RU2470576C2 - Method of automatic assessment of skin and/or wrinkle texture - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: group of inventions relates to cosmetology. Method includes stages at which processing of skin image is performed, which includes cross conversion to binary form and/or processing by comparison of short linear segments to obtain 10 or more types of physical parameters, related to dermal furrow; then physical parameters, obtained at previous stage are substituted into prediction formula. Prediction formula is determined by multi-variant analysis of 10 or more types of physical parameters, related to dermal furrow, and visual assessment values, distributes by categories from third to tenth degree, related to skin and/or wrinkle, obtained from a number of samples in order to obtain assessment value by categories from 3 to 10 degree and assessment of obtained assessment value as assessment result with respect to skin and/or wrinkle texture. Described are device for assessment of skin texture and method of selecting medication for external application on skin, which uses assessment method or device.

EFFECT: providing assessment of epidermal tissues.

7 cl, 11 dwg, 10 tbl

Description

FIELD OF THE INVENTION

The present invention relates to a technology for assessing skin condition, and more particularly to a technology for evaluating skin texture and / or wrinkles using physical parameters of the skin as an indicator.

State of the art

Important facts when using cosmetics are the appropriate selection and use of cosmetics suitable for the user and, at the same time, ensuring the effect of their use. The use of cosmetics that are not suitable for the user is not only unable to provide their effect, but can also cause undesirable effects. Therefore, the wrong choice of cosmetics that are not suitable for the skin of the user should be avoided. Such complaints as “this cosmetic product is not suitable for my skin” are often heard. However, such a phenomenon often arises because a cosmetic product suitable for a given skin has not been selected, i.e. in other words, a cosmetic product that is not suitable for the skin was mistakenly selected. An important factor in choosing cosmetics suitable for a given skin and to ensure the effect of the application is the assessment of skin texture and wrinkles.

In connection with the above, a variety of technologies have been developed for the selection of cosmetics. For example, a technology is described, including the isolation and analysis of dermal furrow samples obtained by irradiating skin prints (see Patent Document 1), a technology including image analysis with respect to a direct image of the skin surface (see Patent Document 2), a technology including processing imaging, such as gradation processing in relation to the imprint of the image (see Patent Document 3), a technology including measuring wrinkle depth from an imprint using electromagnetic waves (see Patent Document 4), a technology including measuring wrinkle depth from skin imprint material using a standard scale (see Patent Document 5), a technology including performing line thinning processing for a single-color pixel image or processed pixel state skin image to measure through this texture using a thin line as a peak interval indicator (see Patent Document 6) and the like. However, the main problems in the known technologies for choosing cosmetics are: are the results of the evaluation of the texture of the skin and / or wrinkles consistent with the visual estimates related to the texture and / or wrinkle; and, besides, is it possible to conduct a quick assessment at points of sale and consultations, in other words, compatibility between objectivity and speed of assessment is required. That is, although there are evaluation criteria, serious problems are: 1) the fact that visual estimates related to texture and / or wrinkles are determined subjectively by the observer by visual inspection of the print or by the observer with the naked eye under a microscope, and therefore it is difficult to achieve objectivity; 2) the long period of time required for evaluation. Therefore, it was desirable to develop an evaluation technology that is objective and quick and can provide an assessment of skin texture and / or wrinkles.

Under such conditions, image analysis technology carried out in the usual and traditional way, such as filter processing, binary conversion processing, image signal processing or comparison processing, cannot sufficiently extract and quantify the characteristic information about the patterns of dermal furrows and dermal protrusions . Therefore, the applicants of the present invention have developed a completely new technology for extracting target information from such images obtained from a subject, i.e., quantification technology using binary cross-conversion processing and matching of short linear segments (see Patent Document 7).

List of links

Patent documents

Patent Document 1: JP 60-053121 A

Patent Document 2: JP 64-059145 A

Patent Document 3: JP 02-046833 A

Patent Document 4: JP 08-145635 A

Patent Document 5: JP 2000-342556

Patent Document 6: JP 2006-061170 A

Patent Document 7: JP 2008-061892 A

SUMMARY OF THE INVENTION

Technical problem

The present invention was developed in connection with the above circumstances, the purpose of the present invention is to provide technology for assessing the texture of the skin and / or wrinkles, providing a quick and highly accurate assessment of the condition of the skin and / or wrinkles. Another objective of the present invention is to provide a method for selecting a preparation for external application to the skin based on the evaluation results.

Solution

The applicants of the present invention have conducted extensive exploratory studies of a method for evaluating the texture of the skin and / or wrinkles, using which it is possible to quickly and accurately evaluate the texture of the skin and / or wrinkles. As a result, the applicants found that the texture of the skin and / or wrinkles can be quickly estimated with high accuracy by obtaining an estimated value calculated by substituting in a pre-prepared formula for predicting the physical parameters of the skin obtained by processing the image of the skin by cross-conversion into binary form and / or processing by comparing short linear segments, and thus, the present invention was created. That is, the present invention relates to the following technologies.

(1) A method for evaluating skin texture and / or wrinkles, comprising the steps of: processing a skin image by performing binary cross-conversion and / or processing by matching short linear segments to obtain physical skin parameters; and substitute the physical parameters obtained in the previous step in a pre-prepared forecast formula to obtain an estimated value and evaluate the obtained estimated value as an estimated result in relation to skin texture and / or wrinkles.

(2) A device for evaluating skin texture and / or wrinkles, including: means for inputting a pre-prepared forecast formula; means for capturing images of the skin; means for calculating the physical parameters of the skin from the captured image of the skin; means for calculating the estimated value related to the texture of the skin and / or wrinkle according to a previously prepared forecast formula and calculated physical parameters of the skin; and means for displaying the calculated estimated value.

(3) A program for evaluating the skin, which allows the computer to function as: a means for calculating the physical parameters of the skin from a captured image of the skin; and means for calculating the estimated value related to the skin texture and / or wrinkle, according to a previously prepared forecast formula and calculated physical parameters of the skin.

(4) A method for selecting a preparation for external application to the skin, comprising the steps of: evaluating the texture of the skin and / or wrinkles using the evaluation method described in paragraph (1), or the evaluation device described in paragraph (2) ; and choosing a preparation for external application to the skin containing a component to improve the texture or to prevent deterioration of the texture when the evaluation indicates that the condition of the skin texture and / or wrinkles in the test subject is unsatisfactory, or choose a preparation for external application on the skin containing only a moisturizing component, in the case when the evaluation result indicates that the condition of the skin texture and / or wrinkles in the test subject is good, each of which is based on the estimated value related to the texture of the skin and / or wrinkle, estimated at the stage of evaluation.

The technical result of the invention

In accordance with the present invention, it is possible to provide a technology for evaluating the texture of the skin and / or wrinkles, which can quickly and accurately evaluate the texture of the skin and / or wrinkles. In addition, this technology can be used to provide a formulation suitable for the user to be applied to the outer surface of the skin.

Brief Description of the Drawings

Figure 1. Photo showing the image brightness of the print (photo replacing the picture).

Figure 2. A photograph illustrating the method of processing by cross-conversion to binary form (photograph replacing the picture).

Figure 3. A photograph illustrating the image of the fingerprint obtained by cross-conversion to binary form (photograph replacing the picture).

Figure 4. A photograph showing the processing method by comparing short linear segments (photograph replacing the figure).

Figure 5. A photograph illustrating the image of the fingerprint obtained by processing the comparison of short linear segments (photograph replacing the picture).

6. Photographs showing standard images of the skin texture (left: estimated value 1; right: estimated value 5) (photos replacing drawings).

7. Photographs, each of which shows a photograph of the standard image of wrinkles (left: estimated value 1; right: estimated value 3) (photographs replacing drawings).

Fig. 8. Photos, each of which shows the area of the dermal furrows, which is a physical parameter after processing by the method of cross-conversion to binary form and comparison of short linear segments (photo replacing the picture).

Fig.9. Photographs, each of which shows the average width of the dermal furrows, which is a physical parameter after processing by the method of cross-conversion to binary form and comparison of short linear segments (photo replacing the picture).

Figure 10. A diagram illustrating an example configuration of a device for evaluation.

11. A flowchart related to processing by an evaluation device.

Description of Embodiments

The evaluation method of the present invention can evaluate the state of the skin texture, the condition of skin wrinkles, or both according to the image of the skin.

Capturing an image of skin to be used in the present invention

The present invention uses an image of the skin. A method for capturing a skin image may be a method for acquiring a skin image by directly acquiring a skin image, or a method for acquiring a skin image through a skin imprint sample. The image capturing method may include, for example, capturing an image using a digital video camera through a stereo microscope, or using a commercially available digital microscope. Examples of a digital microscope include a cosmetology microscope manufactured by MORITEX and a digital microscope manufactured by KEYENCE.

In the present invention, it is preferable to capture an image of the skin by means of the aforementioned fingerprint sample obtained from the skin. When a skin image is captured by a print pattern obtained from the skin, only drawing information can be captured, excluding color information about the skin surface, thereby avoiding interference such as freckles. In addition, irregularities on the surface of the skin that are not needed for analysis (larger than irregularities due to dermal grooves or dermal protrusions) are eliminated by imprinting, and therefore, the analysis is easily performed. A method for capturing an image of a skin through a print sample is described below.

When the microscope lens is positioned at an angle of 90 ° relative to the fingerprint sample to lightly irradiate the fingerprint sample at an appropriate angle, the image of a shadow cast by irregularities on the surface of the fingerprint sample formed by incident light can be captured as an image through a microscope. In this case, the fingerprint sample refers to the technology of indirect inspection of irregularities on the skin by applying a solvent in the form of a coating on a transparent plastic plate in order to soften it, then pressing the softened part to the skin to display the unevenness of the skin on the softened part and inspecting the unevenness. The Kawai Method is known as a typical technology. This technology has been widely used in the fields of cosmetology, because this technology can reliably display bumps on the skin and can provide excellent preservation of the properties of bumps. Some kits for preparing such a fingerprint sample have been released by the industry and can be used. Such a fingerprint sample is preferably taken from the cheek or from the corner of the eyes in their lower part (1.5 cm × 1.5 cm). The general method includes irradiating the surface of the print of such a sample with light from a lower surface perpendicular to it, and observing the transmitted light. That is, irregularities are observed in the form of an image using the phenomenon of the amount of transmitted light reduced by the scattering of the emitted light due to the displayed irregularities. In the evaluation method of the present invention, the fingerprint is preferably inspected as follows. That is, the surface of the imprint on which there are irregularities is rotated in the direction of the image acquisition direction, and the light is emitted at an angle of 10 to 40 degrees or more, preferably 20 to 30 degrees relative to the surface, followed by obtaining an image formed by reflected light ( capture in the form of an image). When this approach is adopted, the irregularities displayed on the surface of the print more clearly appear as a difference in light intensity. Table 1 shows the clarity of the estimates obtained by observing an identical sample at various angles of incidence. Evaluation criteria were those shown below: o: transparent; Δ: slightly opaque; and x: opaque.

Table 1 The angle of incidence of light (angle of elevation) Clear unevenness 10 degrees From o to Δ 25 degrees ABOUT 40 degrees From o to Δ

Cross binary conversion as image processing

In the present invention, image processing, including cross-conversion to binary and / or processing by matching short linear segments, is performed for the above captured image of the skin. Image processing is described in JP 2008-061892 A (Patent Document 7) and is described below.

As the most fundamental image processing method, a binary conversion processing method is provided, comprising separating the target from the background in the image and extracting the target in the form of a shape. In the case where the contrast between the target and the background is sufficient, binary conversion processing can be easily performed. However, in reality there are minor shading changes mainly in the boundary between the target and the background and, therefore, it is difficult to determine the threshold value for processing by binary conversion in order to extract the shape with high accuracy. In addition, in the case where the background level of shading fluctuates due to fluctuations in lighting, it is difficult to accurately derive the shape using a threshold value fixed on the entire display. In such a case, processing using a dynamic threshold value (processing using a variable threshold value) using different threshold values for different pixels instead of a fixed threshold value is preferably applied, and the method for cross-binary processing in the present invention relates to a processing method using dynamic threshold value. In general, the processing area of the processing method using the dynamic threshold value has a rectangular shape, while the processing area of the processing method by cross-conversion to binary form has a transverse shape suitable for deriving the shape of the dermal sulcus (see FIG. 2). When applying the method of processing by cross-conversion to binary form, the shadow formed by the convex parts in the dermal sulcus can be detected without exposure to fluctuations in the illumination that irradiates the imprint, through this uniform and high-precision image obtained by cross-conversion to binary form (see Fig. 3) from the entire display, including the dermal furrows, from deep and distinct to small dermal furrows.

It should be noted that the above processing using cross-conversion to binary form can be carried out using the device for the quantification of epidermal tissue described in JP 2008-061892 A.

Matching short linear segments as image processing

A method for comparing short linear segments is a method for calculating the physical parameters of a target picture in an image obtained by converting to binary form. In the traditional method, the number of pixels of the target is calculated based on the pixels of the image obtained by converting to binary form to calculate physical parameters such as area, length and center of gravity. On the other hand, in the method of matching short linear segments, the physical parameters are calculated based on a short line formed from a plurality of pixels (having a length of from several pixels to several tens of pixels and a width of one pixel). In particular, the end point of the target area is defined as the starting point of the short line, and when the end point of the short line is within the target area, the end point is determined as another starting point and connected to the next short line. When the endpoint of the short line is not within the target area, the connection is completed. The procedure is repeated until the target area is covered with short line segments. Then, the following are measured: the number, angles, and the like of the characteristics of the short line segments in the target region for calculating the characteristic parameters of the target object (see FIG. 4). In accordance with this method, by comparing short linear segments, it is possible to obtain an image having the characteristics of dermal grooves that are narrow, continuous and directional (see FIG. 5).

It should be noted that the above processing by matching short linear segments can be carried out using the device for the quantitative determination of epidermal tissue described in JP 2008-061892 A.

Any of the above image processing methods may be performed. However, when both image processing methods are performed, the physical parameters can be calculated with greater accuracy. In addition, if necessary, other image processing methods can be performed, such as processing by converting the brightness of illumination, processing by conversion to binary form, processing by filtering, general image processing (area, border length, aspect ratio, center of gravity, needle ratio, magnification and inversion).

Calculation of physical parameters of texture and / or wrinkles

In the present invention, the physical parameters of the image of the skin can be obtained by image processing, including the aforementioned processing by cross-conversion to binary form and / or processing by matching short linear segments. Physical parameters are obtained by quantitatively determining the characteristics of dermal furrows, dermal protrusions, and similar structural properties of the skin. Examples of physical parameters include physical parameters such as the area of the dermal furrows, the average width of the dermal furrows, the change in the width of the dermal furrows, the gap between the dermal furrows, the parallelism of the dermal furrows, the direction of the dermal furrows and the density of the dermal furrows; and more detailed physical parameters, such as the maximum number of short linear segments at 95 ° or more of the number of segments for different angles, the maximum number of short linear segments at 10 ° or more and 90 ° or less of the number of segments for different angles, the maximum number of short linear segments from the numbers of segments for different widths, the width at the maximum number of short linear segments from the numbers of segments for different widths, the total value of the data of the frequency of the connected number of short linear segments and the total value in widths of short linear segments from the number of segments for different widths. In the present invention, physical parameters are calculated from physical quantitative parameters that are considered to be deeply related to texture and wrinkle.

In particular, physical parameters are defined as follows:

dermal furrow area = occupied area of dermal furrows in the image range to be processed as a target, or the total number of matching short linear segments;

the average width of the dermal sulcus = (the sum of the widths of the dermal sulci of various coincident starting points / total number of starting points);

change in the width of the dermal furrows = standard deviation or dispersion calculated from the histogram of the widths and the number of widths of the dermal furrows;

the average interval between the dermal furrows = 1 / (area of the dermal furrow / average width of the dermal furrow);

concurrency of dermal furrows = degree of concentration or dispersion of peaks calculated from the histogram of the angle and the number of dermal furrows; and

direction and density of dermal furrows = number of short linear segments at an angle θ (histogram height) / total length of dermal furrows.

Other physical parameters can be appropriately calculated using the above calculation formulas. As described above, there are many physical parameters obtained in the present invention, and from these physical parameters are selected that are preferred for calculation according to the prediction formula described below. The calculation of such physical parameters can be carried out by a computer program, including the above image processing, such as processing by cross-conversion to binary form and / or processing by comparison of short linear segments.

Prediction formula

To evaluate the texture of the skin and / or wrinkles, a prediction formula is determined in advance that shows the relationship between the above physical parameters of the skin and the visual estimate related to the skin texture and / or wrinkle. The forecast formula can, for example, be created in the following way.

Imprints of the skin obtained by taking into account the characteristics of the skin condition, age and the like (hereinafter referred to as samples) are subjected to a visual assessment of the texture of the skin and / or wrinkles by the evaluators. On the other hand, the physical parameters of the skin of the samples are calculated as described above. The number of samples is preferably 100 or more, or more preferably 500 or more. In a visual assessment of skin texture and / or wrinkles, samples are determined by a plurality of (preferably five or more) evaluators suitable for representing a third party with respect to standard photographs of the third to tenth degree for assessing textures ranging from good to bad, or the number of wrinkles from small to large, for the purpose of showing, by means of this, evaluation criteria corresponding to standard photographs. The evaluators mentioned above, suitable for presenting the third moan, are preferably persons with at least one year of experience in research on beauty, aesthetics or determination of skin condition and constantly train their skills in determining skin condition. The maximum value and the minimum value of the evaluation criteria for each sample are excluded, and their average value is calculated as a visual evaluation value related to the texture and / or wrinkle of the sample.

It should be noted that the degrees of skin texture and wrinkles to be evaluated as skin conditions were discussed at the Society of Cosmetic Pharmacists of Japan and the International Federation of Societies of Cosmetic Pharmacists (IFSCC), and recognition of these degrees is widely divided as the degrees of assessment of skin texture and wrinkles that may be objectively recognized by a third party. Figures 6 and 7 show examples of standard photographs of texture (five-degree grade) and wrinkles (three-grade grade), which are standardized based on statistical processing. In the technical field, such standard photographs can be judged to be somewhat reliable if the aggregate, which serves as the basis for the creation, includes approximately 100 photographs. If the collection includes more than 1000 photographs, then standard photographs can be rated as very reliable, and there is no need to take into account the differences between standard photographs. In the above sample evaluation, standard photographs standardized on the basis of such statistical processing can be used in the present invention, and the population when creating standard photographs preferably includes 1000 or more photographs.

Subsequently, the forecast formula is determined by the thus estimated evaluation criteria for texture and / or wrinkles and the calculated physical parameters of the skin. With respect to the prediction formula, the formula obtained by multivariate analysis of the physical parameters of the skin and visual evaluation criteria related to texture and / or wrinkle can be defined as the prediction formula of the present invention. Multivariate analysis is preferably analysis based on the relationship between the explanatory variable and the objective function variable, and its examples include discriminant analysis, analysis of the main components, factor analysis, quantitative theory I, quantitative theory II, quantitative theory III, regression analysis (MLR (limits of maximum remainder), PLS (partial minimum quadratic), PCR (regression of the main component) and logistics), multidimensional scaling, controlled clustering, neural network and group study. In addition, a forecast formula can be created using free computer software and a commercially available product. Of these, particularly preferred are: multiple regression analysis, discriminant analysis, and quantitative theory I. For example, it is preferable that multiple regression analysis be performed using physical parameters of the skin as explanatory variables and using visual evaluation criteria related to texture and / or wrinkles calculated above as variables of the objective function for determining, by this, the multiple regression equation, and the multiple regression equation divided as a forecast formula.

As described above, the physical parameters of the skin to be used to calculate the prediction formula include a variety of physical parameters. However, from the point of view of increasing the accuracy of the estimation method of the present invention, the physical parameters preferably include a physical parameter related to the dermal sulcus, and from the point of view of further improving the accuracy of the estimation method, the physical parameters preferably include 10 or more kinds of physical parameters related to the dermal sulcus. The total number of physical parameters to be used to calculate the forecast formula is preferably 10 or more.

Evaluation Method

When the above physical parameters of the skin are substituted into the prediction formula defined as described above to obtain evaluation criteria, it is possible to evaluate the texture of the skin and / or wrinkles. When the physical parameters of the skin, calculated from the captured images, are substituted into the forecast formula, visual evaluation criteria can be obtained relating to the skin texture and / or wrinkle. In accordance with the present invention, it is possible to very accurately evaluate the texture of the skin and / or wrinkles through the above steps. In addition, physical parameters, visual evaluation criteria, and similar characteristics of the new sample are included in the database. When the database is updated and corrected, the accuracy of the forecast formula is further improved, which is expected to allow an accurate assessment.

Device and evaluation program

In addition, another aspect of the present invention is a program for carrying out the above steps. That is, the program is a program for evaluating the skin, which is used to enable the computer to function as a means for calculating physical parameters from captured skin images and as a means for calculating assessment criteria related to skin texture and / or wrinkle, according to a previously prepared forecast formula and calculated physical parameters of the skin. The evaluation program of the present invention can be used by installing the program in equipment such as a personal computer.

In addition, another aspect of the present invention is an evaluation device for carrying out the above steps. That is, the device is a device for assessing the texture of the skin and / or wrinkles, including a means for entering a previously prepared forecast formula, a means for obtaining a skin image, a means for calculating the physical parameters of the skin from the captured skin image, a means for calculating an estimated value, related to the texture of the skin and / or wrinkle, according to a previously prepared forecast formula and calculated physical parameters of the skin, and means for displaying the calculated estimated value.

One aspect of the above evaluation apparatus is described with reference to FIG. 10. The evaluation device of the present invention may be a general purpose computer, such as a personal computer, or a dedicated evaluation computer. Part 1 of the input is a means for entering the above forecast formula, and the forecast formula used for the assessment is entered into it in advance. For example, an input device such as a keyboard may be used. Part 2 of the image capture is a means for capturing images of the skin, and you can use a digital video camera or a commercially available digital microscope. The Central Processing Unit (CPU) 3 is a means for calculating the physical parameters of the skin from captured skin images and a means for calculating the evaluation criteria related to the skin texture and / or wrinkle, according to a previously prepared forecast formula and calculated physical parameters of the skin. When the above evaluation program is installed, the device functions as such a means. Random Access Memory (RAM) 4 is a storage medium for temporarily storing data. The display portion 5 is a means for outputting the calculated estimated value and may be, for example, a display device, such as a liquid crystal display, or an output device, such as a printer.

The sequence of processing operations performed by the above evaluation device is described with reference to FIG.

First, a skin image is captured from an image capturing portion, such as a digital video camera. As described above, the image can be taken directly from the skin of the test subject or by means of a fingerprint sample. The above captured skin image is subjected to image processing, such as binary cross-conversion processing and short linear line processing processing on the CPU, while the physical parameters of the skin image are calculated. The types of physical parameters of the skin image to be calculated depend on the types of physical parameters used to calculate the prediction formula introduced in advance from the input means, and they are determined accordingly. The calculated physical parameters of the skin image are preliminarily substituted into the forecast formula in the CPU for the calculation of the evaluation criteria. Then, the calculated evaluation criteria are derived from the output means, such as a liquid crystal display.

It should be noted that in the present invention, when each of the formulas - the prediction formula related to the skin texture and the prediction formula related to the skin wrinkles - is pre-entered into the above evaluation device, the skin texture and wrinkles can be evaluated immediately or separately .

A method for selecting a preparation for external application to the skin based on an estimated estimated value related to skin texture

Based on the estimated value related to the texture evaluated by the evaluation method described above or the device for evaluation, a preparation may be selected for external application to the skin of the test subject whose skin image was used. Using the evaluation method or evaluation device of the present invention, the evaluation can be carried out quickly and with high accuracy, similar to the determination of the condition of the skin by a specialist. Therefore, based on the results, a preparation for external application to the skin can be selected to maintain, prevent deterioration or improve the texture of the skin.

When choosing a preparation for external application to the skin, in particular, when choosing a cosmetic product in the case when the displayed evaluation value indicates that the skin texture of the test subject is unsatisfactory, a cosmetic product containing a component to improve the texture or to prevent deterioration of texture suitable for the skin of the test subject. This component includes a component that stimulates metabolism, a stimulator of collagen synthesis, a stimulator of peeling of the epidermis and a pathogen for restoring bundles of collagen fibers, as well as a cosmetic product may contain one or more types of these components. Of these, the causative agent for restoring bundles of collagen fibers is the most effective for improving texture.

The aforementioned metabolism stimulating component includes retinoic acid, phytosteside, phytosterol, sphingosine and steroid. The above collagen synthesis stimulant includes malt root extract. The above epidermis peeling stimulant includes α-hydroxy acid. The causative agent for the restoration of bundles of collagen fibers, includes rosemary extract and derivatives of ursolic acid.

On the other hand, in the case where the display of the evaluation value indicates that the condition of the skin texture of the test subject is good, then a cosmetic product containing only a moisturizing component to maintain a state of texture suitable for the skin of the test subject can be selected. The above moisturizing component includes heparinoids. Such components may be contained in cosmetics in an amount to provide their effects, and preferably contained in an amount of from 0.01 to 5.0 wt.%. When a person skilled in the art properly selects components based on the displayed texture evaluation values, suitable cosmetics may be selected. The following is an example of choosing cosmetic components for five degrees of texture rating values. It should be noted that the components can naturally be contained not only in cosmetics, but also in other skin preparations intended for external use.

The value of texture evaluation, cosmetic component

1 (good texture), moisturizing component;

2 collagen synthesis stimulator and moisturizing component;

3 collagen synthesis stimulator, epidermis peeling stimulant and moisturizing component;

4 pathogen collagen fiber bundle recovery, collagen synthesis stimulator and moisturizing component;

5 (unsatisfactory texture) - the causative agent of restoration of bundles of collagen fibers, a stimulator of collagen synthesis, a stimulator of peeling of the epidermis and a moisturizing component.

A method for selecting a preparation for external application to the skin based on an estimated estimated value related to skin wrinkles

As in the case of texture, a preparation for external application to the skin, suitable for the evaluated image of the skin of the test subject, can be selected based on the evaluation criteria related to wrinkles, evaluated by the above evaluation method or device for evaluation. As examples of cosmetics selected on the basis of assessed evaluation criteria related to wrinkles, examples of the selection of cosmetic components for three degrees of wrinkle assessment values are given below.

Wrinkle score value, cosmetic component

1 (good condition), moisturizing component;

2 collagen synthesis stimulator, epidermis peeling stimulator and moisturizing component;

3 (unsatisfactory state) - the causative agent of the restoration of bundles of collagen fibers, a stimulator of collagen synthesis, a stimulator of peeling of the epidermis and a moisturizing component.

With the above selection of cosmetic components, not only texture or wrinkle assessment values, but also other indicators, such as various values of skin properties, results of a review of skin surface patterns and preference can be used in combination. Such combinations with other indicators are included in the technical scope of the present invention. Specific examples of such other indicators include: skin property values, such as loss of water through the epidermis (TEWL) and conductivity; the value of the properties of keratinized cells obtained by peeling off the adhesive tape from the skin surface (such as cell area, cell volume, distribution of cell area, flattening of cells, uniform distribution of cells, multilayer peeling of the stratum corneum cells and the presence of nucleated cells); the skin’s ability to retain water, the amount of sebum secreted, the degree of skin aging, the ability to produce melanin, skin color, skin property and skin quality, each of which is evaluated by property values. Of these, water retention is particularly preferred, which is closely related to the state of texture and the state of wrinkles.

Examples of the present invention are described below, but the scope of the present invention is not limited to them.

EXAMPLES

Example 1

Processing for visual assessment of texture and wrinkles

Fingerprint samples were collected from the central cheek regions of 30 women from adolescence to 60 years old, and images obtained from fingerprint samples using a cosmetic microscope manufactured by MORITEX were stored as digital data. The images were processed to eliminate interference using a general-purpose personal computer with the installed software for the above image processing to convert, by means of this, images into images obtained under bright lighting, and processed by cross-conversion to binary form, and processing by comparing short linear segments performed to calculate the physical parameters related to the dermal sulcus. 17 physical parameters were calculated as physical parameters, such as the area of the dermal furrows (see Fig. 8), the average width of the dermal furrows (see Fig. 9), the change in the width of the dermal furrows, the gap between the dermal furrows, the parallelism of the dermal furrows, the direction of the dermal furrows and density of dermal furrows. As is evident from Figs. 8 and 9, the physical parameters clearly show the characteristics of the irregularities of the dermal grooves and dermal protrusions and are indicators that are very easily visually determined.

Example 2

Automatic assessment of visual texture assessment

Using standard photographs to evaluate the five degrees of texture of fingerprint samples obtained from the central regions of the cheeks of women (see FIG. 6, created by the applicants of the present invention based on an aggregate including 1000 photographs), a total of 1000 digital images of the evaluation criteria related to to skin texture, in the range from 1 to 5 (1: good texture to 5: poor condition), including 200 images for each evaluation criterion, according to 15,000 digital images and visual evaluation criteria relating to skin prints with their use, is determined by three evaluators who are specialists dermatologists. Appraisers had at least one year of experience in research on beauty, aesthetics, or determination of skin conditions and periodically train skills to determine skin conditions. For 1000 images, the physical parameters were calculated by the methods described in Example 1. Subsequently, the images were divided into two groups, group A and group B, so that each group included 100 images of each of the visual assessment criteria from 1 to 5. Then, 500 images from group A were subjected to multiple regression analysis (manufactured by SPSS Inc.) using visual texture estimates as variables of the objective function and using 17 physical parameters as explanatory variables to calculate the equation multiple regression as a prediction formula (multiple correlation coefficient = 0.909). For the other 500 images from Group B, the physical parameters calculated as described above were substituted into the explanatory variables in the multiple regression equation to evaluate the visual evaluation criteria related to the texture (automatic estimation values). The list of physical parameters used is shown in table 2, and the results are shown in table 3.

table 2 The histogram ranges of angles of short linear segments from 0 ° to 180 ° at intervals of 5 ° Name of the physical parameter Definition of a physical parameter The number of short linear segments_2 The total number of short linear segments The average width of the dermal furrow_2 The average width of the dermal sulcus Dispersion of the width of the dermal sulcus_2 Dispersion (change) of the width of the dermal sulcus The interval between the dermal furrows_2 The average gap between the dermal furrows Parallelism of dermal furrows_2 Dispersion of the number of short linear segments for different angles Dispersion of the density of dermal furrows_2 Dispersion (change) in the density of dermal furrows Asymmetry 90-180_2 The asymmetry of the histogram of the number of short linear segments for different angles, when the histogram at angles of 90 to 180 degrees is considered as a normal distribution Peak formation 90-180_2 The formation of a histogram peak of the number of short linear segments for different angles, when the histogram at angles of 90 to 180 degrees is considered as a normal distribution The maximum frequency in the last half The maximum number at 95 ° or more on the histogram of the angles of short linear segments The maximum frequency in the first half The maximum number at 10 ° and 90 ° or less on the histogram of the angles of short linear segments Width type The maximum value on the histogram of the width Modal width position Width at maximum value in the width histogram Total number of connectivity The total magnitude of the data of the magnitude of the connected frequency of short linear segments Maximum in the first half ± 1 The total number of short linear segments at an angle of maximum frequency in the first half and at angles of ± 1 (5 °) angle Subtotal in the first half The total number at 10 ° or more and 90∈ or less on the histogram of the angles of short linear segments Subtotal in the last half Total number at 95 ° or more in the histogram of angles of short linear segments The maximum frequency in the last half_2 The maximum number at 95 ° or more on the histogram of the angles of short linear segments The maximum frequency in the first half_2 The maximum number at 10 ° or 90 ° or less on the histogram of the angles of short linear segments Width Type_2 The maximum value on the histogram of the width Number of measuring points for width_2 Total Width Frequency Data The maximum value of the starting point of short linear segments_2 The maximum value of the starting point of short linear segments on the histogram of the state of aggregation Width frequency after peak_2 The total value of the data of the frequency of the width from 4 to 12

Table 3 Visual Evaluation Criteria (automatic rating value) one 2 3 four 5 Visual assessment one 47 49 four 0 0 2 19 68 13 0 0 3 four 33 57 5 one four 0 0 fourteen 54 32 5 0 0 0 fifteen 85

Table 3 is a summary table of visual texture determination criteria (automatic estimation values) obtained by the present invention and visual texture evaluation criteria. Spearman's correlation coefficient is 0.887, the exact coincidence of both evaluation criteria is 62%, and if a shift of one degree is accepted, then the coincidence is 98%. Therefore, the texture estimation method of the present invention has satisfactory accuracy.

Example 3

Group A and group B were exchanged in Example 2, the multiple regression equation (multiple correlation coefficient = 0.935) created for images of group B was used to evaluate the visual assessment criteria (automatic rating values) of group A. The results are presented in table 4.

Table 4 Visual Evaluation Criteria (automatic rating value) one 2 3 four 5 Visual assessment one 29th 66 5 0 0 2 8 70 21 one 0 3 2 28 60 10 0 four 0 one 39 55 5 5 0 0 four 47 49

Table 4 is a summary table of visual assessment criteria related to texture (automatic estimation values) obtained by the present invention and visual assessment criteria related to texture. Spearman's correlation coefficient is 0.861, the coincidence of both evaluation criteria is 53%, and if a shift of one degree is accepted, then the coincidence is 97%. The results reveal that the method can continuously accurately evaluate unknown data with high accuracy.

Example 4

Automatic assessment of visual assessment of wrinkles

In Example 2, a total of 600 digital images of the evaluation criteria related to skin wrinkles were selected, ranging from 1 to 3, including 200 images for each evaluation value, and the procedure of Example 2 was carried out in the same manner as described above. The multiple correlation coefficient of the multiple regression equation obtained by multiple regression analysis was 0.912; Spearman's correlation coefficient of visual evaluation criteria related to wrinkles (automatic evaluation values) and visual evaluation criteria related to wrinkles is 0.705; the coincidence of both assessment criteria is 65%, and if a shift of one degree is accepted, then the exact match is 100%. Therefore, the method for evaluating wrinkles of the present invention has satisfactory accuracy.

Example 5

In Example 4, Group A data and Group B data were exchanged, and automatic evaluation was performed in the same manner as described above. As a result, the multiple correlation coefficient of the final multiple regression equation is 0.820, the Spearman correlation coefficient of visual evaluation criteria related to wrinkles (automatic judgment values) and visual evaluation criteria related to wrinkles is 0.880, the exact coincidence of both evaluation criteria is 84%, and if a shift of one degree is accepted, then the match is 100%. The results reveal that the method can continuously accurately evaluate unknown data also with the accuracy of wrinkles.

Example 6

Automatic assessment of visual texture assessment

In Example 2, instead of multiple regression analysis, a neural network (manufactured by NeuralWare) was used for group A in a controlled study to conduct a study based on physical parameters using visual assessment criteria related to texture as dependent variables, thereby obtaining a prediction formula . Physical parameters were substituted into the obtained forecast formula for evaluating visual assessment criteria related to the texture (automatic judgment values) of group B. The results are presented in table 5.

Table 5 Visual assessment criteria (automatic rating value) one 2 3 four 5 Visual assessment one 48 47 5 0 0 2 eighteen 62 19 one 0 3 0 37 55 7 one four 0 0 fifteen 56 29th 5 0 0 0 eighteen 82

Table 5 is a summary table of visual assessment criteria related to texture (automatic estimation values) obtained by the present invention and visual assessment criteria related to texture. Spearman's correlation coefficient is 0.871, the coincidence of both evaluation criteria is 62%, and if a shift of one degree is accepted, then the coincidence is 99%. The results reveal that even if a prediction formula is created using a multivariate analysis unit other than multiple regression analysis, an automatic assessment can be carried out with high accuracy.

Example 7

In Example 2, multiple regression analysis (manufactured by SPSS Inc.) was performed using the three physical parameters shown in Table 6 as explanatory variables for calculating the multiple regression equation (multiple correlation coefficient = 0.880). Spearman's correlation coefficient is 0.831, the exact coincidence of both evaluation criteria is 47%, and if a shift of one degree is accepted, then the coincidence is 95%. The results show that a greater number of physical parameters leads to higher accuracy than if the number of physical parameters is small, and the assessment can be carried out with high accuracy.

Table 6 The histogram ranges of angles of short linear segments from 0 ° to 180 ° at intervals of 5 ° Name of the physical parameter Definition of a physical parameter The intermediate amount of the first half Total at 10 ° or more and 90 ° or less in the histogram of the angles of short linear segments Width Type_2 The maximum value on the histogram of the width Width frequency after peak_2 The total value of the data of the frequency of the width from 4 to 12

Example 8

As examples and comparative examples of texture and wrinkle estimates, the coincidence rate of the estimates and the required time to evaluate the sample (s) are shown in Tables 7 and 8. That is, the estimates include texture estimation (comparative example 1) and wrinkle evaluation (comparative example 2) three dermatologists, appraisers (trained appraisers), which are used as standards for visual assessment; and texture evaluation (comparative example 3) and wrinkle evaluation (comparative example 4) by untrained evaluators (evaluating using standard photographs shown in FIGS. 6 and 1, which were explained earlier). In addition, in examples 1 and 2, binary conversion and refinement processing is performed (the connection diagram in the binary converted image obtained by sampling is converted to a linear image so as not to eliminate the connectivity) instead of being processed by cross-conversion to binary and comparing short linear segments, and the assessment was carried out in the same manner as in example 2, using statistical physical parameters, as a rule, obtained by the above processing methods ki, such as total area, standard deviation, total number, unit of area related to the dermal protrusion, average, standard deviation, standard error and coefficient of variation related to the intervals between the peaks of the thin line. Visual evaluation criteria related to texture and wrinkle (automatic evaluation values) are respectively defined as comparative examples 5 and 6.

Table 7 Texture Exact match rate (%) Frequency of coincidence, including shift by one degree (%) The required time (sec) Example 2 62 98 one Example 3 53 97 one Example 6 62 99 2 Comparative Example 1 - - from 2 to 3 Reference Example 3 45 82 21 Reference Example 5 fifty 84 one

Table 8 Wrinkle Exact match rate (%) Frequency of coincidence, including shift by one degree (%) The required time (sec) Example 4 65 one hundred one Example 5 84 one hundred one Reference Example 2 - - 2 Reference Example 4 49 88 eighteen Reference Example 6 54 89 one

Performance test by cosmetic selection method based on automatic texture rating value

A test of long-term use of a cosmetic product in female subjects was performed to determine the applicability of the method of choosing a cosmetic product based on texture evaluation values.

First, based on the following compositions, five types of cosmetics (cosmetics 1-5) for various conditions of skin texture were prepared in accordance with the usual method of obtaining cosmetics.

Cosmetic 1: cosmetic for value 1 automatic texture evaluation

Component Content Glycerol 5 wt.% 1,3-butanediol 5 wt.% Soy protein 0.1 wt.% heparinoid 0.1 wt.% Ethanol 5 wt.% Methylparaben 0.1 wt.% Water The rest of the amount

Cosmetic 2: cosmetic for value 2 automatic texture evaluation

Component Content Glycerol 5 wt.% 1,3-butanediol 5 wt.% Malt Root Extract 0.1 wt.% Soy protein 0.1 wt.% Heparinoid 0.1 wt.% Ethanol 5 wt.% Methylparaben 0.1 wt.% Water The rest of the amount

Cosmetic 3: cosmetic for value 3 automatic texture evaluation

Component Content Glycerol 5 wt.% 1,3-butanediol 5 wt.% Malt Root Extract 0.1 wt.% Sodium lactate 0.1 wt.% Soy protein 0.1 wt.% heparinoid 0.1 wt.% Ethanol 5 wt.% Methylparaben 0.1 wt.% Water The rest of the amount

Cosmetic 4: cosmetic for value 4 automatic texture evaluation)

Component Content Glycerol 6 wt.% 1,3-butanediol 5 wt.% Rosemary extract 0.1 wt.% Malt Root Extract 0.1 wt.% Soy protein 0.1 wt.% Stearylursolate 0.1 wt.% Heparinoid 0.1 wt.% Ethanol 10 wt.% Methylparaben 0.1 wt.% Water The rest of the amount

Cosmetic 5: cosmetic for value 5 automatic texture evaluation)

Component Content Glycerol 7 wt.% 1,3-butanediol 5 wt.% Rosemary extract 0.1 wt.% Malt Root Extract 0.1 wt.% Sodium lactate 0.1 wt.% Soy protein 0.1 wt.% Stearylursolate 0.1 wt.% Heparinoid 0.1 wt.% Ethanol 15 wt.% Methylparaben 0.1 wt.% Water The rest of the amount

Consequently, 80 healthy female participants (aged 32 to 57 years) were selected by random sampling, and they were divided into 2 groups A and B so that there was no difference in age distribution between the two groups. For group A, texture estimation values were calculated automatically from cheek fingerprints using the multiple regression equation obtained in Example 2, and cosmetics from 1 to 5 corresponding to automatic texture estimation values from 1 to 5 were used. For group B, automatic estimation values were calculated as in the same manner as described above, but the values were not taken into account, and cosmetic 5 was used for the value 5 of the automatic texture estimation. Participants from groups A and B used the prescribed cosmetics for three months, and then the values for automatic texture evaluation were calculated in the same manner as described above. Texture improvement values were calculated by the equation “texture improvement value” = “value of automatic texture evaluation before operational testing” - “value of automatic texture evaluation after operational testing” to determine average values for the corresponding groups. In the case when the test subject complained that the cosmetic product was not suitable for her skin during the test period, the subject stopped using the cosmetic product and was excluded from the assessment. The results are shown in Table 9. Table 9 shows that if the method of choosing cosmetics of the present invention was adopted, there were no participants who applied cosmetics that were not suitable for their skin, and cosmetics had effects that improved the texture of the skin.

Table 9 Adopting a method of choosing a cosmetic The value of the texture assessment before application (number of participants) Average texture improvement The number of participants complaining that the cosmetic product is not suitable for their skin Group A (adopting a selection method) 15 people) 0.00 0 people 2 (7 people) 0.29 0 people 3 (10 people) 0.60 0 people 4 (11 people) 0.82 0 people 5 (7 people) 1.00 0 people Group B (not accepting the selection method) 1 (6 people) 0.00 4 people 2 (8 people) 0.17 2 people 3 (10 people) 0.56 1 person 4 (9 people) 0.78 0 people 5 (7 people) 1.00 0 people

Performance test by cosmetic selection method based on automatic wrinkle rating value

A test of the long-term use of a cosmetic product in female subjects was performed to determine the applicability of the method of choosing a cosmetic product based on the values of the wrinkle score.

First, based on the following compositions, three types of cosmetics (cosmetics 1-3) for various conditions of skin wrinkles were prepared in accordance with the usual method of obtaining cosmetics.

Cosmetic 1: cosmetic for magnitude 1 automatic wrinkle assessment

Component Content Glycerol 5 wt.% 1,3-butanediol 5 wt.% Malt Root Extract 0.1 wt.% Soy protein 0.1 wt.% Heparinoid 0.1 wt.% Ethanol 10 wt.% Methylparaben 0.1 wt.% Water The rest of the amount

Cosmetic 2: cosmetic for magnitude 2 automatic wrinkle assessment

Component Content Glycerol 5 wt.% 1,3-butanediol 5 wt.% Malt Root Extract 0.1 wt.% Sodium lactate 0.1 wt.% Soy protein 0.1 wt.% heparinoid 0.1 wt.% Ethanol 10 wt.% Methylparaben 0.1 wt.% Water The rest of the amount

Cosmetic 3: cosmetic for magnitude 3 automatic wrinkle assessment

Component Content Glycerol 7% of the mass 1,3-butanediol 5 wt.% Rosemary extract 0.1 wt.% Malt Root Extract 0.1 wt.% Sodium lactate 0.1 wt.% Soy protein 0.1 wt.% Stearylursolate 0.1 wt.% Heparinoid 0.1 wt.% Ethanol 15 wt.% Methylparaben 0.1 wt.% Water The rest of the amount

Consequently, 72 healthy women participants (aged 30 to 58 years) were selected by random sampling, and they were divided into 2 groups A and B so that there was no difference in age distribution between the two groups. For group A, wrinkle scores were calculated automatically from the cheek prints using the multiple regression equation obtained in Example 4, and cosmetics from 1 to 3 were used, corresponding to the values for automatic texture estimation from 1 to 3. For group B, the automatic scores were calculated in the same manner as described above, but the values were not taken into account, and cosmetic 2 was used for the value 2 of the automatic assessment of wrinkles. Participants from groups A and B used the prescribed cosmetics for six months, and then the values for the automatic assessment of wrinkles were calculated in the same manner as described above. The wrinkle improvement values were calculated by the equation “wrinkle improvement value” = “the value of the automatic wrinkle assessment before the operational test” - “the value of the automatic wrinkle assessment after the operational test” to determine the average values for the corresponding groups. The results are shown in Table 10. Table 10 shows that, if the method of choosing the cosmetics of the present invention was adopted, the cosmetics had the effects of improving wrinkles.

Table 10 Adopting a method of choosing a cosmetic Wrinkle score before application (number of participants) Average wrinkle improvement Group A (adopting a selection method) 1 (10 people) 0.00 2 (15 people) 0.27 3 (11 people) 0.45 Group B (not accepting the selection method) 1 (9 people) 0.00 2 (15 people) 0.27 3 (12 people) 0.25

Industrial applicability

In accordance with the present invention, technology can be provided to facilitate evaluation of skin texture or wrinkles, which can be carried out anywhere, quickly and with high accuracy. As a result, you can get information useful for analyzing the condition of the skin and beauty or choosing cosmetics, for example, in department stores or stores.

Claims (7)

1. A method for evaluating skin texture and / or wrinkles, comprising the steps of:
performing image processing of the skin, including cross-conversion into binary form and / or processing by comparing short linear segments to obtain 10 or more kinds of physical parameters related to the dermal furrow; and
substitute the physical parameters obtained at the previous stage in the forecast formula, which is determined by multivariate analysis of 10 or more types of physical parameters related to the dermal sulcus, and visual estimated values distributed into categories from the third to tenth degree related to the skin texture and / or wrinkle obtained from a number of samples to obtain an estimated value in categories from 3 to 10 degrees and to evaluate the obtained estimated value as an estimated result in relation to skin texture and / or wrinkles. 2. The method for evaluating skin texture and / or wrinkles according to claim 1, wherein the skin image is a skin image obtained by a skin imprint sample. 3. The method for evaluating skin texture and / or wrinkles according to claim 2, wherein the skin image is a skin image obtained by taking an image formed by reflected light caused by light irradiation at an angle of 10 to 40 ° with respect to the print sample. 4. A device for evaluating the texture of the skin and / or wrinkles, including:
means for entering a prediction formula determined by multivariate analysis of 10 or more kinds of physical parameters related to the dermal sulcus and visual estimated values distributed in categories from the third to tenth degree related to the skin texture and / or wrinkle obtained from a number of samples ;
means for capturing images of the skin;
means for calculating 10 or more kinds of physical parameters related to the dermal sulcus, according to the captured image of the skin;
means for calculating the estimated value related to the texture of the skin and / or wrinkle in the category from third to tenth degree according to the forecast formula and calculated physical parameters; and
means for displaying the calculated estimated value. 5. A method for selecting a preparation for external application to the skin, comprising the steps of:
evaluate the texture of the skin and / or wrinkles using the evaluation method according to claim 1 or the evaluation device according to claim 4; and
choosing a preparation for external application to the skin containing a component to improve the texture or to prevent deterioration of the texture when the assessment indicates that the condition of the skin texture and / or wrinkles in the test subject is unsatisfactory, or choosing a preparation for external application to skin containing only a moisturizing component, in the case where the assessment indicates that the condition of the skin texture and / or wrinkles in the test subject is good, each of and / or wrinkle, ATEGORY from the third to the tenth power estimated at step evaluation. 6. The method of selecting a preparation for external application to the skin according to claim 5, wherein the preparation for external application to the skin is a cosmetic product. 7. The method of selecting a preparation for external application to the skin according to claim 6, in which the cosmetic product contains one or two or more types selected from the group consisting of a moisturizing component, a collagen synthesis stimulant, an epidermis peeling stimulator and a collagen fiber bundle recovery pathogen.

Images