Human body quantitative evaluation method for anti-oxidation effect of cosmetics and application

Abstract

The invention relates to the technical field of data processing, in particular to a human body quantitative evaluation method for the anti-oxidation effect of cosmetics and application, cosmetic attribute data and body part skin data when to-be-evaluated cosmetics are smeared on a human body part are collected, and the smearing sequence of the to-be-evaluated cosmetics is evaluated; performing feature detection on the cosmetic attribute data to obtain skin abnormal information with smearing part data along a cosmetic smearing range, and performing feature detection on the body part skin data to obtain space skin abnormal information with smearing part data; based on the smearing sequence of the to-be-evaluated cosmetics, whether the to-be-evaluated cosmetics cover the position area corresponding to the space skin abnormal information or not is judged; if yes, calculating superposition information and outputting the superposition information; and if not, outputting space skin abnormal information. The anti-oxidation synergistic effect of the cosmetic is identified according to multiple operation sequences of the cosmetic on the skin and the smearing position of the cosmetic, and the actual application environment is better met.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a human quantitative evaluation method for the antioxidant efficacy of cosmetics and its application. Background Art

[0002] Traditional cosmetic safety assessments mainly rely on animal experiments and clinical trial data, with limited sample sizes, which make it difficult to fully reflect actual consumer usage. The antioxidant response of cosmetics is a dynamic process that is affected by the characteristics of the cosmetics themselves, as well as by the user's skin condition, environmental factors, and other factors. Traditional static statistical models are difficult to describe this complex dynamic relationship. Cosmetic antioxidants should differ significantly among different users, and some people are more sensitive to specific ingredients or doses of multiple cosmetics used in combination. Traditional group average models are difficult to reflect this user diversity.

[0003] Therefore, in the development of cosmetics, it is very important to master the various operation sequences of cosmetics on the skin and the application positions of cosmetics. Therefore, in the beauty industry, a human quantitative evaluation method and application for the antioxidant effects of various cosmetics are urgently needed. Summary of the invention

[0004] To achieve the above objectives, this application provides the following technical solutions: According to the first aspect of the present invention, the present invention claims a method for quantitatively evaluating the antioxidant efficacy of cosmetics on human body, comprising: Collecting cosmetics property data and body part skin data when the cosmetics to be evaluated are applied to human body parts, and evaluating the application order of the cosmetics to be evaluated; the cosmetics property data include state data, ingredient data and application part data; the body part skin data include original skin bare face information and / or skin with makeup image data; the original skin bare face information and the skin with makeup image data both include application part data; Performing feature detection on the cosmetics attribute data to obtain skin abnormality information along the cosmetics application range with application site data, and performing feature detection on the body site skin data to obtain spatial skin abnormality information with application site data; Based on the application order of the cosmetics to be evaluated, determining whether the cosmetics to be evaluated cover the position area corresponding to the spatial skin abnormality information; If yes, then superimposed information is calculated and output; the superimposed information is a feature obtained by superimposing the skin abnormality information along the cosmetic application range and the spatial skin abnormality information at the same position, or the superimposed information is the skin abnormality information along the cosmetic application range; If not, the spatial skin abnormality information is output.

[0005] Furthermore, the method further comprises: The output superposition information and the spatial skin abnormality information are aggregated to determine the skin abnormality information of the human body part.

[0006] Furthermore, the data of the properties of the cosmetics to be evaluated when applied to a part of the human body is collected, specifically including: Collecting the state data and ingredient data of the cosmetics to be evaluated when applied to the human body, collected by the allergy sensor; The collection tracking unit collects application site data of the cosmetics to be evaluated when applied on a part of the human body.

[0007] Furthermore, the collecting of the skin data of the body part when the cosmetic to be evaluated is applied to the body part of the human body specifically includes: The original skin information when the cosmetics to be evaluated are applied on the human body is collected by the infrared sensor, and / or the skin image data with makeup when the cosmetics to be evaluated are applied on the human body is collected by the camera.

[0008] Furthermore, the evaluating the application order of the cosmetics to be evaluated specifically includes: Collecting the central application site data of the cosmetic to be evaluated when applied to the body part of the human body calculated by the tracking unit; Collecting the average value of the thickness and center color of the cosmetics to be evaluated when applied to the body part of the human body calculated by the fluorescence intensity analysis unit; Calculating the cosmetic sample application site data when the cosmetic to be evaluated is applied to the human body part based on the central application site data of the cosmetic to be evaluated, the cosmetic application thickness, the central color mean, and the skin perception current; The application order of the cosmetics to be evaluated is evaluated based on the cosmetic sample application site data.

[0009] Furthermore, when the skin data of a body part includes original skin bare face information and skin makeup image data, feature detection is performed on the skin data of the body part to obtain spatial skin abnormality information with makeup part data, specifically including: Performing denoising processing on the original bare-face information and the skin-with-makeup image data respectively; Perform feature detection on the original bare face information after denoising based on the spatial skin abnormality information detection model to obtain a first record; Performing feature detection on the denoised skin makeup image data based on the spatial skin abnormality information detection model to obtain a second record; At a preset position, when the first record stores free radical data antioxidant capacity information containing skin abnormality information and the second record does not store a depth image containing skin abnormality information, the free radical data antioxidant capacity information is processed to determine the spatial skin abnormality information with application site data corresponding to the preset position; the preset position is any position of the human body part; At a preset position, when the first record does not store free radical data antioxidant capacity information containing skin abnormality information and the second record stores a depth image containing skin abnormality information, the depth image is processed to determine the spatial skin abnormality information with application site data corresponding to the preset position; At a preset position, when the first record stores free radical data antioxidant capacity information containing skin abnormality information and the second record stores a depth image containing skin abnormality information, the free radical data antioxidant capacity information and the depth image are subjected to time-space synchronization processing to determine the spatial skin abnormality information with application site data corresponding to the preset position; The spatial skin abnormality information detection model includes a facial skin abnormality information detection sub-model, an upper limb skin abnormality information detection sub-model and a lower limb skin abnormality information detection sub-model; and the face, the upper limb and the lower limb are respectively input into the facial skin abnormality information detection sub-model, the upper limb skin abnormality information detection sub-model and the lower limb skin abnormality information detection sub-model to determine the probability values ​​of the face, the upper limb and the lower limb corresponding to each preset skin attribute in a plurality of preset skin attributes; Determine, according to the probability values, the original skin attribute information of the face, the upper limbs, and the lower limbs corresponding to the original skin without makeup or with makeup; For any preset skin attribute among the plurality of preset skin attributes, performing the following weighted sum operation: calculating the weighted sum of the probability values ​​of the face, the upper limbs and the lower limbs respectively on the preset skin attribute as the probability value of the user's face on the preset skin attribute; Determine skin attribute information of the user's face according to probability values ​​of the user's face at various preset skin attributes; The skin abnormality information detection model is a convolutional neural network model; The skin abnormality information detection model is trained using a plurality of images in which characteristic regions are divided and skin attribute annotations are added to the characteristic regions.

[0010] Furthermore, the calculating of the superposition information specifically includes: Using a superposition weighted formula, the skin abnormality information along the cosmetic application range and the spatial skin abnormality information at the same position are superimposed to obtain superimposed information; Alternatively, when the spatial skin abnormality information does not exist in the position area corresponding to the spatial skin abnormality information, the skin abnormality information along the cosmetic application range is determined as the superimposed information; Wherein, the weight value of the superposition weighted formula is determined according to the weight value formula; The weight value formula is a formula constructed by taking the skin site area deviation of the sample of the cosmetic to be evaluated and the temperature and humidity of the sample of the cosmetic to be evaluated as input; The regional deviation of the sample skin part is obtained by superimposing the deviation value of the fluorescence intensity analysis unit and the deviation value of the tracking unit; The deviation value of the fluorescence intensity analysis unit is a deviation value generated by the fluorescence intensity analysis unit when the fluorescence intensity analysis unit and the tracking unit jointly determine the application site data of the cosmetic sample when the cosmetic to be evaluated is applied to the body part of the human body; The deviation value of the tracking unit is a deviation value generated by the tracking unit when the fluorescence intensity analysis unit and the tracking unit jointly determine the application site data of the cosmetic sample when the cosmetic to be evaluated is applied to the body part of the human body; The output sample skin area deviation σ and sample temperature and humidity B are used as input to construct the weight value formula, and the weight value ξ of the superimposed weighted formula is calculated according to the weight value formula y and 1-ξ y , where ξ y is the weight value of skin abnormality information along the cosmetic application range, 1-ξ y is the weight value of spatial skin abnormality information; ; According to the calculated weight value β of the superposition weighted formula, the obtained abnormal information is extended along the y-axis range. In the section where the abnormal information covers the vertical x-axis range of point M, the coordinates of point M are (x1, 0, z1) and the coordinates of point N are (x1, y2, z2); ; The depth of each point of skin abnormality information is the formula of the horizontal and vertical coordinates corresponding to the point, that is, z = f(x, y); Construct the weighted formula of the superposition model according to the weight value: ; The coordinates in the above formula are all in the sample coordinate system; g(x,y) is the depth of each coordinate point of the visually detected skin abnormality information, f(x,y) is the depth of each coordinate point of the response detected skin abnormality information, and (x,y,z) is the coordinate of the skin abnormality information after superposition.

[0011] According to the second aspect of the present invention, the present invention seeks protection for the application of the human quantitative evaluation method for the antioxidant efficacy of cosmetics in the evaluation of cosmetic effects.

[0012] The present application relates to the field of data processing technology, and in particular to a method and application for quantitative evaluation of the antioxidant efficacy of cosmetics on the human body, which collects cosmetics attribute data and body part skin data when the cosmetics to be evaluated are applied to human body parts, and evaluates the application order of the cosmetics to be evaluated; performs feature detection on the cosmetics attribute data to obtain skin abnormality information along the cosmetic application range with application part data, and performs feature detection on the body part skin data to obtain spatial skin abnormality information with application part data; based on the application order of the cosmetics to be evaluated, determines whether the cosmetics to be evaluated cover the position area corresponding to the spatial skin abnormality information; if so, calculates the superposition information and outputs the superposition information; if not, outputs the spatial skin abnormality information. The present invention identifies the synergistic antioxidant efficacy of cosmetics through the various operation sequences of cosmetics on the skin and the application positions of cosmetics, which is more in line with the actual application environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a workflow diagram of a method for quantitatively evaluating the antioxidant efficacy of cosmetics on humans as claimed in an embodiment of the present application. DETAILED DESCRIPTION

[0014] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0015] The terms "first", "second" and "third" in this application are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined as "first", "second" and "third" can explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. All range indications in the embodiments of this application (such as up, down, left, right, front, back...) are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the range indication also changes accordingly. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0016] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0017] The traditional method of testing the antioxidant capacity of cosmetics is mainly to add various chemicals, solubilizers and extraction solutions to the samples to be tested. However, these methods change or destroy the original chemical formula of the cosmetics. Therefore, the traditional detection methods cannot be used to test cosmetic samples. In addition, the traditional methods only use detection agents that are sensitive to one type of active oxygen or active nitrogen, and cannot comprehensively evaluate the antioxidant capacity of cosmetics.

[0018] In view of this, the purpose of the present invention is to provide a human quantitative evaluation method and application for the antioxidant efficacy of various cosmetics, so as to achieve accurate and rapid detection of skin quality.

[0019] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] Embodiment 1 Abnormal skin information: a type of skin surface feature that is inconsistent with the statistical properties of the skin at a long spatial scale. In particular, it refers to features that are formed due to significant changes in external skin contact within a smaller spatial range (such as rosacea, dermatitis, rashes, blackheads, etc.).

[0021] Multi-source input superposition model and its function: A method of superimposing and correcting the test records of multiple cosmetics to achieve skin quality (abnormal information) detection.

[0022] This embodiment takes into account the limitations of spatial detection and allergy sensor detection, and performs skin quality detection under the superposition of multi-source input signals such as ultraviolet sensors, air quality sensors, and photosensors. The superposition between the two is achieved based on the cosmetic position coordinates, and the position deviation value range is fully considered during the superposition process. Specifically, this embodiment includes constructing a skin detection correction model using the distribution of cosmetic position deviation values, and using the skin abnormality information detected by the allergy sensor to correct the skin abnormality information detected by the perception sensor, and obtaining a perception record by superimposing the two. The skin detection method consists of two parts: multi-source input superposition abnormal information detection and superposition weight calculation. Therefore, the present invention provides a human quantitative evaluation system for the antioxidant efficacy of cosmetics, including: a sample position unit, an information acquisition unit, a sample position deviation value calculation unit, a skin abnormality information identification unit, a weighted formula generation unit, and a skin abnormality information superposition unit.

[0023] The information collection unit includes an allergy sensor information collection unit and a perception sensor data collection unit.

[0024] The allergy sensor information acquisition unit is used to collect cosmetic status data (such as cream, water, and lotion) through the allergy sensor, and to detect the vertical sliding characteristics of cosmetics along the skin and the changes in the cosmetic status in real time; the perception sensor data acquisition unit is used to collect the original skin bare face information during the cosmetics application process through the infrared sensor, and to collect the skin makeup image data during the cosmetics application process through the camera, and send the records obtained from the above two to the skin abnormality information recognition unit.

[0025] The skin abnormality information recognition unit includes a response-based skin abnormality information recognition model and a spatial skin abnormality information recognition model.

[0026] The response-based skin abnormality information recognition model is used to perform feature detection on cosmetic ingredient data to obtain skin abnormality information along the cosmetic application range with application site data. The response-based skin abnormality information recognition model includes a cluster analysis model, a perturbation observer detection model, and a threshold-based detection method.

[0027] The spatial skin abnormality information recognition model includes an information analysis unit and a data superposition unit; wherein the information analysis unit includes an infrared sensor information analysis unit and a camera information analysis unit. The camera information analysis unit performs grayscale, noise reduction, binarization and other denoising and feature space-based clustering analysis and segmentation processing on the skin makeup image data collected by the camera to obtain a depth image with skin abnormality information. The infrared sensor information analysis unit performs noise reduction, image pixel segmentation, image pixel clustering and other processing on the original skin makeup information collected by the infrared sensor to obtain free radical data antioxidant capacity information containing skin abnormality information. The data superposition unit determines whether the infrared sensor information analysis unit and the camera information analysis unit both output information containing skin abnormality information, so as to select different superposition models, and then output the final spatial skin abnormality information with application site data.

[0028] The visual skin abnormality information recognition model, the infrared sensor skin abnormality information recognition model, and the superposition model described in this embodiment are not limited to the above methods.

[0029] The sample position deviation value calculation unit takes the tracking system deviation value and the fluorescence intensity analysis system deviation value as input, obtains the sample tracking deviation value (expressed by standard deviation σ) by superposition calculation of the deviation values, and inputs the sample tracking deviation value into the weighted formula generation unit.

[0030] The weighted formula generation unit is used to construct a weighted formula using the sample tracking information deviation value (σ) as input (the weighted formula described below is only an example).

[0031] The skin abnormality information superposition unit is used to judge whether the cosmetics cover the position area corresponding to the spatial skin abnormality information based on the evaluated cosmetic position (here is the application order of the cosmetics mentioned below) and the spatial skin abnormality information with the application site data. If it is judged that it does not cover, the spatial skin abnormality information is output; if it is judged that it covers, the input multi-source detected skin abnormality information (i.e., the spatial skin abnormality information with the application site data and the skin abnormality information along the cosmetic application range with the application site data) is weighted using a weighting formula to obtain the superimposed skin abnormality information with the application site data.

[0032] The above is an introduction to the various parts of the skin quality detection system. The following is a detailed introduction to the corresponding methods of the described skin quality detection system: First, two global coordinate systems are established, namely the human body coordinate system (X, Y, Z) and the sample coordinate system (x, y, z). The right-handed coordinate system, i.e., the sample coordinate system, is established with the point where the sample touches the human body as the origin, the cosmetic application range as the positive range of the x-axis, and the vertical upward as the positive range of the z-axis.

[0033] The ordinates of the human body coordinate system (X, Y, Z) and the sample coordinate system (x, y, z) are the same, that is, Z = z. The coordinate transformation of the abnormal information in the sample coordinate system and the human body coordinate system is as follows: Assume that the coordinates of the abnormal information in the sample coordinate system are (x1, y1, z1) and the coordinates in the human body coordinate system are (X1, Y1, Z1). The transformation matrix between the sample coordinate system and the human body coordinate system is: ,but , Z1=z1.

[0034] in, , .

[0035] R is the rotation matrix, and T is the coordinate of the origin of the sample coordinate system in the human body coordinate system.

[0036] Step 1: When cosmetics are applied normally, the allergy sensor starts to collect cosmetic attribute data, which includes but is not limited to application status data, ingredient data, etc.; the perception sensor collects skin data of body parts, among which the infrared sensor collects the original skin information during the cosmetics application process, and the camera collects the skin makeup image data during the cosmetics application process as the input of the skin abnormality information recognition unit.

[0037] Step 2: After the skin abnormality information recognition unit collects the information input by the information collection unit, it processes the cosmetics attribute data based on the corresponding skin abnormality information recognition model to obtain abnormal information. Specifically, firstly, an abnormal information threshold is set, and then whether it is abnormal information is determined based on whether the z-axis acceleration of the cosmetics input by the allergy sensor is greater than the abnormal information threshold. The length l1 and depth z of the abnormal information along the central axis of the sample are calculated to obtain an image of the skin abnormality information along the central axis of the sample; wherein, in the image of the skin abnormality information along the central axis of the sample, in the sample coordinate system, the x-axis is the cosmetics application range, the z-axis is the vertical range, and the xz coordinate system is the spatial coordinate, that is, what is obtained is the depth of the abnormal information along the cosmetics application range.

[0038] Among them, the depth z of each point along the x range of skin abnormality information can be regarded as the formula f(x,y) of the horizontal and vertical coordinates (x,y). Here, the coordinates are fixed to the center of the skin where the cosmetic sample is connected, so its vertical coordinate is zero, that is, y=0. The horizontal coordinate of point M is x1, and the depth is z1, then z1=f(x1,0).

[0039] The spatial skin abnormality information recognition model includes an information parsing unit and a data superposition unit.

[0040] After receiving the skin data of the body part sent in step 1, the information analysis unit processes the original skin information, including: first performing image pixel denoising, image pixel segmentation and other denoising, using a clustering model to perform clustering analysis on the denoised original skin information, and obtaining free radical data antioxidant capacity information with skin abnormality information as input to the data superposition unit. The denoised skin image is segmented using a segmentation model to obtain a depth image containing skin abnormality information as input to the data superposition unit.

[0041] The data superposition unit determines whether the infrared sensor information analysis unit and the camera information analysis unit in step 2 have both output information containing skin abnormality information, which is divided into three situations: 1. When only the infrared sensor information analysis unit outputs information containing skin abnormality information, the bounding box is fitted on the antioxidant capacity information of the free radical data containing skin abnormality information to obtain the abnormal information and use it as the spatial detection of skin abnormality information.

[0042] 2. Only the camera information analysis unit outputs the information containing skin abnormality information. The convolutional neural network model is used to perform feature detection on the depth image containing skin abnormality information to obtain the abnormal information and use it as the spatial detection of skin abnormality information.

[0043] 3. When both the infrared sensor information analysis unit and the camera information analysis unit output information containing skin abnormality information, the data are first synchronized in time to obtain the infrared sensor antioxidant capacity information and camera data after feature detection at the same time; then spatial synchronization is performed to convert the antioxidant capacity information into the image coordinate system to achieve spatial data matching between the infrared sensor and the camera; then the sensor data are superimposed to obtain superimposed data; finally, the superimposed data is fitted with a bounding box to obtain abnormal information and use it as spatial detection of skin abnormality information.

[0044] The depth z of each point in the spatially detected skin abnormality information is a formula corresponding to the horizontal and vertical coordinates, that is, z=g(x,y).

[0045] Step 3: While the allergy sensor is collecting information, the sample position unit is also collecting information and calculating the sample application site data in real time. The sample application site data can be calculated through the center position, the mean value of the cosmetic color, the current sensed by the cosmetic skin, etc. Specifically, the data of the cosmetic center application site is first collected in real time through the GPS tracking system, and the cosmetic application thickness θ, the center color mean β, etc. are collected in real time through the fluorescence intensity analysis system. Then, the cosmetic coverage area ψ is calculated by formula (1), and then the sample position is calculated based on the cosmetic coverage area ψ and the cosmetic center application site data.

[0046] Evaluate the order of applying cosmetics based on the sample application site data.

[0047] Step 4: The deviation of the fluorescence intensity analysis system conforms to the normal distribution with a mean of 0 and a standard deviation of σ1, and the deviation of the tracking system conforms to the normal distribution with a mean of 0 and a standard deviation of σ2. In the process of calculating the sample position, it is assumed that the deviation of the fluorescence intensity analysis system and the tracking system are independent of each other. The deviation of the sample position is calculated by superimposing the deviation of the fluorescence intensity analysis system and the deviation of the tracking system. Therefore, the deviation of the sample position also conforms to the normal distribution with a mean of 0 and a standard deviation of σ.

[0048] Step 5: Construct a superposition model weighting formula based on the sample position deviation value output from step 4 (the superposition model weighting formula shown below is for example only).

[0049] Step 6: Determine whether the cosmetics cover the position area corresponding to the spatial skin abnormality information according to the position corresponding to the spatial skin abnormality information output in step 2 and the cosmetics application order output in step 3.

[0050] If the cosmetics do not cover the location area corresponding to the spatial skin abnormality information, the spatial detected skin abnormality information is output. If the cosmetics cover the location area corresponding to the spatial skin abnormality information, the different skin abnormality information output in step 3 is superimposed and weighted by using the superposition model weighting formula output in step 5 to obtain the superimposed skin abnormality information.

[0051] The superposition model weighted formula is: The sample skin area deviation σ and sample temperature and humidity B output in step 4 are used as input to construct the weight value formula, and the weight value ξ of the superimposed weighted formula is calculated according to the weight value formula y and 1-ξ y (where ξ y is the weight value of skin abnormality information along the cosmetic application range, 1-ξ y is the weight value of spatial skin abnormality information).

[0052] ; According to the calculated weight value β of the superposition weighted formula, the abnormal information obtained in step 2 is extended along the y-axis range. In the section where the abnormal information covers the vertical x-axis range of point M, the coordinates of point M are (x1,0,z1) and the coordinates of point N are (x1,y2,z2). .

[0053] The depth of each point of skin abnormality information is the formula of the horizontal and vertical coordinates corresponding to the point, that is, z=f(x,y).

[0054] According to the weight value, the weighted formula of the superposition model is constructed as shown in formula (1): ; The coordinates described in the above formula are all in the sample coordinate system. g(x,y) is the depth of each coordinate point of the visually detected skin abnormality information, f(x,y) is the depth of each coordinate point of the response detected skin abnormality information, and (x,y,z) is the coordinate of the superimposed skin abnormality information.

[0055] Embodiment 2 See also Figure 1 This embodiment provides a method for quantitatively evaluating the antioxidant efficacy of cosmetics on human body, comprising: Step 101: collecting cosmetic property data and body part skin data when the cosmetic to be evaluated is applied to a body part, and evaluating the application order of the cosmetic to be evaluated; The cosmetics attribute data includes status data, ingredient data and application site data; the body skin data includes original skin information and / or skin makeup image data; the original skin information and the skin makeup image data both contain application site data.

[0056] Step 102: Perform feature detection on the cosmetics attribute data to obtain skin abnormality information along the cosmetics application range with application site data, and perform feature detection on the body site skin data to obtain spatial skin abnormality information with application site data.

[0057] Step 103: Based on the application order of the cosmetics to be evaluated, determine whether the cosmetics to be evaluated cover the position area corresponding to the spatial skin abnormality information; if so, execute step 104; if not, execute step 105.

[0058] Step 104: Calculate and output the superposition information; the superposition information is a feature obtained by superimposing the skin abnormality information along the cosmetic application range and the spatial skin abnormality information at the same position, or the superposition information is the skin abnormality information along the cosmetic application range.

[0059] Step 105: Output the spatial skin abnormality information.

[0060] Furthermore, the method for quantitatively evaluating the antioxidant efficacy of cosmetics on human body provided in this embodiment also includes: Step 106: Summarize the output superposition information and the spatial skin abnormality information to determine the skin abnormality information of the human body part.

[0061] The step described in this embodiment: collecting the cosmetic property data when the cosmetic to be evaluated is applied to a part of the human body, specifically includes: The state data and ingredient data of the cosmetics to be evaluated collected by the allergy sensor when applied on the human body are collected; the application site data of the cosmetics to be evaluated collected by the tracking unit when applied on the human body are collected.

[0062] The steps described in this embodiment: collecting the skin data of the body part when the cosmetics to be evaluated are applied to the body part of the human body specifically include: The original skin information when the cosmetics to be evaluated are applied on the human body is collected by the infrared sensor, and / or the skin image data with makeup when the cosmetics to be evaluated are applied on the human body is collected by the camera.

[0063] The steps described in this embodiment: evaluating the application order of the cosmetics to be evaluated specifically include: The central application site data of the cosmetic to be evaluated when applied to the human body part calculated by the collection tracking unit; the cosmetic application thickness and the central color mean value of the cosmetic to be evaluated when applied to the human body part calculated by the fluorescence intensity analysis unit; based on the central application site data of the cosmetic to be evaluated, the cosmetic application thickness, the central color mean value and the skin perception current, the cosmetic sample application site data when the cosmetic to be evaluated is applied to the human body part; and the application order of the cosmetic to be evaluated is evaluated based on the cosmetic sample application site data.

[0064] When the skin data of a body part includes original skin information without makeup and skin image data with makeup, the steps described in this embodiment: performing feature detection on the skin data of the body part to obtain spatial skin abnormality information with makeup part data specifically include: The original bare-face information and the skin with makeup image data are denoised respectively; feature detection is performed on the denoised original bare-face information based on the spatial skin abnormality information detection model to obtain a first record; feature detection is performed on the denoised skin with makeup image data based on the spatial skin abnormality information detection model to obtain a second record.

[0065] Wherein, in this embodiment, the spatial skin abnormality information detection model includes a facial skin abnormality information detection sub-model, an upper limb skin abnormality information detection sub-model and a lower limb skin abnormality information detection sub-model; and the face, the upper limb and the lower limb are respectively input into the facial skin abnormality information detection sub-model, the upper limb skin abnormality information detection sub-model and the lower limb skin abnormality information detection sub-model, and the probability values ​​of the face, the upper limb and the lower limb corresponding to each preset skin attribute in a plurality of preset skin attributes are determined; The original skin attribute information without makeup or with makeup corresponding to each of the face, the upper limbs and the lower limbs is determined according to the probability value.

[0066] For any preset skin attribute among the plurality of preset skin attributes, the following weighted sum operation is performed: a weighted sum of the probability values ​​of the face, the upper limbs and the lower limbs on the preset skin attribute is calculated as the probability value of the user's face on the preset skin attribute; The skin attribute information of the user's face is determined according to the probability values ​​of the user's face at various preset skin attributes.

[0067] The skin abnormality information detection model is a convolutional neural network model; The skin abnormality information detection model is trained using a plurality of images in which characteristic regions are divided and skin attribute annotations are added to the characteristic regions.

[0068] At a preset position, when the first record stores free radical data antioxidant capacity information containing skin abnormality information and the second record does not store a depth image containing skin abnormality information, the free radical data antioxidant capacity information is processed to determine the spatial skin abnormality information with application site data corresponding to the preset position; the preset position is any position of the human body part.

[0069] At the preset position, when the first record does not store the free radical data antioxidant capacity information containing skin abnormality information and the second record stores a depth image containing skin abnormality information, the depth image is processed to determine the spatial skin abnormality information with the application site data corresponding to the preset position.

[0070] At a preset position, when the first record stores free radical data antioxidant capacity information containing skin abnormality information and the second record stores a depth image containing skin abnormality information, the free radical data antioxidant capacity information and the depth image are processed synchronously in time and space to determine the spatial skin abnormality information with application site data corresponding to the preset position.

[0071] When the skin data of a body part includes original skin makeup information, the steps described in this embodiment: performing feature detection on the skin data of the body part to obtain spatial skin abnormality information with application part data specifically include: The original bare face information is denoised; the denoised original bare face information is feature detected based on a spatial skin abnormality information detection model to obtain a first record; at a preset position, when the first record stores free radical data antioxidant capacity information containing skin abnormality information, the free radical data antioxidant capacity information is processed to determine the spatial skin abnormality information with application site data corresponding to the preset position; the preset position is any position of the human body part.

[0072] When the skin data of a body part includes image data of skin with makeup, the steps described in this embodiment: performing feature detection on the skin data of a body part to obtain spatial skin abnormality information with makeup part data specifically include: The skin-with-makeup image data is denoised; and feature detection is performed on the denoised skin-with-makeup image data based on a spatial skin abnormality information detection model to obtain a second record.

[0073] At a preset position, when the second record stores a depth image containing skin abnormality information, the depth image is processed to determine the spatial skin abnormality information with the application site data corresponding to the preset position. The preset position is any position of the human body part.

[0074] The step of calculating the superposition information in this embodiment specifically includes: The skin abnormality information along the cosmetic application range and the spatial skin abnormality information at the same position are superimposed by using a superposition weighted formula to obtain superimposed information.

[0075] Alternatively, when the spatial skin abnormality information does not exist in the position area corresponding to the spatial skin abnormality information, the skin abnormality information along the cosmetic application range is determined as the superimposed information; Among them, the weight value of the superposition weighted formula is determined according to the weight value formula; the weight value formula is a formula constructed with the sample skin site area deviation of the cosmetic to be evaluated and the sample temperature and humidity of the cosmetic to be evaluated as input; the sample skin site area deviation is obtained by superimposing the deviation value of the fluorescence intensity analysis unit and the deviation value of the tracking unit; the deviation value of the fluorescence intensity analysis unit is the deviation value generated by the fluorescence intensity analysis unit when the fluorescence intensity analysis unit and the tracking unit jointly determine the cosmetic sample application site data when the cosmetic to be evaluated is applied to the human body part; the deviation value of the tracking unit is the deviation value generated by the tracking unit when the fluorescence intensity analysis unit and the tracking unit jointly determine the cosmetic sample application site data when the cosmetic to be evaluated is applied to the human body part.

[0076] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0077] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

[0078] In the several embodiments provided in the present application, it should be understood that the disclosed systems, units and methods can be implemented in other ways. For example, the unit embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, units or indirect coupling or communication connection of units, which can be electrical, mechanical or other forms.

[0079] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. The above is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the specification and drawings of this application, or directly or indirectly used in other related technical fields, is also included in the patent protection scope of the present application.

[0080] The specific implementation methods of the invention are described in detail above, but they are only examples, and the present application is not limited to the specific implementation methods described above. For those skilled in the art, any equivalent modification or substitution of the invention is also within the scope of the present application, and therefore, the equalization, modification, and improvement made without departing from the spirit and principle of the present application should be included in the scope of the present application.

Claims

1. A method for quantitatively evaluating the antioxidant efficacy of cosmetics on human body, characterized in that: include: Collecting cosmetics property data and body part skin data when the cosmetics to be evaluated are applied to human body parts, and evaluating the application order of the cosmetics to be evaluated; the cosmetics property data include state data, ingredient data and application part data; the body part skin data include original skin bare face information and / or skin with makeup image data; the original skin bare face information and the skin with makeup image data both include application part data; Performing feature detection on the cosmetics attribute data to obtain skin abnormality information along the cosmetics application range with application site data, and performing feature detection on the body site skin data to obtain spatial skin abnormality information with application site data; Based on the application order of the cosmetics to be evaluated, determining whether the cosmetics to be evaluated cover the position area corresponding to the spatial skin abnormality information; If yes, then superimposed information is calculated and output; the superimposed information is a feature obtained by superimposing the skin abnormality information along the cosmetic application range and the spatial skin abnormality information at the same position, or the superimposed information is the skin abnormality information along the cosmetic application range; If not, the spatial skin abnormality information is output.

2. The method for quantitative evaluation of the antioxidant efficacy of cosmetics according to claim 1, characterized in that: Also includes: The output superposition information and the spatial skin abnormality information are aggregated to determine the skin abnormality information of the human body part.

3. The method for quantitatively evaluating the antioxidant efficacy of cosmetics according to claim 1, characterized in that: The data on the properties of the cosmetics to be evaluated when applied to a part of the human body is collected, specifically including: Collecting the state data and ingredient data of the cosmetics to be evaluated when applied to the human body, collected by the allergy sensor; The collection tracking unit collects application site data of the cosmetics to be evaluated when applied on a part of the human body.

4. The method for quantitative evaluation of the antioxidant efficacy of cosmetics according to claim 1, characterized in that: The collecting of the skin data of the body part when the cosmetic to be evaluated is applied to the body part specifically includes: The original skin information when the cosmetics to be evaluated are applied on the human body is collected by the infrared sensor, and / or the skin image data with makeup when the cosmetics to be evaluated are applied on the human body is collected by the camera.

5. The method for quantitative evaluation of the antioxidant efficacy of cosmetics on human body according to claim 1, characterized in that: The evaluation of the application order of the cosmetic to be evaluated specifically includes: Collecting the central application site data of the cosmetic to be evaluated when applied to the body part of the human body calculated by the tracking unit; Collecting the average value of the thickness and center color of the cosmetics to be evaluated when applied to the body part of the human body calculated by the fluorescence intensity analysis unit; Calculating the cosmetic sample application site data when the cosmetic to be evaluated is applied to the human body part based on the central application site data of the cosmetic to be evaluated, the cosmetic application thickness, the central color mean, and the skin perception current; The application order of the cosmetics to be evaluated is evaluated based on the cosmetic sample application site data.

6. The method for quantitatively evaluating the antioxidant efficacy of cosmetics on human body according to claim 1, characterized in that: When the skin data of a body part includes original skin information without makeup and skin image data with makeup, feature detection is performed on the skin data of the body part to obtain spatial skin abnormality information with makeup part data, specifically including: Performing denoising processing on the original bare-face information and the skin-with-makeup image data respectively; Perform feature detection on the original bare face information after denoising based on the spatial skin abnormality information detection model to obtain a first record; Performing feature detection on the denoised skin makeup image data based on the spatial skin abnormality information detection model to obtain a second record; At a preset position, when the first record stores free radical data antioxidant capacity information containing skin abnormality information and the second record does not store a depth image containing skin abnormality information, the free radical data antioxidant capacity information is processed to determine the spatial skin abnormality information with application site data corresponding to the preset position; the preset position is any position of the human body part; At a preset position, when the first record does not store free radical data antioxidant capacity information containing skin abnormality information and the second record stores a depth image containing skin abnormality information, the depth image is processed to determine the spatial skin abnormality information with application site data corresponding to the preset position; At a preset position, when the first record stores free radical data antioxidant capacity information containing skin abnormality information and the second record stores a depth image containing skin abnormality information, the free radical data antioxidant capacity information and the depth image are subjected to time-space synchronization processing to determine the spatial skin abnormality information with application site data corresponding to the preset position; The spatial skin abnormality information detection model includes a facial skin abnormality information detection sub-model, an upper limb skin abnormality information detection sub-model and a lower limb skin abnormality information detection sub-model; and the face, the upper limb and the lower limb are respectively input into the facial skin abnormality information detection sub-model, the upper limb skin abnormality information detection sub-model and the lower limb skin abnormality information detection sub-model to determine the probability values ​​of the face, the upper limb and the lower limb corresponding to each preset skin attribute in a plurality of preset skin attributes; Determine, according to the probability values, the original skin attribute information of the face, the upper limbs, and the lower limbs corresponding to the original skin without makeup or with makeup; For any preset skin attribute among the plurality of preset skin attributes, performing the following weighted sum operation: calculating the weighted sum of the probability values ​​of the face, the upper limbs and the lower limbs respectively on the preset skin attribute as the probability value of the user's face on the preset skin attribute; Determine skin attribute information of the user's face according to probability values ​​of the user's face at various preset skin attributes; The skin abnormality information detection model is a convolutional neural network model; The skin abnormality information detection model is trained using a plurality of images in which characteristic regions are divided and skin attribute annotations are added to the characteristic regions.

7. The method for quantitative evaluation of the antioxidant efficacy of cosmetics on human body according to claim 1, characterized in that: The calculating of the superposition information specifically includes: Using a superposition weighted formula, the skin abnormality information along the cosmetic application range and the spatial skin abnormality information at the same position are superimposed to obtain superimposed information; Alternatively, when the spatial skin abnormality information does not exist in the position area corresponding to the spatial skin abnormality information, the skin abnormality information along the cosmetic application range is determined as the superimposed information; Wherein, the weight value of the superposition weighted formula is determined according to the weight value formula; The weight value formula is a formula constructed by taking the skin site area deviation of the sample of the cosmetic to be evaluated and the temperature and humidity of the sample of the cosmetic to be evaluated as input; The regional deviation of the sample skin part is obtained by superimposing the deviation value of the fluorescence intensity analysis unit and the deviation value of the tracking unit; The deviation value of the fluorescence intensity analysis unit is a deviation value generated by the fluorescence intensity analysis unit when the fluorescence intensity analysis unit and the tracking unit jointly determine the application site data of the cosmetic sample when the cosmetic to be evaluated is applied to the body part of the human body; The deviation value of the tracking unit is a deviation value generated by the tracking unit when the fluorescence intensity analysis unit and the tracking unit jointly determine the application site data of the cosmetic sample when the cosmetic to be evaluated is applied to the body part of the human body; The output sample skin area deviation σ and sample temperature and humidity B are used as input to construct the weight value formula, and the weight value ξ of the superimposed weighted formula is calculated according to the weight value formula y and 1-ξ y , where ξ y is the weight value of skin abnormality information along the cosmetic application range, 1-ξ y is the weight value of spatial skin abnormality information; ； According to the calculated weight value β of the superposition weighted formula, the obtained abnormal information is extended along the y-axis range. In the section where the abnormal information covers the vertical x-axis range of point M, the coordinates of point M are (x1, 0, z1) and the coordinates of point N are (x1, y2, z2); ； The depth of each point of skin abnormality information is the formula of the horizontal and vertical coordinates corresponding to the point, that is, z = f(x, y); Construct the weighted formula of the superposition model according to the weight value: ； The coordinates in the above formula are all in the sample coordinate system; g(x,y) is the depth of each coordinate point of the visually detected skin abnormality information, f(x,y) is the depth of each coordinate point of the response detected skin abnormality information, and (x,y,z) is the coordinate of the skin abnormality information after superposition.

8. Use of the human quantitative evaluation method for the antioxidant efficacy of cosmetics according to any one of claims 1 to 7 in the evaluation of cosmetic effects.

Images