Skin multi-parameter detection method and system based on optical signal reflection

Abstract

This invention discloses a method and system for multi-parameter skin evaluation based on optical signal reflection, belonging to the field of optical skin detection technology. The evaluation method includes: simultaneously irradiating the skin to be tested with five wavelength light sources (365nm, 660nm, 940nm, 1050nm, and 1450nm); receiving the mixed reflected light signals through a first photodetector and a second photodetector respectively; and obtaining independent reflected light signals of each wavelength through frequency division multiplexing demodulation; calculating skin oxygen saturation, hydration index, perfusion index, oil content index, and sensitivity index based on the reflected light signals of the corresponding wavelengths; and performing a weighted summation of the five parameters to output a comprehensive multi-parameter skin score. This invention avoids the time deviation introduced by time-division detection by simultaneously irradiating with multiple wavelength light sources and simultaneously acquiring signals, achieving synchronous acquisition and comprehensive evaluation of multiple skin parameters.

Description

Technical Field

[0001] This invention relates to the field of optical skin detection technology, and more specifically, to a method and system for multi-parameter skin detection based on optical signal reflection. Background Technology

[0002] The skin is the largest organ in the human body, and its physiological state is closely related to human health. Accurate and comprehensive assessment of the skin's condition is of great value in skin health management, disease early warning, and evaluation of cosmetic effects.

[0003] Optical detection methods have been widely used in skin detection due to their advantages such as being non-invasive and real-time. Among the existing methods, there are detection methods for single skin indicators, as well as methods that detect multiple indicators separately and output the results independently.

[0004] However, existing methods for obtaining multi-dimensional skin condition information typically require multiple independent testing processes with time intervals between them. Since skin physiological state fluctuates in real time due to factors such as heart rate, blood pressure, and environmental temperature and humidity, test results acquired at different times reflect skin information under different conditions, leading to time bias when these results are combined for evaluation. Furthermore, different test items correspond to different signal acquisition and calculation processes, ultimately only outputting the test values ​​for each item separately, lacking a unified processing step for multi-dimensional information. This makes it difficult for users to obtain intuitive and consistent evaluation conclusions.

[0005] Therefore, there is an urgent need to provide a solution to the above problems. Summary of the Invention

[0006] To overcome the time discrepancies caused by the inability to acquire various parameters synchronously in existing skin detection methods, and the fact that each parameter is output independently, making it impossible to form a unified comprehensive evaluation conclusion, this invention provides a multi-parameter skin detection method and system based on optical signal reflection.

[0007] The technical solution of this invention is as follows:

[0008] A method for detecting multiple skin parameters based on optical signal reflection includes the following steps:

[0009] The skin to be tested is simultaneously irradiated by light sources with wavelengths of 365nm, 660nm, 940nm, 1050nm and 1450nm, and the reflected light signals with wavelengths of 365nm, 660nm and 940nm are received by a first photodetector, and the reflected light signals with wavelengths of 1050nm and 1450nm are received by a second photodetector.

[0010] The reflected light signals received by the first photodetector and the second photodetector are frequency-division multiplexed and demodulated to separate independent reflected light signals of each wavelength.

[0011] Skin oxygen saturation was calculated based on the ratio of 660nm reflected light signal to 940nm reflected light signal.

[0012] The skin hydration index was calculated based on the ratio of 1050nm reflected light signal to 1450nm reflected light signal.

[0013] The skin perfusion index is calculated based on the ratio of the AC component to the DC component of the reflected light signal output by the second photodetector.

[0014] The skin oil index is calculated based on the ratio of the fluorescence signal generated by 365nm ultraviolet light excitation to the reflected light signal of 660nm.

[0015] Apply pressure perturbation of predetermined frequency and amplitude to the skin to be tested, collect independent reflected light signals of all wavelengths before and after the perturbation, and calculate the skin sensitivity index based on the relative change of each wavelength reflected light signal.

[0016] The skin oxygen saturation, skin hydration index, skin perfusion index, skin oil content index, and skin sensitivity index are weighted and summed to output a comprehensive skin multi-parameter score.

[0017] Preferably, the formula for calculating the skin oxygen saturation is as follows;

[0018] ;

[0019] in, The The intensity of the reflected light signal at 660nm, the The intensity of the reflected light signal at 940nm, the , , This is a pre-calibration coefficient specifically for blood oxygen saturation.

[0020] Preferably, the process of calculating the skin hydration index based on the ratio of 1050nm reflected light signal to 1450nm reflected light signal includes:

[0021] The original ratio was calculated based on the 1050nm reflected light signal and the 1450nm reflected light signal.

[0022] The surface temperature of the skin being measured is collected in real time. The temperature deviation is obtained based on the standard reference temperature. The preset temperature compensation calibration parameters are retrieved. The original ratio is corrected based on the temperature deviation to obtain the first intermediate ratio after temperature compensation.

[0023] The actual contact pressure between the detection probe and the skin is collected in real time. The pressure deviation is obtained based on the standard skin contact pressure. The preset pressure compensation calibration parameters are retrieved. The first intermediate ratio is corrected again based on the pressure deviation to obtain the second intermediate ratio after dual compensation of temperature and pressure.

[0024] Based on the ratio of reflected light signals at wavelengths of 660nm and 940nm used to collect blood oxygen saturation as an indicator of skin melanin content, preset skin color compensation calibration parameters are retrieved, and the second intermediate ratio is finally corrected in combination with the melanin content indicator to output the corrected ratio.

[0025] The skin hydration index is calculated based on the correction ratio using a pre-calibrated polynomial specific to the hydration index.

[0026] Preferably, the specific pre-calibration polynomial for the hydration index is:

[0027] ;

[0028] in, The ratio is obtained after temperature compensation, pressure compensation, and skin color compensation. , , This is a pre-calibration coefficient specifically for the hydration index.

[0029] Preferably, the perfusion index is calculated using the following formula:

[0030] ;

[0031] Among them, the The AC component of the reflected signal from the second photodetector, the The DC component of the signal reflected by the second photodetector; the AC component The DC component is extracted by a bandpass filter. Extracted by a low-pass filter.

[0032] Preferably, the process of calculating the skin oil index based on the ratio of the fluorescence signal generated by 365nm ultraviolet light excitation to the reflected light signal of 660nm includes:

[0033] Extracting fluorescence signals from skin sebum produced by 365nm ultraviolet light stimulation. The fluorescence signal The fluorescence emission intensity detected in the 450nm to 500nm band was obtained by separating it using a long-pass filter placed in front of the photodetector.

[0034] Calculate fluorescence signal With 660nm reflected light signal The ratio, substituted into the pre-calibration curve. The skin oil index is obtained by solving the problem. ; wherein, the The nonlinear fitting curve obtained based on multi-gradient sebum real-person sample calibration is used as input with real-time fluorescence ratio as input, and outputs an oil content index with a unified quantization range to offset the offset error of fluorescence signal caused by optical path loss of different devices and human skin color difference.

[0035] Preferably, in the process of calculating the skin sensitivity index, the predetermined frequency of the pressure disturbance is 1 Hz and the predetermined amplitude is 10 kPa; the formula for calculating the sensitivity index is:

[0036] ;

[0037] in, For the first The weighting coefficients corresponding to each wavelength For the first The change in the reflected signal before and after the disturbance of each wavelength. For the first The original reflected signal intensity at each wavelength.

[0038] Preferably, the formula for calculating the comprehensive skin multi-parameter score is as follows:

[0039] ;

[0040] in, These are the weighting coefficients corresponding to blood oxygen saturation, hydration index, perfusion index, oil content index, and sensitivity index, respectively.

[0041] Preferably, the evaluation method further includes an adaptive calibration step, specifically including:

[0042] Stores complete historical measurement data for the current user, as well as baseline skin quality data covering different skin tones and age groups;

[0043] After each round of complete multi-parameter skin testing, the user's historical measurement data is retrieved and compared with the group's skin quality benchmark data. Based on the deviation between the two sets of data, all pre-calibrated coefficients corresponding to the fitting curves of blood oxygen saturation, hydration index, and oil content index are dynamically corrected.

[0044] A second aspect of the present invention also provides a multi-parameter skin assessment system based on optical signal reflection, comprising a skin detection module, a signal processing module, an oil content assessment module, a blood oxygen assessment module, a hydration assessment module, a perfusion index assessment module, a sensitivity assessment module, and a parameter fusion output module, wherein:

[0045] The skin detection module is used to simultaneously irradiate the skin to be tested with light sources of wavelengths of 365nm, 660nm, 940nm, 1050nm and 1450nm, and to receive reflected light signals of wavelengths of 365nm, 660nm and 940nm based on a first photodetector, and to receive reflected light signals of wavelengths of 1050nm and 1450nm based on a second photodetector.

[0046] The signal processing module is used to perform frequency division multiplexing demodulation on the reflected light signals received by the first photodetector and the second photodetector to separate independent reflected light signals of each wavelength.

[0047] The blood oxygen assessment module is used to calculate skin blood oxygen saturation based on the ratio of 660nm reflected light signal to 940nm reflected light signal.

[0048] The hydration assessment module is used to calculate the skin hydration index based on the ratio of 1050nm reflected light signal to 1450nm reflected light signal.

[0049] The perfusion index evaluation module is used to extract the AC and DC components of the output signal of the second photodetector and calculate the skin perfusion index based on the ratio of the AC and DC components.

[0050] The oil content assessment module is used to calculate the skin oil content index based on the ratio of the fluorescence signal generated by 365nm ultraviolet light excitation to the reflected light signal of 660nm.

[0051] The sensitivity assessment module is used to apply pressure perturbation of predetermined frequency and amplitude to the skin to be tested, collect independent reflected light signals of all wavelengths before and after the perturbation, and calculate the skin sensitivity index based on the relative change of each wavelength reflected light signal.

[0052] The parameter fusion output module is used to perform weighted summation of the skin blood oxygen saturation, skin hydration index, skin perfusion index, skin oil content index, and skin sensitivity index, and output a comprehensive score of multiple skin parameters.

[0053] According to the above-described solution, the beneficial effects of this invention are as follows:

[0054] 1. This invention obtains all the optical signals required to calculate multiple parameters at the same time by simultaneously irradiating with multiple wavelength light sources and receiving them synchronously with multiple photodetectors. Each parameter reflects the true state of the skin at the same time, avoiding the time deviation introduced by multiple independent detections or time-sharing switching in the prior art.

[0055] 2. This invention separates the synchronously acquired mixed signal into independent signals of each wavelength through frequency division multiplexing demodulation, calculates multiple skin parameters separately, and then calculates the weighted sum to output a comprehensive score. This achieves unified processing of the entire process from signal acquisition to result output, overcoming the shortcomings of existing technologies that only output isolated parameter values ​​and make it difficult for users to obtain intuitive evaluation conclusions.

[0056] 3. This invention introduces a three-level compensation mechanism of temperature, pressure, and skin color to eliminate interference factors in hydration measurement. It achieves convenient measurement of perfusion index by synchronously extracting AC / DC components through a reflective architecture. Furthermore, it achieves objective quantification of skin sensitivity by using a weighted calculation of the rate of change of multi-wavelength signals combined with a predetermined pressure disturbance. This overcomes the limitations of existing methods, such as insufficient detection accuracy of the above parameters or reliance on specific hardware conditions. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a schematic flowchart of the detection method provided by the present invention. Detailed Implementation

[0059] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0060] It should be noted that when a component is referred to as "fixed," "set," or "connected" to another component, it may be located directly or indirectly on that other component. The terms "upper," "lower," "left," "right," "front," "rear," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or position based on the accompanying drawings, and are for ease of description only, and should not be construed as limiting the technical solution. The terms "first," "second," etc., are used for ease of description only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features. "Many" means two or more, unless otherwise explicitly specified. "Several" means one or more, unless otherwise explicitly specified.

[0061] This invention provides a method for multi-parameter skin detection based on optical signal reflection, see [link to relevant documentation]. Figure 1 This includes the following steps:

[0062] S1. The skin to be tested is simultaneously irradiated by light sources with wavelengths of 365nm, 660nm, 940nm, 1050nm and 1450nm, and the reflected light signals with wavelengths of 365nm, 660nm and 940nm are received by a first photodetector, and the reflected light signals with wavelengths of 1050nm and 1450nm are received by a second photodetector.

[0063] S2. Frequency division multiplexing demodulation is performed on the reflected light signals received by the first photodetector and the second photodetector to separate the independent reflected light signals of each wavelength.

[0064] S3. Calculate skin oxygen saturation based on the ratio of 660nm reflected light signal to 940nm reflected light signal;

[0065] S4. Calculate the skin hydration index based on the ratio of 1050nm reflected light signal to 1450nm reflected light signal;

[0066] S5. Calculate the skin perfusion index based on the ratio of the AC component to the DC component of the reflected light signal output by the second photodetector.

[0067] S6. Calculate the skin oil index based on the ratio of the fluorescence signal generated by 365nm ultraviolet light excitation to the reflected light signal of 660nm.

[0068] S7. Apply pressure perturbation of predetermined frequency and predetermined amplitude to the skin to be tested, collect independent reflected light signals of all wavelengths before and after the perturbation, and calculate the skin sensitivity index based on the relative change of each wavelength reflected light signal.

[0069] S8. The skin blood oxygen saturation, skin hydration index, skin perfusion index, skin oil content index and skin sensitivity index are weighted and summed to output a comprehensive skin multi-parameter score.

[0070] In some embodiments, the execution of step S1 includes: using time-division modulation to drive five light sources (365nm, 660nm, 940nm, 1050nm, and 1450nm) to simultaneously emit modulated light signals to irradiate the skin to be tested; in terms of optical path structure, the front end of the first photodetector is equipped with an optical filter that can pass through the 365nm, 660nm, and 940nm wavelength bands to receive the reflected light signals of the above three wavelength bands after reflection by the skin; the front end of the second photodetector is equipped with a special short-wave infrared filter that only allows light in the 1050nm and 1450nm wavelength bands to pass through, thereby receiving the reflected light signals of the two short-wave infrared wavelength bands; the two photodetectors simultaneously collect the simulated reflected light signals and convert the collected simulated light signals into corresponding electrical signals for transmission.

[0071] In reality, due to factors such as component manufacturing tolerances, operating temperature variations, and long-term aging of the light source, the actual output wavelength of the light source will have a certain deviation. Therefore, the filter configured at the front end of the first photodetector allows light in the 365nm±30nm, 660nm±30nm, and 940nm±30nm bands to pass through, which can accommodate reasonable wavelength deviations during the normal operation of the above three light sources. The filter configured at the front end of the second photodetector allows light in the 1050nm±30nm and 1450nm±40nm bands to pass through, while setting an OD4 depth cutoff for bands below 1000nm to prevent visible light sources from shifting their wavelengths and entering the short-wave infrared acquisition channel, causing signal crosstalk.

[0072] In some embodiments, during step S2, the process includes: five light sources are each frequency-division multiplexed using independent modulation frequencies, so that the multi-band light sources can synchronously irradiate the skin under test without temporal signal overlap. The mixed reflected photoelectric signal acquired by the photodetector contains multi-wavelength superimposed signals. After analog-to-digital conversion of the mixed signal, frequency domain separation and demodulation are performed based on the preset modulation frequencies of each light source, respectively demodulating the independent reflected light signals corresponding to 365nm±30nm, 660nm±30nm, 940nm±30nm, 1050nm±30nm, and 1450nm±40nm.

[0073] The actual emitted wavelength of the light source device has an inherent offset from the nominal wavelength due to manufacturing process tolerances, operating temperature drift, and long-term aging. In this embodiment, during signal demodulation, the filter bandwidth is adaptively matched according to the pre-calibrated wavelength offset range of each light source. While being compatible with the normal wavelength offset of the light source, it effectively filters out cross-band stray light and channel crosstalk signals, ensuring the integrity and independence of the reflected signals of each wavelength and improving the signal stability of multi-parameter synchronous detection.

[0074] In some embodiments, the process of performing step S3 includes: extracting the intensity of the 660nm reflected light signal and the intensity of the 940nm reflected light signal respectively and calculating the ratio between the two signals, and further substituting them into a preset blood oxygen fitting polynomial to obtain the skin blood oxygen saturation parameter.

[0075] The preset blood oxygen fitting polynomial is:

[0076] ;

[0077] in, The The intensity of the reflected light signal at 660nm, the The intensity of the reflected light signal at 940nm, the , , This is a pre-calibration coefficient specifically for blood oxygen saturation.

[0078] In fact, coefficient , , The acquisition method is as follows: multiple groups of test samples covering different skin colors and blood oxygen levels are selected. The optical acquisition device of this application is used to acquire the ratio of reflected light signals in the 660nm and 940nm bands corresponding to each group of samples. At the same time, a standard medical blood oxygen detection device is used to acquire the standard blood oxygen reference value corresponding to each group of samples. The multiple signal ratios and the corresponding standard blood oxygen reference values ​​are imported into a quadratic polynomial fitting model. The model parameters are then solved by regression using the least squares method to finally determine a unique set of coefficients that are suitable for the optical acquisition characteristics of this device. , , The set of coefficients is then stored in the device's local storage module, and in subsequent testing processes, the pre-calibrated coefficients are directly called to complete the quantitative calculation of blood oxygen saturation.

[0079] Furthermore, the process of executing step S4 includes:

[0080] S4.1 Temperature compensation correction;

[0081] S4.2 Contact pressure compensation correction;

[0082] S4.3, Skin tone compensation correction;

[0083] S4.4 Calculation of Skin Hydration Index.

[0084] Specifically, before executing step S4.1, the intensity of the original reflected light signal corresponding to the two wavelengths of 1050nm and 1450nm output after frequency division multiplexing demodulation is extracted, and the ratio of the two signal intensities is calculated. This value is used as the original signal ratio for hydration calculation.

[0085] Furthermore, during step S4.1, the process includes collecting the real-time skin surface temperature under the current detection environment, retrieving the temperature compensation calibration parameters preset before the device left the factory, and correcting the original signal ratio based on the current temperature. The temperature compensation correction formula is as follows:

[0086] ;

[0087] in, This is the first intermediate ratio after temperature compensation. This represents the difference between the measured skin temperature and the factory-calibrated reference temperature. The temperature compensation pre-calibration coefficients are obtained by fitting skin samples with multiple temperature gradients.

[0088] In fact, when the ambient temperature rises, the absorption coefficient of water molecules near the 1450nm band changes, and the dark current of the short-wave infrared detector increases accordingly. Both of these factors cause the ratio of the reflected signals at 1050nm to 1450nm to deviate from the true value, so temperature compensation is required for correction.

[0089] Specifically, during step S4.2, the process of calculating the skin hydration index based on the ratio of the 1050nm reflected light signal to the 1450nm reflected light signal includes: real-time acquisition of the actual contact pressure between the detection probe and the skin; obtaining the pressure deviation based on the standard skin-fitting pressure; retrieving preset pressure compensation calibration parameters; and correcting the first intermediate ratio again based on the pressure deviation to obtain a second intermediate ratio compensated for both temperature and pressure. The correction formula for pressure compensation is:

[0090] ;

[0091] in, This is the second intermediate ratio after dual compensation for temperature and pressure. This represents the difference between the actual probe pressure and the standard calibration pressure. This is the pressure compensation pre-calibration coefficient.

[0092] Furthermore, during step S4.3, the process includes using the ratio of the reflected light signals at wavelengths of 660nm and 940nm, used for collecting blood oxygen saturation, as an indicator of skin melanin content. Preset skin color compensation calibration parameters are retrieved, and the second intermediate ratio is finally corrected based on the melanin content indicator, outputting a corrected ratio. The skin color compensation correction formula is:

[0093] ;

[0094] in, To incorporate the correction ratio after skin color compensation, , is the ratio of the intensity of the reflected light signal at 660nm to 940nm used in the blood oxygen calculation process, which is used to characterize the melanin content of the current measurement area of ​​the skin being tested; These are pre-calibrated coefficients specifically for skin color compensation, among which Used for the ratio Normalization is performed. This is the skin color compensation gain coefficient.

[0095] Specifically, during step S4.4, a skin hydration index is calculated based on a correction ratio using a pre-calibration polynomial specific to the hydration index; the pre-calibration polynomial specific to the hydration index is:

[0096] ;

[0097] in, The ratio is obtained after temperature compensation, pressure compensation, and skin color compensation. , , This is a pre-calibration coefficient specifically for the hydration index.

[0098] In some embodiments, the process of performing step S5 includes:

[0099] S5.1 Signal component separation;

[0100] S5.2 Quantitative calculation of the infusion index.

[0101] Specifically, in the execution of step S5.1, the following steps are included: retrieving the original shortwave infrared reflection signal acquired by the second photodetector and demodulated by frequency division multiplexing, and performing frequency division processing on the original reflection signal through a low-pass filter and a band-pass filter, respectively. Specifically, the low-pass filter filters out high-frequency fluctuation signals and extracts the steady-state component of the signal to obtain the DC component of the reflection signal; the band-pass filter filters out the DC baseline and high-frequency noise and extracts the dynamic fluctuation component caused by skin microcirculation pulsation to obtain the AC component of the reflection signal.

[0102] Further, step S5.2 is executed, based on the extracted AC and DC components, the skin perfusion index is calculated according to a preset perfusion index calculation formula, which is:

[0103] ;

[0104] Among them, the The AC component of the reflected signal from the second photodetector, the The DC component of the signal reflected by the second photodetector; the AC component The DC component is extracted by a bandpass filter. Extracted by a low-pass filter.

[0105] In fact, by using the short-wave infrared band signal corresponding to the second photodetector for calculation, the interference of visible light band from melanin and oil on the skin surface can be avoided. Relying on the ratio of dynamic light intensity changes generated by microcirculation pulsation to the static light intensity of the substrate, the level of peripheral blood perfusion of the skin can be stably and accurately characterized.

[0106] In some embodiments, the process of performing step S6 includes:

[0107] S6.1, Precise extraction of fluorescence signals;

[0108] S6.2 Signal ratio calculation and error cancellation processing;

[0109] S6.3 Solve the skin oil index based on the pre-calibrated curve.

[0110] Specifically, during step S6.1, when the skin to be tested is irradiated with a 365nm ultraviolet light source, it can excite the sebum and oily substances attached to the skin surface to produce specific fluorescence emission. In this embodiment, a long-pass filter is set at the front end of the photodetector to accurately filter fluorescence signals in the 450nm to 500nm wavelength range, filtering out the 365nm ultraviolet excitation light, ambient stray light, and interference light reflected by the skin itself, thus separating the pure oily fluorescence emission intensity signal. The fluorescence emission intensity detected in this wavelength range is defined as the fluorescence signal. .

[0111] Further, step S6.2 is executed to retrieve the intensity of the independently reflected light signal in the 660nm band obtained by frequency division multiplexing demodulation. Calculate fluorescence signal With reflected light signal The real-time fluorescence ratio. Among them, the 660nm visible light reflection signal is not affected by the skin's oil content, and only represents the transmission loss of the current detection optical path, the fluctuation of the light source power, and the difference in human skin color. Using this signal as a normalization benchmark can effectively offset the offset error caused by the optical path loss of different devices and the difference in skin color on the fluorescence detection signal, and solve the problem of poor stability of single fluorescence signal detection.

[0112] Furthermore, step S6.3 is executed, and the calculated real-time fluorescence ratio is substituted into the pre-calibrated nonlinear fitting curve. The skin oil index is obtained by solving the problem. .

[0113] Among them, the nonlinear fitting curve The fitting model is obtained through exclusive calibration before the equipment leaves the factory. The specific calibration method is as follows: Real skin samples with multi-gradient sebum content are selected, covering different sebum secretion levels, skin tones, and textures. Fluorescence ratios and corresponding real sebum contents of the samples are collected in batches under different light source powers and detection environments. Using the measured sebum content as the standard reference value and the fluorescence ratio as the input variable, a nonlinear fitting algorithm is used to complete data modeling and construct a one-to-one pre-calibration curve. This curve can convert the fluctuating real-time fluorescence ratio into a standardized and comparable oil content index, further eliminating detection errors caused by differences in detection conditions and ensuring the accuracy and consistency of oil content index detection.

[0114] In some embodiments, the process of performing step S7 includes:

[0115] S7.1, Static reference signal acquisition;

[0116] S7.2 Controllable pressure disturbance application and dynamic signal acquisition;

[0117] S7.3, Solve for the skin sensitivity index.

[0118] Specifically, during step S7.1, the detection probe is kept in stable contact with the skin to be tested without any additional pressure disturbance. The steady-state independent reflected light signals corresponding to all wavelengths of 365nm, 660nm, 940nm, 1050nm, and 1450nm are obtained through synchronous illumination by five light sources and synchronous acquisition by dual detectors. Each set of steady-state reflected signals is used as the static reference signal when the skin is not disturbed by external forces.

[0119] Further, in step S7.2, a periodic micro-pressure disturbance with a preset fixed frequency and fixed amplitude is applied to the skin to be tested through the built-in pressure driving module of the device. The preset frequency of the pressure disturbance is 1Hz and the preset amplitude is 10kPa to ensure that the external force stimulation conditions are consistent for each test. During the continuous application of the pressure disturbance, the dynamic reflected light signals corresponding to each wavelength are collected synchronously and continuously, and the dynamic reflected signals corresponding to the peak time of the pressure disturbance are extracted.

[0120] Specifically, during step S7.3, the relative change rate of the reflected light signal at each wavelength before and after the pressure disturbance is calculated to obtain the standardized skin sensitivity index. The formula for calculating the sensitivity index is as follows:

[0121] ;

[0122] in, For the first The weighting coefficients corresponding to each wavelength For the first The change in the reflected signal before and after the disturbance of each wavelength. For the first The original reflected signal intensity at each wavelength.

[0123] In some embodiments, the process of performing step S8 includes:

[0124] S8.1, Parameter normalization preprocessing;

[0125] S8.2, Multi-parameter weighted fusion calculation;

[0126] S8.3 Outputs a standardized comprehensive score for multiple skin parameters.

[0127] Specifically, during step S8.1, the skin oxygen saturation, skin hydration index, skin perfusion index, skin oil content index, and skin sensitivity index calculated in the aforementioned steps are obtained respectively. Since the original numerical dimensions and ranges of the five parameters are not consistent, they cannot be directly weighted and calculated. Therefore, the original parameters are normalized to their maximum and minimum values, mapping all parameters to the standard range of 0~1 to eliminate the calculation bias caused by the difference in dimensions.

[0128] Further, in step S8.2, the device's factory-preset weighted coefficients for each parameter are retrieved. A weighted summation algorithm is then used to fuse and calculate the five normalized skin parameters, constructing a multi-parameter comprehensive skin scoring model. The specific calculation formula is as follows:

[0129] ;

[0130] in, These are the weighting coefficients for blood oxygen saturation, hydration index, perfusion index, oil content index, and sensitivity index, respectively, and they satisfy the weight normalization constraint. .

[0131] Specifically, during step S8.3, the weighted fusion calculation results in range scaling, which converts the values ​​into standardized scores in the range of 0 to 100. Finally, a comprehensive multi-parameter skin score that can be intuitively quantified and compared is output, completing an integrated assessment of the overall health status of the skin.

[0132] Among them, the weighting coefficient These are fixed parameters obtained through big data sample training and calibration before the device leaves the factory. The specific calibration method is as follows: collect a large number of skin samples with different health levels and skin conditions, and simultaneously obtain the five skin parameters and the true value of the professional medical aesthetic skin quality comprehensive score for each sample; with the goal of minimizing the error between the true value of the sample parameters and the score calculated by the model, determine the optimal set of weight coefficients through machine learning iterative training and gradient optimization, and store this set of weight coefficients in the device's storage module. It can be directly called during detection to ensure that the comprehensive score result closely matches the real skin health state and has extremely high evaluation accuracy and universality.

[0133] In fact, the evaluation method also includes an adaptive calibration step, specifically including:

[0134] It stores the current user's complete historical measurement data, as well as the group skin quality benchmark data covering different skin colors and age groups; after each round of complete skin multi-parameter detection, it retrieves the user's historical measurement data and compares it with the group skin quality benchmark data to make mean comparisons. Based on the deviation between the two sets of data, it dynamically corrects all the pre-calibrated coefficients corresponding to the fitting curves of blood oxygen saturation, hydration index and oil content index.

[0135] A second aspect of the present invention also provides a multi-parameter skin assessment system based on optical signal reflection, comprising a skin detection module, a signal processing module, an oil content assessment module, a blood oxygen assessment module, a hydration assessment module, a perfusion index assessment module, a sensitivity assessment module, and a parameter fusion output module, wherein:

[0136] The skin detection module is used to simultaneously irradiate the skin to be tested with light sources of wavelengths of 365nm, 660nm, 940nm, 1050nm and 1450nm, and to receive reflected light signals of wavelengths of 365nm, 660nm and 940nm based on a first photodetector, and to receive reflected light signals of wavelengths of 1050nm and 1450nm based on a second photodetector.

[0137] The signal processing module is used to perform frequency division multiplexing demodulation on the reflected light signals received by the first photodetector and the second photodetector to separate independent reflected light signals of each wavelength.

[0138] The blood oxygen assessment module is used to calculate skin blood oxygen saturation based on the ratio of 660nm reflected light signal to 940nm reflected light signal.

[0139] The hydration assessment module is used to calculate the skin hydration index based on the ratio of 1050nm reflected light signal to 1450nm reflected light signal.

[0140] The perfusion index evaluation module is used to extract the AC and DC components of the output signal of the second photodetector and calculate the skin perfusion index based on the ratio of the AC and DC components.

[0141] The oil content assessment module is used to calculate the skin oil content index based on the ratio of the fluorescence signal generated by 365nm ultraviolet light excitation to the reflected light signal of 660nm.

[0142] The sensitivity assessment module is used to apply pressure perturbation of predetermined frequency and amplitude to the skin to be tested, collect independent reflected light signals of all wavelengths before and after the perturbation, and calculate the skin sensitivity index based on the relative change of each wavelength reflected light signal.

[0143] The parameter fusion output module is used to perform weighted summation of the skin blood oxygen saturation, skin hydration index, skin perfusion index, skin oil content index, and skin sensitivity index, and output a comprehensive score of multiple skin parameters.

[0144] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for detecting multiple skin parameters based on optical signal reflection, characterized in that, Includes the following steps: The skin to be tested is simultaneously irradiated by light sources with wavelengths of 365nm, 660nm, 940nm, 1050nm and 1450nm, and the reflected light signals with wavelengths of 365nm, 660nm and 940nm are received by a first photodetector, and the reflected light signals with wavelengths of 1050nm and 1450nm are received by a second photodetector. The reflected light signals received by the first photodetector and the second photodetector are frequency-division multiplexed and demodulated to separate independent reflected light signals of each wavelength. Skin oxygen saturation was calculated based on the ratio of 660nm reflected light signal to 940nm reflected light signal. The skin hydration index was calculated based on the ratio of 1050nm reflected light signal to 1450nm reflected light signal. The skin perfusion index is calculated based on the ratio of the AC component to the DC component of the reflected light signal output by the second photodetector. The skin oil index is calculated based on the ratio of the fluorescence signal generated by 365nm ultraviolet light excitation to the reflected light signal of 660nm. Apply pressure perturbation of predetermined frequency and amplitude to the skin to be tested, collect independent reflected light signals of all wavelengths before and after the perturbation, and calculate the skin sensitivity index based on the relative change of each wavelength reflected light signal. The skin oxygen saturation, skin hydration index, skin perfusion index, skin oil content index, and skin sensitivity index are weighted and summed to output a comprehensive skin multi-parameter score.

2. The method for multi-parameter skin detection based on optical signal reflection as described in claim 1, characterized in that, The formula for calculating skin oxygen saturation is as follows; ； in, The The intensity of the reflected light signal at 660nm, the The intensity of the reflected light signal at 940nm, the , , This is a pre-calibration coefficient specifically for blood oxygen saturation.

3. The method for multi-parameter skin detection based on optical signal reflection as described in claim 1, characterized in that, The process of calculating the skin hydration index based on the ratio of 1050nm reflected light signal to 1450nm reflected light signal includes: The original ratio was calculated based on the 1050nm reflected light signal and the 1450nm reflected light signal. The surface temperature of the skin being measured is collected in real time. The temperature deviation is obtained based on the standard reference temperature. The preset temperature compensation calibration parameters are retrieved. The original ratio is corrected based on the temperature deviation to obtain the first intermediate ratio after temperature compensation. The actual contact pressure between the detection probe and the skin is collected in real time. The pressure deviation is obtained based on the standard skin contact pressure. The preset pressure compensation calibration parameters are retrieved. The first intermediate ratio is corrected again based on the pressure deviation to obtain the second intermediate ratio after dual compensation of temperature and pressure. Based on the ratio of reflected light signals at wavelengths of 660nm and 940nm used to collect blood oxygen saturation as an indicator of skin melanin content, preset skin color compensation calibration parameters are retrieved, and the second intermediate ratio is finally corrected in combination with the melanin content indicator to output the corrected ratio. The skin hydration index is calculated based on the correction ratio using a pre-calibrated polynomial specific to the hydration index.

4. The method for multi-parameter skin detection based on optical signal reflection as described in claim 3, characterized in that, The specific pre-calibration polynomial for the hydration index is: ； in, The ratio is obtained after temperature compensation, pressure compensation, and skin color compensation. , , This is a pre-calibration coefficient specifically for the hydration index.

5. The method for detecting multiple skin parameters based on optical signal reflection as described in claim 1, characterized in that, The formula for calculating the infusion index is as follows: ； Among them, the The AC component of the reflected signal from the second photodetector, the The DC component of the signal reflected by the second photodetector; the AC component The DC component is extracted by a bandpass filter. Extracted by a low-pass filter.

6. The method for multi-parameter skin detection based on optical signal reflection as described in claim 1, characterized in that, The process of calculating the skin oil index based on the ratio of the fluorescence signal generated by 365nm ultraviolet light excitation to the reflected light signal of 660nm includes: Extracting fluorescence signals from skin sebum produced by 365nm ultraviolet light stimulation. The fluorescence signal The fluorescence emission intensity detected in the 450nm to 500nm band was obtained by separating it using a long-pass filter placed in front of the photodetector. Calculate fluorescence signal With 660nm reflected light signal The ratio, substituted into the pre-calibration curve. The skin oil index is obtained by solving the problem. ; wherein, the The nonlinear fitting curve obtained based on multi-gradient sebum real-person sample calibration is used as input with real-time fluorescence ratio as input, and outputs an oil content index with a unified quantization range to offset the offset error of fluorescence signal caused by optical path loss of different devices and human skin color difference.

7. The method for multi-parameter skin detection based on optical signal reflection as described in claim 1, characterized in that, In calculating the skin sensitivity index, the predetermined frequency of the pressure disturbance is 1 Hz and the predetermined amplitude is 10 kPa; the formula for calculating the sensitivity index is: ； in, For the first The weighting coefficients corresponding to each wavelength For the first The change in the reflected signal before and after the disturbance of each wavelength. For the first The original reflected signal intensity at each wavelength.

8. The method for multi-parameter skin detection based on optical signal reflection as described in claim 1, characterized in that, The formula for calculating the comprehensive skin score using multiple parameters is as follows: ； in, These are the weighting coefficients corresponding to blood oxygen saturation, hydration index, perfusion index, oil content index, and sensitivity index, respectively.

9. The method for detecting multiple skin parameters based on optical signal reflection as described in claim 1, characterized in that, The evaluation method also includes an adaptive calibration step, specifically including: Stores complete historical measurement data for the current user, as well as baseline skin quality data covering different skin tones and age groups; After each round of complete multi-parameter skin testing, the user's historical measurement data is retrieved and compared with the group's skin quality benchmark data. Based on the deviation between the two sets of data, all pre-calibrated coefficients corresponding to the fitting curves of blood oxygen saturation, hydration index, and oil content index are dynamically corrected.

10. A multi-parameter skin assessment system based on optical signal reflection, characterized in that, It includes a skin detection module, a signal processing module, an oil content assessment module, a blood oxygen assessment module, a hydration assessment module, a perfusion index assessment module, a sensitivity assessment module, and a parameter fusion output module, among which: The skin detection module is used to simultaneously irradiate the skin to be tested with light sources of wavelengths of 365nm, 660nm, 940nm, 1050nm and 1450nm, and to receive reflected light signals of wavelengths of 365nm, 660nm and 940nm based on a first photodetector, and to receive reflected light signals of wavelengths of 1050nm and 1450nm based on a second photodetector. The signal processing module is used to perform frequency division multiplexing demodulation on the reflected light signals received by the first photodetector and the second photodetector to separate independent reflected light signals of each wavelength. The blood oxygen assessment module is used to calculate skin blood oxygen saturation based on the ratio of 660nm reflected light signal to 940nm reflected light signal. The hydration assessment module is used to calculate the skin hydration index based on the ratio of 1050nm reflected light signal to 1450nm reflected light signal. The perfusion index evaluation module is used to extract the AC and DC components of the output signal of the second photodetector and calculate the skin perfusion index based on the ratio of the AC and DC components. The oil content assessment module is used to calculate the skin oil content index based on the ratio of the fluorescence signal generated by 365nm ultraviolet light excitation to the reflected light signal of 660nm. The sensitivity assessment module is used to apply pressure perturbation of predetermined frequency and amplitude to the skin to be tested, collect independent reflected light signals of all wavelengths before and after the perturbation, and calculate the skin sensitivity index based on the relative change of each wavelength reflected light signal. The parameter fusion output module is used to perform weighted summation of the skin blood oxygen saturation, skin hydration index, skin perfusion index, skin oil content index, and skin sensitivity index, and output a comprehensive score of multiple skin parameters.

Images