Optical Characteristics Measurement Of Human Skin Using Video Imaging And Signal-Weighted Method

Noncontact measurement of optical characteristics on human skin is very important to investigate human vital physical signals because it has a close relation with the physiological state.In the paper,we proposed a novel noncontact methodology that optical characterizes of human skin was discussed correctly through measuring and analyzing the color variance for video recordings using common camera and signal-weighted method.The approach was applied to noncontact measurement of human physiological parameters such as heart rate,and on its base,we investigated the optical specificity of human skin and explored the optical characteristic of the skin around an acupuncture point to locate the acupoints correctly.This study will provide a simple,safe and noncontact method for the measurement of human physiological parameters,and introduce a novel scientific research and engineering method for the investigation of the meridians and acupuncture pointsFirstly,we introduced and described the development of the optical measurement technique of human skin.Combined biological tissue optics,we analyzed the mechanics of color changes for human complexion,and obtained the conclusion that the visible light is the suitable for measurement of human physical state.For the measurement requirement of human vital physical signals,we presented the experimental conditions,and analyzed the influence on the measurement results.Then,we built a CCD camera-based vision measurement system for human skin successfully and applied it to capturing video recordings of the right fore-arm involving the acupoints after we discussed the optimal illumination conditions for the measurement.Secondly,for all experiments,color videos were stabilized and enhanced respectively using image registration algorithm and homomorphic filtering algorithm to reduce the effect of uncontrolled motion on the measurement results.On the base,we measured the physiological parameters such as heart rate correctly by spatially averaging three RGB channels,yielding the raw traces,blind source separation and time-frequency transform for the videos.The results indicate that the physiological parameters and the optical characteristics of human skin around acupuncture points can be measured through the analysis of the skin's optical characteristic.Finally,the measurement approach was studied systematically to extract the acupoint information from color videos,and obtained human vital physical signals using a signal-weighted method.Then,we analyzed the optical characteristics of the human skin through the analysis of the physiological measurement signal using kurtosis,and found the skin around acupuncture points.Its characteristics were also explored using a based-AR bi-spectrum analysis method.The results show that the specification within the acupooint region can be observed under the visible light.The main innovations are asfollows:(1)A novel method was proposed to determine the location of acupoints using optical images on the surface of human skin,and was applied to accomplish the localization of the acupoints.The results indicate that it will provide a novel way to further study acupuncture points quantitatively.(2)An approach was presented systematically for the extraction of vital physical signals from video recordings through the separation of the RGB three channel signals and the analysis of blind sources.The results show that heart rate can be measured succesffully.(3)An approach was presented for the extraction of the signals relative with the acupoint from video recordings through the signal weighting analysis,and the coincidence between the areas with the specification of optical characteristics and the acupoints was found preliminarily.