevention campaign. Infrared thermometer is actually a not a new product. During the outbreak of SARS in 2003, airports and stations across the country began to use infrared temperature measuring instruments..

The principle of the infrared temperature measuring instrument is very simple. In nature, objects greater than absolute zero will radiate infrared rays into the environment. The magnitude of this infrared energy is positively related to the temperature of the object. So as long as the instrument detects infrared rays and converts them into electrical signals of different sizes, the temperature of the object can be determined.

The infrared thermometer solves queuing problem. In addition it has the functions of long-distance, large area, high efficiency, and non contact temperature measurement. Compared with the traditional temperature measurement method, it is more suitable for crowded places such as airports, subways, hospitals, and high-speed railway stations.

However, infrared thermometers have their natural limitations, and relevant industry insiders said: “After a lot of testing and verification, I think infrared as a body temperature diagnostic screening still has some problems, it is not accurate enough and is subject to The environment is very disturbing. However, at this special moment, it is also a good thing to have a means to help the functional department to quickly screen people with abnormal temperature.”

Infrared thermometers can only recognize the temperature in the environment, but they cannot distinguish the type of object. For example, in a train station, if a passenger with normal temperature carries 40°C hot water, the infrared thermometer may also sound an alarm.

Therefore, in order to solve the problems of inaccurate recognition and false alarms of infrared thermometers, AI infrared thermal imaging temperature measurement technology has been pushed to the foreground.

AI infrared thermal imaging temperature measurement, as the name implies, is the combination of infrared thermal imaging and computer technology: infrared thermal imaging technology can measure non-contact temperature from a distance, and computer vision can quickly locate the person’s forehead in the environment.

From the technical logic, it needs to use machine learning algorithm for face recognition and tracking, and then combined with infrared thermometer to calculate the temperature of the person.

The face recognition algorithm is very developed, but the pandemic brings a new question: how to accurately locate face recognition after wearing a mask?

Since the outbreak of the pandemic, AI companies have rapidly launched the research and development of AI infrared thermal imaging temperature measurement, quickly completed the monitoring model of wearing masks, and launched various solutions. 

From the perspective of popular landing solutions in the market, the industry mostly uses infrared/visible light dual sensors, combined with infrared thermal imaging and human body recognition + face recognition dual recognition, and the temperature measurement accuracy is mostly within ±0.3℃.

AI infrared thermal imaging temperature measurement technology combines face recognition with infrared thermal imaging technology, which can reduce the false alarm rate of the instrument, improve the detection efficiency, and reduce the risk of staff infection.

First of all, in any complex environment, the accuracy of temperature measurement is high; Secondly, non-contact, non-inductive state, no presets. Thirdly, the detection is rapid, and multiple people can be detected at the same time. Fourth, the ease of use is high, and the staff can quickly get started and use it without any training or only simple training.

We know that objects above absolute zero will emit infrared light, but the emissivity of different objects is different, and the emissivity is determined by the temperature, wavelength, material, and surface environment of the object. This one is not accurate. The accuracy will decrease.

Due to the shortage of infrared temperature measurement components, and in order to ensure the accuracy of infrared thermal imaging, companies in the industry have adopted the black body scheme to ensure the accuracy of the AI ​​infrared thermal imaging temperature measurement system through real-time black body temperature correction, But this implementation is costly.

According to the prediction of relevant industry researchers: under the influence of the pandemic, the demand for AI infrared thermal imaging temperature measurement market has surged recently, and the pandemic will completely open the market.

Of course, this demand can only be concentrated in specific scenarios such as stations, airports, schools, and hospitals, and the overall market is relatively niche.

In brief, the combination of AI and industry is an inevitable result of the industrial structure upgrade. In the future, the combination of AI and industry will also produce many forms and create more possibilities, including intelligent office systems, intelligent production systems, and intelligent Service system, etc.

[launchpad_feedback]

Scroll to Top