ThermalWrist: Smartphone Thermal Camera Correction Using a Wristband Sensor †

Abstract

Thermal images are widely used for various healthcare applications and advanced research. However, thermal images captured by smartphone thermal cameras are not accurate for monitoring human body temperature due to the small body that is vulnerable to temperature change. In this paper, we propose ThermalWrist, a dynamic offset correction method for thermal images captured by smartphone thermal cameras. We fully utilize the characteristic that is specific to thermal cameras: the relative temperatures in a single thermal image are highly reliable, although the absolute temperatures fluctuate frequently. To correct the offset error, ThermalWrist combines thermal images with a reliable absolute temperature obtained by a wristband sensor based on the above characteristic. The evaluation results in an indoor air-conditioned environment shows that the mean absolute error and the standard deviation of face temperature measurement error decrease by 49.4% and 64.9%, respectively. In addition, Pearson’s correlation coefficient increases by 112%, highlighting the effectiveness of ThermalWrist. We also investigate the limitation with respect to the ambient temperature where ThermalWrist works effectively. The result shows ThermalWrist works well in the normal office environment, which is 22.91 °C and above.

Non-invasive (non-contact) measurements of human thermal physiology signals and thermal comfort/discomfort poses -A review

“…Infrared cameras are usually high cost and big size. Oliveira et al, 2007Bouzida et al, 2009Burzo et al, 2014Ranjan and Scott;2016Burzo et al, 2017Pavlin et al, 2017Metzmacher et al, 2018Yoshikawa et al, 2019 Non-invasive measurements (crossvalidation of infrared camera, RGB camera and wearable devices)…”

“…A wristband was used to record photoplethysmogram (PPG) signals, from which Inter-beat interval (IBI) signals were extracted and sent to a smart phone for pulse rate variability (PRV) calculation and real time thermal comfort prediction (Nkurikiyeyezu and Lopez, 2018). The wristband was also used to dynamically correct offset errors for thermal images captured by smartphone thermal cameras (Yoshikawa et al, 2019). Measuring results of skin temperature and heart rate (HR)/heart rate variability (HRV) from smartwatch were used to develop thermal sensation estimation models .…”

“…Compared to high-end models, the accuracy of low cost thermal camera is insufficient because of uncooled infrared detectors. A dynamic offset correction method was proposed (Yoshikawa et al, 2019). Infrared camera technology was also compared with traditional invasive measurements of ambient air temperature and semi invasive measurements of wrist-type wearable devices .…”

TSVNet: Combining Time-Series and Opportunistic Sensing by Transfer Learning for Dynamic Thermal Sensation Estimation

“…For masking without the shift, it is necessary to map each pixel in the visible image to one of the pixels in the thermal image. Therefore, we apply an overlay method [44] based on the edges that typically appear in both images. Using the above processes, we accurately extract the skin temperature distribution around the face.…”

Estimating Departure Time Using Thermal Camera and Heat Traces Tracking Technique

“…To obtain the relationship between actual temperature and time, we need to further study function h i (k|θ). Except for some special processing methods [31], the conventional method in most papers is to assume that the noise obey Gaussian distribution [32][33][34][35]. We assume noise ε i,k obeys Gaussian distribution.…”