Personalized skin care product formula generation method and system based on skin state analysis

Abstract

The invention discloses a personalized skin care product formula generation method and system based on skin state analysis, and relates to the technical field of skin care product blending methods, and the method comprises the steps: obtaining raw material formula data corresponding to each skin parameter or combination thereof, generating an association relationship between each skin parameter or combination thereof and each formula raw material, and storing the association relationship as a standard reference library; acquiring various skin parameters of the current user and generating standard formula data based on the standard reference library; preparing a skin care product sample for verification based on the standard formula data, detecting skin parameters of a current user in different environments before and after using the sample, and performing fitting to generate each skin parameter and a first association relationship between the change trend of each skin parameter and the environmental parameters; the environment parameters and the target skin parameters of the environment where the current user is located in the set time period in the future are obtained, the standard formula data are combined, the raw material ratio of the standard formula data is adjusted based on the first incidence relation, personalized skin care product formula data are generated, and the skin care effect of the personalized skin care product is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of skin care product formulation methods, and particularly relates to a personalized skin care product formula generation method and system based on skin state analysis. BACKGROUND

[0002] There are various skin care products on the market, but because the skin quality of each user and the daily environment they are in are not the same, the same skin care product used on different user skin often produces different use effects.

[0003] How to accurately customize personalized skin care products according to the skin parameters of users is a problem that needs to be solved for the current development of skin care products towards high-end. SUMMARY

[0004] In view of the problem that skin care products in the prior art cannot meet the personalized skin care needs of different users, the purpose of the present application is to provide a personalized skin care product formula generation method based on skin state analysis, which accurately generates a personalized skin care product formula by obtaining skin parameter data of a user and the law of changes in skin parameters with changes in the environment, in combination with environmental parameter data of the user during use of the skin care product, so that the skin care product produced can be more suitable for the current skin care needs of the user and improve the skin care effect; to implement the above-mentioned personalized skin care product formula generation method, the purpose of the present application is to propose a personalized skin care product formula generation system based on skin state analysis, and the purpose of the present application is to propose a computer readable storage medium to facilitate the popularization and use of the above-mentioned method, and the specific scheme is as follows: A personalized skin care product formula generation method based on skin state analysis, comprising: Obtaining raw material formula data corresponding to each skin parameter or combination thereof, analyzing and generating the correlation between each skin parameter or combination thereof and each formula raw material, and storing it as a standard reference library; Obtaining each skin parameter of the current user and generating corresponding standard formula data based on the above-mentioned standard reference library; Based on the above-mentioned standard formula data, a verification skin care product sample is made, the skin parameters of the current user in different environments before and after using the above-mentioned sample are detected, and a first correlation between each skin parameter and its change trend and the environmental parameter is fitted and generated; Obtaining environmental parameters and target skin parameters of the environment in which the current user is located in a future set period, combining the standard formula data and based on the first correlation, adjusting the raw material ratio in the standard formula data, and generating personalized skin care product formula data; The skin parameters include moisture content data, sebum secretion data, pigment deposition data, barrier function data, elasticity data, pore state data, skin pH value data, and skin temperature data. The environmental parameters include environmental temperature and humidity data, ultraviolet intensity data, air micro-particle concentration data, and set allergen concentration data.

[0005] By the above technical solution, first, the standard formula data is generated according to the current user's skin parameters, combined with the correlation between the skin parameters and the environmental parameters to prepare the skin care product sample, and then the optimization of the standard formula is confirmed by obtaining the trend of the user's skin parameters in different environments, and finally the personalized skin care product formula highly adapted to the user's needs is generated according to the user's itinerary, i.e. the environment the user may be in the future, which can greatly improve the skin care effect of personalized skin care products.

[0006] Further, the method of generating a standard reference library further comprises: dividing the skin into a plurality of different skin types according to different skin parameters or combinations thereof; directly configuring at least one corresponding basic formula data for each skin type based on existing technology; The personalized skin care product formula generation method further comprises: obtaining the current user's skin parameters and determining the user's skin type based on an AI recognition algorithm; According to the skin type determined, the corresponding basic formula is retrieved; According to the user's specified or based on the environmental parameters of the environment the user is in in the future setting period and the target skin parameters, the standard formula data is adjusted, the raw material ratio in the standard formula data is adjusted, and the personalized skin care product formula data is generated; Wherein, the dimensions of skin type include skin quality, sensitivity, wrinkles, pigments and pore feature data.

[0007] By the above technical solution, first, the basic formula data corresponding to each type of skin is directly stored, and then the type of the user's skin quality is accurately determined by the AI recognition algorithm, and then the basic formula suitable for the current user is obtained according to the determination result. After obtaining the basic formula, the skin care product is automatically or manually customized according to the special effect the user needs to achieve, and finally the personalized skin care product formula data is generated.

[0008] Further, the correlation between each skin parameter or its combination and each formula raw material is analyzed and generated as a standard reference library, including: From the skin quality test database, find subjects similar to the current user in age and gender, obtain their corresponding skin parameter data after using different formula skin care products, and use correlation analysis method to obtain the correlation coefficient between each formula raw material and each skin parameter or its combination; extracting skin parameter data and its corresponding raw material formula data whose correlation coefficient is greater than a set threshold value to form a plurality of correlation data sets and storing them as a first standard reference database; judging whether there is trial data of the current user in the skin quality test large database: if there is no trial data of the current user, using the first standard reference database as the standard reference library; if there is trial data of the current user, extracting skin parameter data of the current user and its corresponding raw material formula data and temporarily storing them in association; judging the data amount of the skin parameter data of the current user and its corresponding raw material formula data: if it exceeds a set value, determining the correlation between each skin parameter or its combination of the current user and each raw material formula by using a correlation analysis method to generate a plurality of sets of correlation relationships as a first modified reference; if it does not exceed a set value, using the first standard reference database as a second modified reference; setting a target skin parameter of the current user, calculating the efficacy difference (including difference direction and value) between the already stored skin parameter and the target skin parameter; extracting skin parameters whose efficacy difference exceeds a set value as parameters to be adjusted, obtaining raw material formula associated with the above-mentioned parameters to be adjusted according to the correlation relationship between the skin parameter and the raw material formula data stored in the above-mentioned first modified reference or second modified reference, and determining the adjustment coefficient of at least one set of raw material formula; storing the adjusted raw material formula data in association with the target skin parameter to form the standard reference library.

[0009] Through the above technical solution, based on the trial data in the skin quality test large database and combined with the correlation analysis algorithm, the raw material formula associated with each skin parameter or its combination can be accurately obtained, and the corresponding relationship between the raw material ratio of each skin care product and the skin parameter can be easily determined. The standard reference library generated by the above method is more suitable for the needs of the current user, and subsequent customization of personalized skin care products only needs to be fine-tuned in the standard formula data, which is convenient and fast, and at the same time can protect the skin of the user from being damaged by the skin care sample. By introducing the use effect data of the current user on the specific formula skin care product, the standard reference library can be more accurately customized, so that the skin quality of the user after using the skin care sample can be closer to the expected value, and more accurate personalized skin care product formula data can be generated subsequently.

[0010] Further, the skin parameters of the current user in different environments before and after using the sample are detected, including: using a hyperspectral skin detector to obtain a hyperspectral image of a set skin area of the current user, and generating each skin parameter data by a hyperspectral image analysis algorithm; The detection of the skin parameters is performed in a test environment cabin for providing different environmental parameters at different test periods.

[0011] By the above technical solution, the skin parameters of the current user under various environmental parameter conditions can be collected in a short time, and thus the skin parameters or combinations thereof of the current user and the correlation between the skin parameter change trend and the environmental parameters can be obtained through analysis, facilitating the subsequent generation of more accurate and applicable personalized skin care product raw material formula data.

[0012] Further, the personalized skin care product formula generation method further comprises: obtaining skin parameter data of the current user in different environments before and after using the small sample, and generating a relationship model representing the linkage change relationship between the skin parameters under different environmental parameter conditions through deep neural network training; The linkage change relationship includes a numerical proportional relationship and a time sequence relationship between different skin parameters.

[0013] By the above technical solution, the remaining skin parameter data can be calculated based on one or more skin parameter data by means of the relationship model, that is, only a small amount of data needs to be collected to obtain the required skin parameter data, thereby providing complete data support for the generation of personalized skin care product raw material formula, and the generation time of the personalized skin care product formula data can be greatly shortened.

[0014] Further, the personalized skin care product formula generation method further comprises: based on the above standard formula data, a verification skin care product small sample is prepared, the skin parameters of the current user under different physiological parameter conditions before and after using the small sample are detected, and a second correlation relationship between the skin parameters and their change trends and the physiological parameters is fitted and generated; obtaining the expected physiological parameters and target skin parameters of the current user within a future set period, combining the standard formula data and based on the second correlation relationship, correcting the personalized skin care product formula data; The physiological parameters include skin temperature, heart rate, blood oxygen saturation, sleep quality, and exercise data.

[0015] By the above technical solution, the required personalized skin care product raw material formula can be adjusted according to the physiological parameter data of the current user within the set period, so that the personalized skin care product prepared is more suitable for the current user.

[0016] Further, the personalized skin care product formula generation method further comprises: obtaining the environmental parameter data and physiological parameter data of the current user in each period, and comparing to generate the applicable period of the current personalized skin care product.

[0017] By the above technical solution, in addition to generating a skin care product formula suitable for the current user, the applicable time period of the skin care product can also be generated accordingly, thereby further improving the use effect of personalized skin care products.

[0018] A personalized skin care product formula generation system based on skin state analysis, comprising: A standard reference library generation unit configured to obtain raw material formula data corresponding to each skin parameter or combination thereof, analyze and store the correlation between each skin parameter or combination thereof and each formula raw material; A standard formula generation unit configured to obtain the current user's skin parameters, generate corresponding standard formula data based on the above standard reference library, and store it; A first correlation generation unit configured to obtain the skin parameters of the current user in different environments before and after using a skin care product sample, and fit to generate a first correlation between each skin parameter and its change trend and environmental parameters, the sample being made based on the above standard formula data; A personalized formula generation unit configured to obtain environmental parameters and target skin parameters of the current user in a future set period, combine the standard formula data and generate a personalized skin care product formula data based on the first correlation by adjusting the raw material ratio in the standard formula data; Wherein, the skin parameters include moisture content data, oil secretion data, pigment deposition data, barrier function data, elasticity data, pore state data, skin pH value data and skin temperature data; The environmental parameters include environmental temperature and humidity data, ultraviolet intensity data, air particle concentration data and set allergen concentration data.

[0019] Further, the first correlation generation unit comprises: A skin parameter acquisition subunit connected to the data output end of the hyperspectral skin detector, used to obtain the hyperspectral image of the set skin area of the current user, and generate each skin parameter data through a hyperspectral image analysis algorithm; An environmental parameter acquisition subunit connected to the data output end of the environmental chamber, used to provide different environmental parameters at different test periods; A data fitting subunit configured to obtain the skin parameters of the current user in different environments before and after using a skin care product sample, and fit to generate the first correlation.

[0020] Further, the system further comprises: The second correlation relationship generating unit is configured to obtain skin parameters of the current user under different physiological parameter conditions before and after using the skin care product sample, and fit to generate a second correlation relationship between each skin parameter and its change trend and each physiological parameter; The formula correction data generating unit is configured to obtain expected physiological parameters and target skin parameters of the current user in a future set period, combine the standard formula data, and generate correction parameters for correcting the personalized skin care product formula data based on the second correlation relationship; The applicable period generating unit is configured to obtain environmental parameter data and physiological parameter data of the current user in each period, and compare to generate an applicable period of the current personalized skin care product and output.

[0021] A computer readable storage medium having computer program modules loaded thereon, when the computer program modules are executed by a processor, to implement the personalized skin care product formula generating method based on skin state analysis as described above.

[0022] The present application has at least the following beneficial effects: (1) By introducing the use effect data of the current user on the specific formula skin care product, the standard reference library can be more accurately customized, so that the skin quality of the user after using the skin care product sample can be closer to the expected value, and more accurate personalized skin care product formula data can be generated subsequently; (2) By obtaining the change trend of the skin parameters of the user in different environments to confirm the optimization of the standard formula, a highly personalized skin care product formula adapted to the needs of the user can be generated according to the user's itinerary, i.e. the environment the user may be in the future, which can greatly improve the skin care efficacy of the personalized skin care product. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 The flowchart of the personalized skin care product formula generating method of the present application; Figure 2 The construction method of the standard reference library is shown in the figure; Figure 3 The connection diagram of the functional modules of the personalized skin care product formula generating system of the present application.

[0024] Reference signs: 100, standard reference library generating unit; 200, standard formula generating unit; 300, first correlation relationship generating unit; 301, skin parameter obtaining subunit; 302, environmental parameter obtaining subunit; 303, data fitting subunit; 400, personalized formula generating unit; 500, second correlation relationship generating unit; 501, physiological parameter obtaining subunit; 600, formula correction data generating unit; 700, applicable period generating unit. DETAILED DESCRIPTION

[0025] Embodiments of the present application are described below in detail, examples of which are shown in the accompanying drawings.

[0026] In the description of the present specification, the description of the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in connection with the described embodiments or examples are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Also, the described specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0027] A personalized skin care product formula generation method based on skin state analysis, as shown in Figure 1 The method mainly includes the following steps: S100, obtaining raw material formula data corresponding to each skin parameter or combination thereof, analyzing and generating the correlation between each skin parameter or combination thereof and each formula raw material, and storing it as a standard reference library; S200, obtaining the current user's skin parameters and generating corresponding standard formula data based on the above standard reference library; S300, based on the above standard formula data, a verification skin care sample is made, the skin parameters of the current user in different environments before and after using the sample are detected, and the first correlation between each skin parameter and its change trend and the environmental parameter is fitted and generated; S400, obtaining the environmental parameters of the environment where the current user is in the future set period and the target skin parameters, combining the standard formula data and based on the first correlation, adjusting the raw material ratio in the standard formula data, and generating personalized skin care product formula data.

[0028] In the embodiments of the present application, the skin parameters of the user mainly include moisture content data, sebum secretion data, pigment deposition data, barrier function data, elasticity data, pore state data, skin pH value data and skin temperature data. The above environmental parameters mainly include environmental temperature and humidity data, ultraviolet intensity data, air micro-particle concentration data and set allergen concentration data.

[0029] In step S100, the raw material formula data corresponding to each skin parameter or combination thereof is from a skin test large database. The large database stores the skin data, i.e. skin parameter data, of the subjects of different age stages after using skin care products with different raw material formulas, which reflects the components of the skin care products and their specific effects and functions.

[0030] In one embodiment, in step S100, the correlation between each skin parameter or combination thereof and each formula raw material is analyzed and stored as a standard reference library, which is combined with Figure 2 as shown, specifically including: S110, from the skin test database, find subjects with similar age and gender to the current user, obtain their skin parameter data corresponding to different formula skin care products, and use correlation analysis to obtain the correlation coefficient between each formula raw material and each skin parameter or combination thereof.

[0031] S120, extract skin parameter data and its corresponding raw material formula data with a correlation coefficient greater than a set threshold value, form multiple correlation data groups, and store them as a first standard reference database.

[0032] S130, determine whether the skin test database contains the current user's test data: S131, if the current user's test data does not exist, the first standard reference database is used as the standard reference library; S132, if the current user's test data exists, extract the current user's skin parameter data and its corresponding raw material formula data and temporarily store them in association; S133, determine whether the current user's skin parameter data and its corresponding raw material formula data exceed a set value: S1331, if it exceeds the set value, use correlation analysis to determine the correlation between each skin parameter or combination thereof and each raw material formula of the current user, and generate multiple sets of correlation relationships as the first modified reference; S1332, if it does not exceed the set value, the first standard reference database is used as the second modified reference.

[0033] S140, set the target skin parameter of the current user, calculate the efficacy difference between the already stored skin parameter and the target skin parameter, and the above efficacy difference includes the difference direction (greater or less) and the numerical value.

[0034] S150, extract the skin parameter with an efficacy difference exceeding a set value as a parameter to be adjusted, and according to the correlation between the skin parameter and the raw material formula data stored in the above first modified reference or second modified reference, obtain the raw material formula associated with the above parameter to be adjusted and determine the adjustment coefficient of at least one group of raw material formula.

[0035] S160, store the adjusted raw material formula data in association with the target skin parameter to form the standard reference library.

[0036] In step S110, the correlation analysis method includes a single factor analysis method and a multi-factor analysis method. In the embodiment, the selected main target events include skin moisture content data, oil secretion data, and barrier function data. The secondary target events include pigmentation data, skin pH value data, and pore state data. First, the single factor analysis method is used to determine the important factors affecting the above-mentioned main target events and secondary target events, i.e., the types of formula raw materials. Then, the multi-factor analysis method is used to determine the multiple formula raw materials and the proportions affecting the target events. In a specific embodiment, a nomogram prediction model can be constructed based on the multi-factor analysis method to determine the influence of each formula raw material and proportion on the target events, which is beneficial to intuitively understand the skin parameter data corresponding to different raw material formula data. In actual application, principal component regression analysis can also be used to analyze and obtain the relationship between each formula raw material and each skin parameter.

[0037] In step S120, the multiple correlation data sets include different raw material formulas and corresponding skin parameters. The raw material formulas include raw material types, such as glycerin, titanium dioxide, vitamin B3, etc., and also include the weight proportions of each raw material.

[0038] The purpose of performing step S131 is that when the test data of the current user does not exist in the skin quality test database, the first standard reference data meeting the first standard of most subjects of the same age and gender is directly used as the standard reference library, which can maximize the avoidance of allergic or uncomfortable reactions of the current user to the verification skin care sample, and ensure the safety of the execution of step S300.

[0039] The data amount in step S133 refers to the minimum amount of data required for analyzing the correlation between each skin parameter and formula raw material. Generally, the data amount of different parameters should not be less than 30.

[0040] In another embodiment, the method for generating the standard reference library in step S100 includes: S170, dividing the skin into multiple different skin types according to different skin parameters or combinations thereof. In the embodiment, the dimensions for defining the skin types include skin quality, sensitivity, wrinkles, pigmentation, and pore characteristics data. The skin is divided into not less than 1000 skin types according to different dimensions and stored.

[0041] S180, directly configuring at least one corresponding basic formula data for each skin type based on the existing technology. For example, pigmentation is divided into sunspots, chloasma, freckles, etc. Different types and severity of pigmentation correspond to at least one (usually multiple) basic formula. The above-mentioned basic formula is bound and entered in advance by the staff.

[0042] The personalized skin care product formula generation method further includes: S410, obtain the skin parameters of the current user and determine the skin type of the user based on the AI recognition algorithm. In practical applications, the above-mentioned skin parameters are obtained by using a hyperspectral image recognition algorithm to recognize skin parameters such as skin water content and oil secretion, and then determining the skin type through an AI recognition algorithm.

[0043] S420, according to the skin type determined in step S410, the corresponding basic formula is retrieved.

[0044] S430, according to the user's specified or based on the environmental parameters and target skin parameters of the current user in the future setting period, combined with the standard formula data and based on the first correlation relationship, the proportion of raw materials in the standard formula data is adjusted to generate personalized skin care product formula data. For example, according to the user's specified or automatically generated formula data, different types of water are added to the basic formula, usually distilled water is added, and if moisture control is required, fermented water or alcohol dew is added.

[0045] The above technical solution first directly stores the basic formula data corresponding to each type of skin, then accurately determines the type of the user's skin through an AI recognition algorithm, and then obtains the basic formula suitable for the current user according to the determination result. After obtaining the basic formula, the skin care product is automatically or manually customized according to the special effect required by the user, and finally the personalized skin care product formula data is generated, and the whole generation process does not need to go through a large amount of data analysis, which is suitable for lightweight deployment of related models.

[0046] In step S300, the skin parameters of the current user in different environments before and after using the sample are detected, and the specific method is as follows: a hyperspectral skin detector is used to obtain the hyperspectral image of the set skin area of the current user, and a hyperspectral image analysis algorithm is used to generate skin parameter data. The core principle of hyperspectral detection and analysis of skin state is to quantitatively analyze skin tissue composition and physiological state through spectral reflection and absorption characteristics of different wave bands, such as oil and pore analysis, dermal collagen structure and skin water content, etc., which can realize rapid and high-frequency collection of skin parameter data.

[0047] In the embodiments of the present application, the detection of the above-mentioned skin parameters can be performed in a specific natural environment, for example, collecting skin parameters related to ultraviolet rays in an environment with strong ultraviolet rays outdoors, collecting skin parameters related to skin capillary dilation in an area with high altitude and low oxygen content, and in actual application, the relevant skin parameters can also be obtained through an environmental chamber for testing or a skin tester. By changing the parameters of the environmental chamber or the skin tester, such as simulating the output light intensity, ultraviolet intensity, temperature and humidity, etc., different environmental parameters can be provided in different test periods, and the skin parameters of the current user under each environmental parameter condition can be collected in a short period of time. Thus, the skin parameters or combinations thereof of the current user and the correlation between the skin parameter change trend and each environmental parameter can be obtained through analysis, so as to facilitate the subsequent generation of more accurate and applicable personalized skin care product raw material formula data.

[0048] In order to shorten the generation time of the personalized skin care product formula data and avoid inaccurate generation of the skin care product formula data due to the absence of part of the skin parameter data, the personalized skin care product formula generation method in the embodiments of the present application further comprises: The skin parameter data of the current user in different environments before and after using the sample is obtained, and a relationship model representing the linkage change relationship between the skin parameters under different environmental parameter conditions is generated through deep neural network training. For example, the linkage relationship between the skin moisture content and the skin oil secretion of the current user under different environmental conditions, that is, the corresponding functional relationship between the two skin parameters. The deep neural network is preferably a convolutional neural network (CNN) and a back propagation neural network (BP neural network), and the training process of the relationship model, such as normalizing and dividing the training set, the validation set, etc., will not be described in detail here.

[0049] In the embodiments of the present application, the linkage change relationship includes the numerical proportional relationship and the time sequence relationship between different skin parameters, for example, when the skin moisture content is M, the skin oil secretion is w*M, and for another example, the skin temperature rises by 1℃ and the oil secretion increases by 10% after maintaining for 20 min. With the help of the above-mentioned relationship model, the remaining skin parameter data can be calculated based on one or more skin parameter data, that is, only a small amount of data needs to be collected to obtain the required skin parameter data, thereby providing complete data support for the generation of personalized skin care product raw material formula.

[0050] Research shows that there is a very close correlation between the skin parameters of the user and the physiological parameters thereof, such as the correlation between the skin glossiness and the sleep quality. In order to make the personalized skin care product more suitable for the current user, in the embodiments of the present application, the personalized skin care product formula generation method further comprises: S310, based on the above standard formula data, a verification skin care product sample is made, and the skin parameters of the current user under different physiological parameter conditions before and after using the sample are detected, and a second correlation relationship between each skin parameter and its change trend and each physiological parameter is fitted and generated. The physiological parameters include but are not limited to skin temperature, heart rate, blood oxygen saturation, sleep quality, and motion data.

[0051] S320, obtaining the expected physiological parameters of the current user in a future set period and the target skin parameters, combining the standard formula data and based on the second correlation relationship, correcting the personalized skin care product formula data. Wherein, obtaining the expected physiological parameters of the current user in a future set period includes: obtaining the schedule of the current user, and obtaining the expected physiological parameters of each period such as sleep period, exercise period, and going-out period. The above expected physiological parameters can also be obtained by predicting the historical physiological parameter data of the user, for example, by analyzing the data collected and stored by the smart wearable device worn by the user to obtain the physiological parameter data of each period.

[0052] The above technical solution can adjust the required personalized skin care product raw material formula according to the expected physiological parameter data of the current user in a set period, so that the personalized skin care product made is more suitable for the current user.

[0053] Based on the analysis of the physiological parameter data collected in each period, the personalized skin care product formula generation method described in the application further includes: obtaining the environmental parameter data and the physiological parameter data of each period of the current user, and comparing to generate the applicable period of the current personalized skin care product. Therefore, based on the above technical solution, the applicable period of the skin care product can also be generated corresponding to the generation of the skin care product formula suitable for the current user, further improving the use effect of the personalized skin care product.

[0054] A personalized skin care product formula generation system based on skin state analysis, as shown in Figure 3 The system mainly includes a standard reference library generation unit 100, a standard formula generation unit 200, a first correlation relationship generation unit 300, and a personalized formula generation unit 400.

[0055] The standard reference library generation unit 100 includes a data communication module and a data storage module on the hardware, and is configured to obtain the raw material formula data corresponding to each skin parameter or its combination in the skin quality test database, analyze and generate the correlation relationship between each skin parameter or its combination and each formula raw material, and store it.

[0056] The standard formula generating unit 200 is configured to obtain the skin parameters of the current user, and generate and store the corresponding standard formula data based on the standard reference library. The first correlation relationship generating unit 300 is configured to obtain the skin parameters of the current user in different environments before and after using the skin care sample, and fit to generate the first correlation relationship between the skin parameters and the change trend and the environmental parameters. The sample is made based on the standard formula data. In the embodiments of the present application, the first correlation relationship generating unit 300 includes a skin parameter obtaining subunit 301, an environmental parameter obtaining subunit 302, and a data fitting subunit 303. The skin parameter obtaining subunit 301 is data-connected with the data output end of the hyperspectral skin detector, and is used to obtain the hyperspectral image of the set skin area of the current user, and generate the skin parameter data through the hyperspectral image analysis algorithm. The environmental parameter obtaining subunit 302 is data-connected with the data output end of the environmental cabin, and is used to provide different environmental parameters in different test periods. The data fitting subunit 303 is configured to obtain the skin parameters of the current user in different environments before and after using the skin care sample, and fit to generate the first correlation relationship.

[0057] The personalized formula generating unit 400 is configured to obtain the environmental parameters and the target skin parameters of the current user in the future set period, combine the standard formula data, and generate the personalized skin care formula data based on the first correlation relationship, and adjust the proportion of the raw materials in the standard formula data.

[0058] In order to further improve the adaptability between the personalized skin care formula and the user, the system described in the present application further includes a second correlation relationship generating unit 500, a formula correction data generating unit 600, and a applicable period generating unit 700.

[0059] The second correlation relationship generating unit 500 is internally provided with a physiological parameter obtaining subunit 501, which is configured to obtain the skin parameters of the current user in different physiological parameter conditions before and after using the skin care sample, and fit to generate the second correlation relationship between the skin parameters and the change trend and the physiological parameters. The formula correction data generating unit 600 is configured to obtain the expected physiological parameters and the target skin parameters of the current user in the future set period, combine the standard formula data, and generate the correction parameters for correcting the personalized skin care formula data based on the second correlation relationship. The applicable period generating unit 700 is used to obtain the environmental parameter data and the physiological parameter data of the current user in each period, compare to generate the applicable period of the current personalized skin care product, and output.

[0060] In actual application, each functional module is loaded in the computer, and data transmission between the computer and the external device is realized by using the related data communication module.

[0061] In order to facilitate the popularization and use of the personalized skin care product formula generation method described in the scheme of the present application, a computer readable storage medium is disclosed herein, which loads a computer program module thereon, and when the computer program module is executed by a processor, it is used to realize the personalized skin care product formula generation method based on skin state analysis as described above.

[0062] Although the embodiments of the present application have been shown and described above, it should be understood that the above-mentioned embodiments are exemplary and cannot be understood as limiting the present application, and those of ordinary skill in the art can make changes, modifications, replacements and variations to the above-mentioned embodiments within the scope of the present application.

Claims

1. A method for generating personalized skincare product formulations based on skin condition analysis, characterized in that, include: Obtain the raw material formula data corresponding to each skin parameter or its combination, analyze and generate the correlation between each skin parameter or its combination and each formula raw material, and store it as a standard reference library; Obtain the current user's various skin parameters and generate corresponding standard formula data based on the aforementioned standard reference library; Based on the above standard formula data, a small sample of skin care product was made for verification. Skin parameters of the current user were detected in different environments before and after using the sample. The first correlation between each skin parameter and its changing trend and environmental parameters was generated by fitting. The system obtains environmental parameters and target skin parameters of the current user's environment within a future set time period, combines the standard formula data, and adjusts the ingredient ratios in the standard formula data based on the first correlation to generate personalized skincare formula data. The skin parameters include moisture content data, sebum secretion data, pigmentation data, barrier function data, elasticity data, pore condition data, skin pH data, and skin temperature data. The environmental parameters include ambient temperature and humidity data, ultraviolet radiation intensity data, airborne particulate matter concentration data, and set allergen concentration data.

2. The method for generating personalized skincare product formulations based on skin condition analysis according to claim 1, characterized in that, Methods for generating standard reference libraries also include: Skin is classified into several different skin types based on different skin parameters or combinations thereof; Based on existing technology, at least one corresponding basic formula data can be directly configured for each skin type; The method for generating personalized skincare product formulas also includes: Obtain various skin parameters of the current user and determine the user's skin type based on AI recognition algorithms; Find and retrieve the corresponding basic formula based on the already determined skin type; Based on the environmental parameters specified by the user or based on the environmental parameters of the current user's environment in a future set time period and the target skin parameters, combined with the standard formula data and based on the first correlation, the raw material ratio in the standard formula data is adjusted to generate personalized skin care product formula data; The dimensions for defining skin type include: skin texture, sensitivity, wrinkles, pigmentation, and pore characteristics.

3. The method for generating personalized skincare product formulations based on skin condition analysis according to claim 1, characterized in that, Analyze the correlation between each skin parameter or combination thereof and each formulation ingredient, and store the results as a standard reference library, including: We searched the skin type testing database for subjects whose age and gender were similar to the current user, obtained their skin parameter data after using different formula skin care products, and used correlation analysis to obtain the correlation coefficient between each formula ingredient and each skin parameter or their combination. Skin parameter data with correlation coefficients greater than a set threshold and their corresponding raw material formulation data are extracted, forming multiple associated data groups, which are then stored as the first standard reference database. Determine if the skin type test database contains the current user's trial data: If there is no trial data for the current user, then the first standard reference database will be used as the standard reference library; If trial data for the current user exists, extract the current user's skin parameter data and its corresponding raw material formula data, and temporarily associate and store them. Determine the amount of data for the current user's skin parameters and their corresponding ingredient formulas: If it exceeds the set value, the correlation analysis method is used to determine the correlation between the current user's various skin parameters or their combinations and the formulas of various raw materials, and multiple sets of correlation relationships are generated as the first correction reference. If it does not exceed the set value, then the first standard reference database is used as the second correction reference; Set the target skin parameters for the current user, and calculate the efficacy difference between the stored skin parameters and the target skin parameters; Skin parameters with efficacy differences exceeding a set value are extracted as parameters to be adjusted. Based on the correlation between skin parameters and raw material formula data stored in the first or second correction reference, the raw material formula associated with the parameters to be adjusted is obtained and at least one set of adjustment coefficients for the raw material formula is determined. The adjusted raw material formula data is associated with and stored with the target skin parameters to form the standard reference library.

4. The method for generating personalized skincare product formulations based on skin condition analysis according to claim 1, characterized in that, Detect skin parameters of the current user before and after using the sample in different environments, including: The hyperspectral image of the user-defined skin region is obtained using a hyperspectral skin detector, and various skin parameter data are generated using a hyperspectral image analysis algorithm. The skin parameters were detected in a test environment chamber to provide different environmental parameters at different test periods.

5. The method for generating personalized skincare product formulations based on skin condition analysis according to claim 1, characterized in that, The method for generating personalized skincare product formulas also includes: The system acquires various skin parameter data of the current user before and after using the sample in different environments, and uses a deep neural network to train and generate a relationship model to represent the linkage and change relationship between various skin parameters under different environmental parameter conditions. The aforementioned linkage and change relationships include the numerical ratio relationship and the temporal relationship between different skin parameters.

6. The method for generating personalized skincare product formulations based on skin condition analysis according to claim 1, characterized in that, The personalized skincare product formulation generation method further includes: Based on the above standard formula data, a small sample of skin care product was made for verification. The skin parameters of the current user before and after using the above sample were detected under different physiological parameter conditions. The second correlation relationship between each skin parameter and its changing trend and each physiological parameter was generated by fitting. Obtain the current user's expected physiological parameters and target skin parameters within a future set time period, combine them with the standard formula data, and based on the second correlation, correct the personalized skincare formula data; The physiological parameters include: skin temperature, heart rate, blood oxygen saturation, sleep quality, and exercise data.

7. The method for generating personalized skincare product formulations based on skin condition analysis according to claim 6, characterized in that, The method for generating personalized skincare product formulas also includes: It acquires environmental and physiological parameter data of the current user at various times, compares them, and generates the applicable time period for the current personalized skin care product.

8. A personalized skincare product formulation generation system based on skin condition analysis, used to implement the personalized skincare product formulation generation method based on skin condition analysis as described in any one of claims 1-7, characterized in that, include: The standard reference library generation unit (100) is configured to obtain the raw material formula data corresponding to each skin parameter or its combination, analyze and generate the correlation between each skin parameter or its combination and each formula raw material, and store it. The standard formula generation unit (200) is configured to acquire various skin parameters of the current user, generate corresponding standard formula data based on the above standard reference library, and store it. The first association generation unit (300) is configured to obtain the skin parameters of the current user before and after using a skin care product sample in different environments, and to fit and generate a first association between each skin parameter and its changing trend and the environmental parameters. The sample is made based on the above-mentioned standard formula data. The personalized formula generation unit (400) is configured to obtain environmental parameters and target skin parameters of the environment in which the current user will be located in a future set time period, combine the standard formula data and adjust the raw material ratio in the standard formula data based on the first correlation relationship, and generate personalized skin care product formula data. The skin parameters include moisture content data, sebum secretion data, pigmentation data, barrier function data, elasticity data, pore condition data, skin pH data, and skin temperature data. The environmental parameters include ambient temperature and humidity data, ultraviolet radiation intensity data, airborne particulate matter concentration data, and set allergen concentration data.

9. The personalized skincare formula generation system based on skin condition analysis according to claim 8, characterized in that, The first association generation unit (300) includes: The skin parameter acquisition subunit (301) is connected to the data output terminal of the hyperspectral skin detector and is used to acquire the hyperspectral image of the skin area currently set by the user and generate various skin parameter data through the hyperspectral image analysis algorithm. The environmental parameter acquisition subunit (302) is connected to the data output terminal of the environmental chamber and is used to provide different environmental parameters at different test periods; The data fitting subunit (303) is configured to obtain the skin parameters of the current user before and after using a skin care product sample in different environments, and to fit and generate the first correlation relationship.

10. The personalized skincare formula generation system based on skin condition analysis according to claim 8, characterized in that, Also includes: The second association generation unit (500) is configured to obtain the skin parameters of the current user before and after using a skin care product sample under different physiological parameter conditions, and to fit and generate a second association between each skin parameter and its changing trend and each physiological parameter. The formula correction data generation unit (600) is configured to obtain the expected physiological parameters and target skin parameters of the current user in a future set time period, combine the standard formula data and generate correction parameters for correcting personalized skin care product formula data based on the second correlation. The applicable time period generation unit (700) is used to obtain environmental parameter data and physiological parameter data of the current user at various time periods, compare and generate the applicable time period of the current personalized skin care product and output it.

11. A computer-readable storage medium, characterized in that, It is loaded with a computer program module, which, when executed by a processor, is used to implement the personalized skin care product formula generation method based on skin condition analysis as described in any one of claims 1-7.

Images