Non-Contact Face Temperature Measurement by Thermopile-Based Data Fusion

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Sensor Technology

Background:

• Current non-contact skin temperature measurement methods like thermal imaging cameras and infrared (IR) thermometers have limitations.
• Thermal imaging is often cost-prohibitive for widespread use.
• IR thermometers require manual alignment, impacting efficiency and accuracy.

Purpose of the Study:

• To develop an automated, non-contact, and area-specific method for facial skin surface temperature measurement.
• To overcome the limitations of existing non-contact temperature measurement techniques.
• To enable precise temperature monitoring of specific facial regions.

Main Methods:

• Integration of a low-cost thermopile sensor matrix with a 2D image sensor.
• Fusion of temperature and image data using parametric affine transformation.
• Implementation of face recognition for assigning temperature values to specific facial regions.

Main Results:

• The proposed method enables automated, non-contact, and area-specific skin temperature determination.
• Temperature values are accurately assigned to selected facial regions via face recognition.
• Participant studies show comparable absolute temperature values to commercial IR forehead thermometers.

Conclusions:

• The developed system offers an advantageous alternative for non-contact facial skin temperature measurement.
• The method provides accurate and automated temperature readings without manual alignment.
• This technology has potential for various applications requiring precise facial thermography.