Dynamic Offset Correction for Smartphone Thermal Cameras Using a Wristband Sensor

Abstract

Thermal images are widely used to various healthcare applications. However, thermal images captured by smart-phone thermal cameras have insufficient accuracy to monitor human body temperature. In this paper, we propose an offset correction method for thermal images captured by smartphone thermal cameras. We fully utilize the characteristic which is specific to thermal cameras: the relative temperatures in a single thermal image are highly reliable although the absolute temperatures fluctuate frequently. Our method combines thermal images with a reliable absolute temperature obtained by a wristband sensor based on the above characteristic. The evaluation result shows that the mean absolute error and the standard deviation of face temperature decrease by 26.2% and 70.1%, respectively, highlighting the effectiveness of the proposed method.