Research on Non-contact Multi-person Heart Rate Measurement Method for Intelligent Education

Lian, Yueqi; Ryu, JongSong; Wang, Suqiu; Liang, Shili; Kyongil, Oh

doi:10.1109/ispds56360.2022.9874225

Cited by 1 publication

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the analysis the distribution of facial blood vessels and the SNR of the sROIs, we utilized adaptive weights for the sROIs to conduct spatial averaging, resulting in enhanced raw signal quality during the preprocessing step. The comparison between previous spatial averaging methods based on weights, 19,29 spatial averaging methods based on single ROI selection, and the method proposed in this paper demonstrated the superior performance of our method in enhancing the quality of the raw signal, leading to certain improvement in subsequent HR measurement. In Ref.…”

Section: Discussionmentioning

confidence: 81%

“…2.2 for obtaining the coarse raw signal. Furthermore, in terms of the quality of the pulse signal, the proposed method is compared with the four methods: the first uses the whole face as the ROI (WH); the second uses the facial skin region as the ROI; the third obtains according to a goodness metric (GM); 19 and the fourth obtains by adopting the FW 29 . When the whole face is used to obtain the pulse signal, the region whose width of the detected rectangular face region is reduced by 0.6 times is set as the ROI.…”

Section: Resultsmentioning

confidence: 99%

“…Furthermore, in terms of the quality of the pulse signal, the proposed method is compared with the four methods: the first uses the whole face as the ROI (WH); the second uses the facial skin region as the ROI; the third obtains according to a goodness metric (GM); 19 and the fourth obtains by adopting the FW. 29 When the whole face is used to obtain the pulse signal, the region whose width of the detected rectangular face region is reduced by 0.6 times is set as the ROI. When the facial skin region is used to obtain the raw signal, in the region whose width of the detected rectangular face region is reduced by 0.6 times, the skin region detected by human skin color clustering technology 33 is set as the ROI.…”

Section: Resultsmentioning

confidence: 99%

“…The studies using a face region as the ROI can be divided into two categories: one is using the rectangular region surrounding the whole face or a predefined percentage of it as the ROI; 13 – 17 the other is using a certain region of the face as the ROI. 18 – 29 The comparative analysis of the pulse signals obtained from the forearm (dorsal), forearm (ventral), forehead, palm, hand (dorsal), cheek, nose, and whole face was performed, and the results showed that the forehead and cheek can provide the high-quality pulse signal. 18 – 20 On the other hand, some authors use facial skin region as ROI 21 – 27 or adaptively select ROI providing high-quality pulse signal.…”

Section: Introductionmentioning

confidence: 99%

“…In our previous paper, 29 the facial ROI was divided into seven sub-regions of interest (sROIs) considering the distribution of blood vessels, skin thickness and skin surface temperature, and suitable fixed weights (FWs) were determined for each sROI through experiments conducted on a database. Among these FWs, the two sROIs with the lowest signal-to-noise ratio (SNR) were assigned values less than or equal to 0, whereas the remaining sROIs were assigned positive values.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Spatial averaging method based on adaptive weight for imaging photoplethysmography

Ryu,

Liang

et al. 2023

J. Biomed. Opt.

Self Cite

View full text Add to dashboard Cite

. Significance Imaging photoplethysmography (iPPG) is a non-contact measuring technology for several physiological parameters reflecting personal health status without a special sensor. However, the pulse signal obtained using the iPPG usually is contaminated by various noises, and the intensity of the interesting pulse signal is relatively weak compared to the noises, emphasizing the necessity of obtaining high-quality pulse signals to measure physiological parameters correctly. Aim Various regions of the face harbor distinct pulse information. We propose a spatial averaging method based on adaptive weights, which can obtain high-quality pulse signals by applying different weights to facial sub-regions of interest (sub-ROIs; sROIs). Approach First, the facial ROI is divided into seven sROIs and the coarse heart rate (HR) is calculated from them. Next, the signal-to-noise ratio (SNR) of the raw signal obtained from each sROI is calculated using the coarse HR, and then a high-quality pulse signal is obtained by assigning positive or negative weights to each sROI based on the SNRs. Results We compare our method with others through the quality analysis of the obtained pulse signals using the self-collected database and the public database PURE. The comparison results show that the proposed method can provide a better pulse signal compared to other methods under various resolutions and states. Conclusions This proposed method can obtain the pulse signal with better quality, which is helpful to accurately measure physiological parameters, such as HR and HR variability.

show abstract

Section: Discussionmentioning

confidence: 81%

Section: Resultsmentioning

confidence: 99%