A method of combining frame accumulation technique and dual temperature calibration for improving temperature measurement accuracy of an infrared thermal imager

Infrared thermal imager has gradually become the basis for diagnosing many diseases. For diagnostic accuracy, we need to measure with high temperature resolution to improve the accuracy of body surface temperature field gradient measurements. In this paper, a combination of frame accumulation and dual temperature calibration is used to improve the temperature accuracy of an infrared imager. The frame accumulation effectively suppresses random noise within the image. The dual temperature calibration method adds temperature calibration for each pixel. This method reduces the influence of spatial distribution on temperature measurement and greatly improves the accuracy of temperature difference measurement between different points. The experiments show that the division value of temperature measurement can be improved to 0.01 °C. Moreover, the temperature measurement accuracy was improved by 3.85 times compared to the accuracy of only single calibration. In the edema case experiment, the dual calibration effectively increased the temperature difference between the edematous and surrounding areas from 1.37 °C to 1.49 °C. Therefore, this method can support the clinical application of infrared imaging for body surface temperature measurement.