KR20190091857A - Method and system for personalized skin care based on skin diagnostic - Google Patents

Abstract

A personalized skin care method according to an embodiment of the present invention includes: a step of measuring the skin condition of a user; a step of allowing a skin diagnosis device to transmit a measured value to a device for providing the skin care method; a step of allowing the device for providing the skin care method to provide a skin care method corresponding to the measured value by using a classification model generated through machine learning.

Description

{Method and system for personalized skin care based on skin diagnostic}

The present invention relates to a method and system for personalized skin care through skin diagnosis. More specifically, the present invention relates to a method for inspecting a subject's skin, performing an analysis using machine learning, and a personalized skin care method based on the analysis result, and a system for performing the method.

Recently, with increasing interest in skin health care, devices and medical devices for skin diagnosis or skin care have been developed. A skin diagnostic device is a system for diagnosing abnormal areas of the skin, and includes a magnifying lens, a light source, a transparent plate in contact with the skin, and a liquid substance filled between the skin and the device to minimize reflections on the skin surface.

Recently, instead of using this liquid material, polarized light is used to remove the light reflected from the skin surface. The skin diagnostic device is manufactured in the form of a digital epiluminescence dermascope that can store or analyze an image or a video clip by mounting an image sensor in the form of simply checking with the human eye through the eyepiece.

Recently, with the rapid development of mobile devices such as smartphones and tablet PCs, an imaging system for mobile skin diagnosis / management in which a mobile device and this skin diagnosis device are combined is being developed. This mobile skin diagnosis / management imaging system is a miniaturized skin diagnosis imaging system that can be attached or linked to a smartphone or tablet.It can easily diagnose and manage the skin health of an individual at home, a small hospital, and a skin care center. It is a mobile healthcare imaging system that can quantitatively analyze various related skin diseases.

This mobile skin diagnostic / care imaging system is a compact form of high performance optics that enables sophisticated and complex optical imaging from simply attaching an optical lens. In addition, most of these systems have the ability to analyze and manage the skin image taken through the application linked with the system, and to transfer the image to a professional. Recently, various analytical technologies such as big data analysis and complex spectroscopic imaging analysis technology have been combined to increase the accuracy in skin diagnosis and management.

However, despite the development of such a skin diagnostic device, it does not yet provide a personalized skin care method. Accordingly, there is a growing need for a method that can provide a personalized skin care method.

The problem to be solved by the present invention is to provide a personalized skin care method and system through skin diagnosis.

The problem to be solved by the present invention is not limited to the problem (s) mentioned above, and other object (s) not mentioned will be clearly understood by those skilled in the art from the following description.

According to one or more exemplary embodiments, a method for providing a skin care method includes measuring, by a skin diagnosis device, a skin condition of a user; Transmitting, by the skin diagnosis device, the measured value to a device for providing a skin care method; And providing, by the device for providing skin care methods, a skin care method corresponding to the measured value using a classification model generated through machine learning.

Preferably, the step of measuring the skin condition of the user may include measuring at least one of brightness, skin texture, elasticity, moisture, oil, PH of the skin.

Preferably, the step of measuring one or more of the brightness, skin texture, elasticity, moisture, oil, PH of the skin, the skin diagnostic device, taking the skin of the user, by analyzing the color of the photographed image It may include the step of measuring the brightness of the skin.

Preferably, the step of measuring one or more of the brightness, skin texture, elasticity, moisture, oil content, pH of the skin, the skin diagnostic device, photographing the skin of the user, the uniformity of the color of the image taken Analyzing may include measuring the skin texture.

Preferably, the step of measuring one or more of the brightness, skin texture, elasticity, moisture, oil, PH of the skin, the step of measuring the elasticity by using a moiré image analysis to implement the contour of the elevation of the face can do.

Preferably, the step of measuring one or more of the brightness of the skin, skin texture, elasticity, moisture, oil, PH, by flowing a high frequency current, by measuring the electrical conductivity of the skin of the user to measure the moisture and oil It may include the step of measuring.

Preferably, the apparatus for providing a skin care method may further include learning a skin care method according to a skin condition by using a decision tree on a measurement value of a skin condition previously measured.

Preferably, the device for providing a skin care method may further include learning a skin care method according to a skin condition by using a support vector machine on a measurement value of a skin condition previously measured.

Preferably, the step of providing a skin care method corresponding to the measured value may include providing the user with a link for purchasing cosmetics corresponding to the skin condition.

Preferably, the step of providing a skin care method corresponding to the measured value may include providing a makeup method corresponding to the skin condition to the user in the form of a video.

In order to solve the above technical problem, the skin care method providing system according to another embodiment measures one or more of skin brightness, skin texture, elasticity, moisture, oil content, and pH to measure the skin condition of a user, A skin diagnosis device for transmitting the measured value to a device for providing a skin care method; And a skin care method providing device providing a skin care method corresponding to the measured value by using a classification model generated through machine learning.

By using the present invention, even if the user does not go to the dermatologist, it is possible to easily check the objective information on his or her skin condition, and further obtain product information suitable for each skin condition. In addition, the present invention may be provided with a personalized skin care method tailored thereto instead of simply completing the skin diagnosis.

Effects of the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a view for explaining a personalized skin care system through the skin diagnosis according to an embodiment of the present invention.
2 to 3 are views for explaining a personalized skin care method through the skin diagnosis according to an embodiment of the present invention.
4 is a flowchart illustrating a personalized skin care method through skin diagnosis according to an embodiment of the present invention.

As the present invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description. However, this is not intended to limit the present invention to specific embodiments, it should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. In describing the drawings, similar reference numerals are used for similar elements.

Terms such as first, second, A, and B may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component. The term and / or includes a combination of a plurality of related items or any item of a plurality of related items.

When a component is referred to as being "connected" or "connected" to another component, it may be directly connected to or connected to that other component, but it may be understood that other components may be present in between. Should be. On the other hand, when a component is said to be "directly connected" or "directly connected" to another component, it should be understood that there is no other component in between.

The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting of the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, the terms "comprise" or "have" are intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more other features. It is to be understood that the present invention does not exclude the possibility of the presence or the addition of numbers, steps, operations, components, components, or a combination thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art. Terms such as those defined in the commonly used dictionaries should be construed as having meanings consistent with the meanings in the context of the related art, and shall not be construed in ideal or excessively formal meanings unless expressly defined in this application. Do not.

Hereinafter, with reference to the accompanying drawings, preferred embodiments of the present invention will be described in detail.

1 is a view for explaining a personalized skin care system through the skin diagnosis according to an embodiment of the present invention.

Referring to FIG. 1, the skin care system proposed by the present invention is divided into a skin diagnosis device 100 and a care method providing device 200. The skin diagnosis apparatus 100 is a device for diagnosing a skin condition of a user, and includes a sensor for measuring brightness, skin texture, elasticity, moisture, oil content, pH, and the like of the skin. The care method providing device 200 provides a personalized skin care method based on the measured value.

Specifically, the camera module is built in to measure the brightness of the skin. In addition, a light source module can be built in to assist in shooting. In order to measure the brightness of the skin, a light source may be irradiated to the skin, and the brightness may be measured by taking an image of a portion to which the light source is irradiated. In addition, an image taken to measure the brightness of the skin may be used to measure the skin texture. Skin texture can be measured based on the uniformity of color in the image. The more uniform the color, the clearer the skin.

In order to measure the elasticity of the skin, the moiré image analysis method using the interference phenomenon of light to implement the height difference of the face in the contour line can be used. The light source may be irradiated to the face using the light source module described above, and the elasticity of the skin may be measured based on the interference phenomenon.

And to measure the moisture and oil of the skin can use the electrical conductivity of the skin. The electrostatic load capacity of the skin is directly related to the amount of moisture. That is, the moisture content and the electrostatic load capacity tend to be proportional to each other, so that the higher the moisture level of the skin is, the higher the value can be used to measure the moisture of the skin. To this end, the skin diagnosis apparatus 100 includes two electrodes. By measuring the resistance by flowing a current from one electrode and receiving a current on the other side, it can be converted into a moisture content and an oil content.

The stratum corneum is composed of brick and cement-like structures. The keratinocytes, which act as bricks, have moisture in the cells through natural moisturizing factors, and lipid components surround the keratinocyte membrane and the keratinocytes. Because there is water in the cells, the current in MHz is used to measure skin moisture because a high frequency current must flow through the keratinocytes. Then, the amount of moisture in the stratum corneum can be measured by setting the depth at which the current is transmitted to 10-20 占 퐉 corresponding to the thickness of the stratum corneum.

In addition, the skin diagnosis apparatus 100 may include a PH sensor to measure the pH of the skin. In addition, image analysis can measure skin blemishes and blemishes. It is also possible to measure the size of the pores and the temperature of the skin. In the present invention, the brightness of the skin, skin texture, elasticity, moisture, oil, PH in order to help the understanding of the invention, but this is not intended to limit the invention.

After the information indicating the condition of the skin is measured by the skin diagnosis device 100, the measured value is transmitted to the management method providing device 200 through wired / wireless. In the example of FIG. 1, the management method providing device 200 is illustrated in the form of a smartphone. However, in addition to a smartphone, various wearable devices may be used as the management method providing device 200.

The management method providing device 200 receives the measured value and performs an analysis based on the measured value. In this process, analytical models trained through machine learning can be used. For example, a training model generated by applying supervised learning may be used. For example, a decision tree, a support vector machine (SVM), or the like can be used. This will be described in more detail with reference to FIGS. 2 and 3.

2 to 3 are views for explaining a personalized skin care method through the skin diagnosis according to an embodiment of the present invention.

Decision tree learning uses decision trees as a predictive model that links observations and targets to an item. This is one of the predictive modeling methods used in statistics, data mining, and machine learning. In the tree model, a target tree having a finite number of values is called a classification tree. In this tree structure, the leaf (leaf node) represents the class label and the logical product of the features associated with the class label. In a decision tree, a continuous value of a target variable, generally a real number, is called a regression tree.

In decision analysis, decision trees are used to show the decision making process and decision making in a visual and explicit way. In the field of data mining, decision trees are used to represent the data itself rather than the decision made. However, the classification tree as a result of data mining may be used as an input value of decision analysis. This page focuses on decision trees in the field of data mining.

Decision tree learning is a methodology commonly used in data mining that aims to create models that predict the value of target variables based on some input variables. In the decision tree structure, each internal node corresponds to one input variable, and the branches leading to the child nodes correspond to possible values of the input variable. The leaf node corresponds to the target variable value when each input variable has values corresponding to a path from the root node to the leaf node.

Decision tree learning is one of the most useful techniques in supervised classification learning. Each element of the class's domain is called a class. Input attributes correspond one-to-one to all internal nodes of the decision tree or classification tree. Branches connected to internal nodes of the tree show values that the property can have, and leaf nodes show the class or probability distribution of classes.

The "learning" of the decision tree is the process of dividing the dataset used for learning into subsets according to appropriate partitioning criteria or partitioning tests. This process is recursively repeated for each divided subset of data in a manner called circular partitioning, and continues until the partition no longer adds new predictive values or the nodes in the subset have the same value as the target variable. . This top-down induction of decision trees (TDIDT) is an example of the greed algorithm and is the most common way to learn a decision tree from data.

Referring to FIG. 2, a decision tree based on the brightness of the skin and the oil fraction of the skin is shown. By comparing the brightness of the skin and the threshold c1, and again comparing the oil content of the skin and the threshold c2, in each case, a decision tree recommending a soap suitable for the user can be viewed.

However, this is an example of a decision tree based on the brightness of the skin and the oil content of the skin to aid in the understanding of the invention, and actually generates a decision tree based on the brightness, skin texture, elasticity, moisture, oil content, and pH of the skin. can do. In addition, various products such as cleansing, makeup, cosmetics, soaps, and mask packs and skin care methods can be derived from the decision tree.

In the case of the face-washing method, a crystal tree can be generated based on the moisture and oil of the skin. Depending on the moisture and oil of the skin, it can be divided into neutral skin, oily skin, dry skin, and complex skin, and the washing method is different for each skin. Therefore, by learning these features, we can suggest personalized face washing method according to the moisture and oil of skin.

For example, neutral skin is the most ideal skin compared to other skin types. Neutral skin is more active than other skin types, and the face is more moisturized, making your face look moist. It is a skin that has the advantage that the makeup is eaten well because makeup is kept well, and makeup does not get erased well and trouble does not easily occur on the skin even during daily life. If you take good care of your skin, you may have a very good skin condition, but it is especially important when your skin is very dry or water evaporates.

To clean your normal skin, cleanse your skin with a cleansing product specifically for neutral skin. Neutral skin, like other skins, should not be washed with hot water and strong soap. Neutral skin is rich in oil, so if you wash your face with hot water and a strong soap, it will reduce the amount of oil. Neutral skin should be rinsed with lukewarm water and weak soap, and finished with lotion. Otherwise, you can return to dry skin, so care must be taken.

Dry skin has a low secretion of sebum, which can result in a lack of gloss on your face compared to other skin types. It is the hardest to manage compared to other skin types because it has less grease in spring or summer. Because of its lack of luster, the skin is prone to crumbs and fine lines and blemishes appear. Dry skin can cause dead skin cells. When washing your face, your skin may be pulling or itching may occur. Therefore, it is most important to keep your skin dry. Use a moisturizing product to keep your skin hydrated and moisturized. Even when you exfoliate, it's a good idea to keep the oil to some extent, rather than keeping the oil and fat properly.

Dry skin has a low secretion of sebum and lacks moisture, so it should be moisturized and managed to balance moisture and oil. For dry skin, it's enough to wash your face with water instead of using it. Due to the low oil content, cleansing with foam cleansing may cause oil to run out. Dry skin is less secretion of sebum, so it may cause skin problems with residues of dead skin and cosmetics. After cleansing, care should be taken to avoid pulling the skin with products that have excellent skin moisturizing power.

Oily skin has a large amount of sebum secretion, wide pores and thick stratum corneum. Oily skin has a lot of sebum and oil in the face, so make-up is easy to remove. Since sebum is secreted a lot, bacteria grow and acne and blemishes tend to occur. Poor care can lead to rough pores and skin problems.

Oily skin is oily and sensitive, so irritating your skin can ruin your skin. When removing the makeup, remove the makeup using lotion or cleansing tissue, and use a lot of foam to massage the face. When washing your face, oily skin should not be washed with hot water. Since the pores are already wide, washing with hot water will generate more waste, which can stimulate the pores and produce more oil, so wash your face with lukewarm water. For oily skin, apply lotion only after cleansing. Applying a variety of products can be counterproductive, so it is enough to use only one moisture cream with low oil content.

Combination skin is a condition in which the secretion of sebum is uneven and the balance is broken, the condition of the skin is not uniform, and the skin condition is a combination of dry and neutral oily skin. Wearing makeup or being affected by the weather can cause the skin to lose balance and cause skin problems. For complex skin, apply a lotion or moisture essence that contains less moisturizing oil around the eyes and around the mouth.

Combination skin Because it is a skin with several types of skin, you should wash your face twice more carefully than other skin types. If you wash your entire face with oily cleansing product because the T-zone area is greasy, your cheeks are dry and you are likely to exfoliate. After washing your face gently, wash your face once more. Combination skin is important for proper moisture and oil supply, so pack it.

As such, according to the moisture and oil of the skin, the lotion method and the moisturizing product to be used after washing may be different. In the present invention, the decision tree is learned based on the measured value, and the facial cleansing and moisturizing products suitable for the skin condition of the user are determined according to the decision tree. I can suggest. In particular, by providing the user with the face-wash method through the content such as video can provide a more personalized face-wash method.

In the case of makeup methods, the appropriate makeup method and cosmetics are divided according to the brightness, skin texture, elasticity, moisture, oil content, and pH of the skin. Similarly, by learning a decision tree and suggesting cosmetics and cosmetics for a user's skin condition according to the decision tree, a more personalized skin care method can be provided.

As described above, although a manager or a medical staff provides a customized consulting to a user by referring to a value measured by the skin diagnosis apparatus 100, in the present invention, a learning model is generated in advance through machine learning and automatically based on the measured value. There is an advantage in providing a personalized skin care method.

In addition to decision trees in machine learning, support vector machines (SVMs) are available. Support vector machine is one of the field of machine learning. It is a supervised learning model for pattern recognition and data analysis. It is mainly used for classification and regression analysis. Given a set of data in either category, the SVM algorithm creates a non-probability binary linear classification model that determines which category the new data belongs to based on the given data set.

The classification model is represented as a boundary in the space where the data is mapped. The SVM algorithm finds the boundary with the largest width. SVM can be used in nonlinear classification as well as linear classification. Nonlinear classification requires mapping the given data into a higher-dimensional feature space, which also uses kernel tricks to do this efficiently.

Referring to FIG. 3, it can be seen that data is divided into two sets by applying a support vector machine in a two-dimensional space using two features. In this way, if a model is generated by applying a support vector machine to classify the data in the machine learning process, a training model may be applied to classify which set of data to be measured later.

Applying the support vector machine in terms of the characteristics of various skin conditions measured by the skin diagnosis apparatus 100 can be divided into two areas, and skin care methods suitable for each area can be recommended. have. This can automatically provide personalized skin care methods.

For example, it recommends soaps, mask packs, cosmetics, etc., suitable for dry skin with low oil content and high acidity, and can provide facial cleansing and make-up methods using the smartphone as a video. Alternatively, it is recommended that soaps, mask packs, cosmetics, etc. are suitable for oily and neutral skin with a lot of oil, and a face washing method and make-up method using the same may be provided through a smartphone as a video.

You can even generate sales by providing links to buy products that fit your skin. Through this, it is possible to relay products between users and sellers by providing an environment where users can purchase products beyond simply providing information, and through this, they can continuously provide free services to users.

4 is a flowchart illustrating a personalized skin care method through skin diagnosis according to an embodiment of the present invention.

Referring to FIG. 4, first, a skin condition of a user is measured through the skin diagnosis apparatus 100 (S1100). The indicators measured here are indices such as skin brightness, skin texture, elasticity, moisture, oil, and pH. The measured values are transmitted to the management method providing device 200 via wired / wireless.

The management method providing device 200 applies a classification model trained in advance (S1200) to provide a personalized skin care method according to the measured index (S1300). In this process, the learning model can be created using a decision tree or a support vector machine. In addition to the measurement value of the skin condition measured by the skin diagnostic device 100 receives additional information about the age, height, gender, etc. from the user can utilize this information as a feature to perform machine learning have.

So far I looked at the center of the preferred embodiment for the present invention. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in descriptive sense only and not for purposes of limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope will be construed as being included in the present invention.

Claims (11)

Measuring, by the skin diagnosis device, the skin condition of the user;
Transmitting, by the skin diagnosis device, the measured value to a device for providing a skin care method; And
Providing, by the device for providing skin care methods, a skin care method corresponding to the measured value by using a classification model generated through machine learning;
How to provide skin care. The method of claim 1,
Measuring the skin condition of the user,
Measuring at least one of skin brightness, skin texture, elasticity, moisture, oil content, and pH,
How to provide skin care. The method of claim 2,
Measuring at least one of the brightness, skin texture, elasticity, moisture, oil, PH of the skin,
Wherein the skin diagnosis device, photographing the skin of the user, and analyzing the color of the photographed image to measure the brightness of the skin,
How to provide skin care. The method of claim 2,
Measuring at least one of the brightness, skin texture, elasticity, moisture, oil, PH of the skin,
The skin diagnosis device, comprising the steps of photographing the skin of the user, and measuring the skin texture by analyzing the uniformity of the color of the photographed image,
How to provide skin care. The method of claim 2,
Measuring at least one of the brightness, skin texture, elasticity, moisture, oil, PH of the skin,
Measuring the elasticity by using a moiré image analysis method for realizing the height difference of the face in a contour line,
How to provide skin care. The method of claim 2,
Measuring at least one of the brightness, skin texture, elasticity, moisture, oil, PH of the skin,
Flowing a high frequency current, by measuring the electrical conductivity of the skin of the user comprising the step of measuring the moisture and oil,
How to provide skin care. The method of claim 1,
The skin care method providing device further comprises the step of learning the skin care method according to the skin condition by using the decision tree on the measured value of the skin condition in advance,
How to provide skin care. The method of claim 1,
The skin care method providing device further comprises the step of learning the skin care method according to the skin condition using a support vector machine to the measured value of the skin condition in advance,
How to provide skin care. The method of claim 1,
Providing a skin care method corresponding to the measured value,
Providing the user with a link to purchase cosmetics corresponding to the skin condition,
How to provide skin care. The method of claim 1,
Providing a skin care method corresponding to the measured value,
Providing a makeup method corresponding to the skin condition to the user in the form of a video,
How to provide skin care. A skin diagnosis device for measuring one or more of brightness, skin texture, elasticity, moisture, oil content, and pH of the skin to measure a skin condition of the user, and transmitting the measured value to a device for providing a skin care method; And
And a skin care method providing device providing a skin care method corresponding to the measured value by using a classification model generated through machine learning.
Skin care method providing system.

Images