Motion-Resistant Remote Imaging Photoplethysmography Based on the Optical Properties of Skin

Abstract

Remote imaging photoplethysmography (RIPPG) can achieve contactless monitoring of human vital signs. However, the robustness to a subject's motion is a challenging problem for RIPPG, especially in facial video-based RIPPG. The RIPPG signal originates from the radiant intensity variation of human skin with pulses of blood and motions can modulate the radiant intensity of the skin. Based on the optical properties of human skin, we build an optical RIPPG signal model in which the origins of the RIPPG signal and motion artifacts can be clearly described. The region of interest (ROI) of the skin is regarded as a Lambertian radiator and the effect of ROI tracking is analyzed from the perspective of radiometry. By considering a digital color camera as a simple spectrometer, we propose an adaptive color difference operation between the green and red channels to reduce motion artifacts. Based on the spectral characteristics of photoplethysmography signals, we propose an adaptive bandpass filter to remove residual motion artifacts of RIPPG. We also combine ROI selection on the subject's cheeks with speeded-up robust features points tracking to improve the RIPPG signal quality. Experimental results show that the proposed RIPPG can obtain greatly improved performance in accessing heart rates in moving subjects, compared with the state-of-the-art facial videobased RIPPG methods.

Using imaging photoplethysmography for heart rate estimation in non-human primates

“…In this study, a cascade classifier using the Viola-Jones algorithm [70] allowed detection of monkey faces in recorded video. Furthermore, techniques for automatic selection of regions providing most pulsatile information in humans [38, 46, 71, 72] can be adopted. This might improve quality of iPPG extraction in NHPs, since application of adaptive model-based techniques to ROI refinement results in better iPPG quality and more accurate pulse rate estimation in humans [46, 72].…”

Non-contact, synchronous dynamic measurement of respiratory rate and heart rate based on dual sensitive regions

“…Wang et al, pointed out limitation of VJ detector, introduced a “tracking-by-detection” with kernels method which is superior than other tracking algorithms and a skin/nonskin pixel classification method for achieving high SNR pulse signals [24, 25]. Feng et al selected two golden ROIs on the region of cheek that has a higher SNR for assessment instead of the whole face, utilizing a speeded-up robust features (SURF) detector [26]. Emrah Tasli mentioned shortages in traditional tracking algorithms, and proposed a facial landmark localization method to track golden ROIs for obtaining robust signals [27].…”

A Real-Time Contactless Pulse Rate and Motion Status Monitoring System Based on Complexion Tracking

“…In [14], the author adopted the signal registered in the red color channel to compensate for motion artifacts in the captured rPPG signal. Another effective method was proposed named Blood-volume pulse vector (PBV) in [27], indicating that the rPPG signal performed a unique signature (specific vector) in the normalized RGB color space.…”

“…Once the ROIs are determined, the rPPG signal is acquired by computing the spatial mean of the proper color channel signal (green color channel for bright mode; red color channel for dark mode measurement) within the ROIs as follows (3) and (4) [14]. ${\mathrm{S}}_{\mathrm{Right}/\mathrm{Left}}false[\mathrm{normaln}false]=true\sum _{\mathrm{i}=1}^{30}true\sum _{\mathrm{j}=1}^{40}{\mathrm{S}}_{\mathrm{n}}false({\mathrm{x}}_{\mathrm{i}},{\mathrm{y}}_{\mathrm{j}}false)$ where S Right/Left [n] is the sum of the green/red color signal within right/left ROI, S n (x i ,y j ) is the green/red color signal at location (x i ,y j ), n is the current frame number.…”

“…However, the rPPG signal queried from the variation of blood vessel volume in the skin surface is sensitive to the outside noises produced by changes of illumination, and motion artifacts [11]. To reduce the effects on the rPPG signal, state-of-the-art work is taking place to demonstrate multiple methods to enhance signal quality and to address motion artifacts [12,13,14]. However, current methods still have problems with real-time implementation and show limited improvement in dealing with severe subject movement.…”