Lip evaluation method

Abstract

To provide a technique for calculating a ratio at which, each of a plurality of elements forming an output factor, corresponds to the output factor, and evaluating features of the lip or the features of the body of a subject, on the basis of a calculation result.SOLUTION: A lip evaluation method is a method for evaluating features of the lip or features of the body of a subject, using one or more factors which are selected from a category group formed of: a physiological information category, a sensitiveness information category, and a color information category, as input factors, and using one or more factors different from the input factors, as output factors, where any one or more of the input factors or the output factors are factors related to the lip, each factor is formed of a plurality of elements which indicates a degree of the factor, or classifies the factor. In the method, a ratio at which, each of the plurality of elements forming the output factors, corresponds to the output factors, is calculated, on the basis of a mathematical model indicating a correlation between the input factors and output factors and being acquired in advance, from a parent population, and the input factors acquired from the subject.SELECTED DRAWING: Figure 13

Description

[Technical Field]

[0001] The present invention relates to a method for evaluating lips. [Background technology]

[0002] There is a method for determining the lip color of a user and selecting a lipstick color that suits the user's lip color category using a lipstick selection chart based on the lip color category (Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2008-62042 Summary of the Invention [Problem to be solved by the invention]

[0004] In Patent Document 1, a user's lip color is classified into one of several lip color categories based on lightness, saturation, and hue using a lip color scale. However, as research progressed, it became clear that simply classifying lip colors into one of several lip color categories was not sufficient, and there was room for improvement.

[0005] The present invention has been made in consideration of the above-mentioned problems, and relates to a lip evaluation method in which at least one of the input factors or output factors is a factor related to the lips, and each factor is composed of a plurality of elements, and the method calculates the corresponding ratio of each of the plurality of elements that make up the output factor based on a mathematical model that shows the correlation between the input factors and output factors obtained in advance and the input factors obtained from the subject, and evaluates the lip or physical characteristics of the subject based on the calculation results. [Means for solving the problem]

[0006] The present invention relates to a method for evaluating the lip or body features of a subject using, as input factors, one or more factors selected from a category group consisting of a physiological information category whose factors are physiological features related to the lips or body of the subject, an affective information category whose factors are sensory features related to the lips or body of the subject, and a color information category whose factors are color information related to the lips or body of the subject, and using, as output factors, one or more factors different from the input factors, wherein at least one of the input factors or the output factors is a factor related to the lips, and each factor is composed of a plurality of elements that indicate the degree of or classify the factor, and the lip evaluation method is characterized in that the lip evaluation method calculates the applicable ratio of each of the plurality of elements that make up the output factor based on a mathematical model that is obtained in advance from a population and indicates the correlation between the input factors and the output factors, and the input factors obtained from the subject, and evaluates the lip or body features of the subject based on the calculation result.

[0007] The present invention also relates to a system for evaluating the lip or body features of a subject using one or more factors selected from a category group consisting of a physiological information category whose factors are physiological features related to the lips or body of the subject, an affective information category whose factors are sensory features related to the lips or body of the subject, and a color information category whose factors are color information related to the lips or body of the subject as input factors, and one or more factors different from the input factors as output factors, wherein at least one of the input factors or the output factors is a factor related to the lips, and each factor is composed of a plurality of elements that indicate the degree of or classify the factor, and the lip evaluation system is characterized by comprising: calculation means for calculating the proportions of each of the plurality of elements that constitute the output factor based on a mathematical model that is obtained in advance from a population and indicates the correlation between the input factors and the output factors, and the input factors obtained from the subject; and evaluation means for evaluating the lip or body features of the subject based on the calculation results. [Effects of the Invention]

[0008] According to the method provided by the present invention, the ratios corresponding to each of the multiple elements that make up the output factor are calculated, and the characteristics of the subject's lips or body can be evaluated based on the calculation results, thereby enabling a multifaceted evaluation. [Brief explanation of the drawings]

[0009] [Figure 1] (a) A diagram showing the chromatic characteristics of people with pink lip colors, (b) A diagram showing the chromatic characteristics of people with pale lip colors, (c) A diagram showing the chromatic characteristics of people with purple lip colors, and (d) A diagram showing the chromatic characteristics of people with brown lip colors. [Figure 2] (a) is a diagram showing the distribution of lip L* (lightness) by lip color type, (b) is a diagram showing the distribution of lip C* (saturation) by lip color type, and (c) is a diagram showing the distribution of lip h (hue) by lip color type. [Figure 3] FIG. 10 is a diagram showing the percentage of subjects by lip color type by age group. [Figure 4] FIG. 1 is a diagram showing the relationship between lip color type and melanin index. [Figure 5] FIG. 1 is a diagram showing lip color types and ATI (peak blood flow position). [Figure 6] FIG. 10 is a diagram showing an example of an MBR time waveform. [Figure 7] FIG. 2 is a diagram showing each part of the lips. [Figure 8] FIG. 10 is a diagram showing the correlation between lip color and somatovisual category. [Figure 9] 1 is a flowchart illustrating a lip evaluation method. [Figure 10] FIG. 10 is a diagram showing an overview of implementation pattern 1. [Figure 11] FIG. 10 is a diagram showing an overview of implementation pattern 2. [Figure 12] FIG. 10 is a diagram showing an overview of implementation pattern 3. [Figure 13] FIG. 10 is a diagram showing an overview of implementation pattern 4. [Figure 14]FIG. 10 is a diagram showing an overview of implementation pattern 5. [Figure 15] FIG. 10 is a diagram showing an overview of implementation pattern 6. [Figure 16] FIG. 10 is a diagram showing an overview of implementation pattern 7. [Figure 17] FIG. 10 is a diagram showing an overview of implementation pattern 8. [Figure 18] FIG. 1 is a conceptual diagram of a lip information evaluation system. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. Note that the drawings of the present embodiments are intended to explain the technical concept, configuration, and operation of the present invention, and are not intended to specifically limit the configuration. In addition, in all drawings, similar components are given similar reference numerals, and duplicated explanations will be omitted as appropriate.

[0011] <Summary> An outline of the lip evaluation method according to this embodiment will be described. The lip evaluation method of this embodiment (hereinafter may be referred to as this method) is a method for evaluating the lip or body features of a subject using, as input factors, one or more factors selected from a category group consisting of a physiological information category, which uses as factors physiological features related to the lips or body of the subject, an affective information category, which uses as factors sensory features related to the lips or body of the subject, and a color information category, which uses as factors color information related to the lips or body of the subject, and using, as output factors, one or more factors different from the input factors, wherein at least one of the input factors or the output factors is a factor related to the lips, and each factor is composed of multiple elements that indicate the degree of or classify the factor, and is characterized in that the method calculates the corresponding ratio of each of the multiple elements that make up the output factor based on a mathematical model that is obtained in advance from a population and shows the correlation between the input factors and the output factors, and the input factors obtained from the subject, and evaluates the lip or body features of the subject based on the calculation result.

[0012] In the past, lip color classification was performed by classifying (determining) one lip color category from multiple lip color categories (lip color types, lip colors), as in Patent Document 1, for example. However, lip colors often have no clear boundaries between lip color categories and combine multiple lip color categories in a predetermined ratio. The inventors have discovered that it is important to calculate and evaluate the ratios at which lip colors fall within multiple lip color categories, rather than classifying lip colors into specific lip color categories. Therefore, we propose a lip evaluation method that calculates the ratios at which multiple elements (e.g., multiple color types) constituting an output factor (e.g., color group A) fall, and evaluates the lip or physical characteristics of a subject based on the calculation results.

[0013] In this lip evaluation method, at least one of the input factors or output factors is a factor related to lips, and each factor is composed of multiple elements that indicate the degree of the factor or classify the factor. Lip evaluation is performed by calculating the ratio of each of the multiple elements that make up the output factor based on a mathematical model that is obtained in advance from a population and indicates the correlation between the input factor and the output factor, and the input factor obtained from the subject. To make the present invention easier to understand, among the color group factors that are divided using lip pigment information, color group A factors (factors that belong to the color type category) and factors that belong to the color information category (L * (lightness), C * We will first provide a specific explanation of the correlation between color and the color components (color saturation, h (hue)), and then explain each factor and each category.

[0014] Figure 1 shows the results of lip color classification, based on lip color (mainly the lower lip) assessment of 85 Japanese female subjects in their 20s to 70s. Evaluation was performed visually by three expert evaluators familiar with lip color. First, lip colors with similar colors were assessed to minimize the number of types (first assessment). A reassessment was then performed using the same photographs after a set number of days (second assessment), and the final color classification was determined. If the results of the first and second assessments differed, a reassessment was performed. As a result of the assessment, as shown in Figure 1, the lip colors of Japanese women can be classified into four types. In this specification, each type is given a tentative color name and will be described as (a) pink type, (b) pale type, (c) purple type, and (d) brown type. However, there is no clear boundary, and it is common for multiple types to be combined, such as a pink type and a pale type. Here, the absence of a clear boundary refers to the fact that the color information (L) of the four groups is inconsistent. * , C * , h, etc.) have different central values, but are not completely separated without overlapping, meaning that there are areas that overlap partially. Here, the characteristics of each type will be described.

[0015] Figure 1(a) Pink type has bright red lips. * is high, C * is high and h is low. Figure 1(b) Pale type is a lip color without redness. * is high, C * is low and h is high. Figure 1(c) Purple type is a lip color with low saturation. * is high, C * is low and h is low. Figure 1(d) Brown type has dark lip color. * is low, C * is low and h is high.

[0016] Figures 2(a) to (c) show the L * , C *, and h, and what color types are distributed at what values. Figure 2(a) shows the L of the subjects' lips by color type. * The horizontal axis represents color type, and the vertical axis represents lightness. As is clear from Figure 2(a), the distribution of L * There is no big difference in the values, but the brown type is L * is distributed at lower values ​​than the other three types. There is no significant difference between the pink, pale, and purple types, but there is a significant difference between the purple and brown types, and there are also significant differences between the pink and brown types, and between the pale and brown types. Figure 2(b) shows the C of the subjects' lips by color type. * The horizontal axis represents the color type, and the vertical axis represents the saturation value. As is clear from Figure 2(b), the pink type is distributed in the C * The pale, purple and brown types are distributed in high values, but the C * The pink type is significantly different from each of the other types, and there are also significant differences between the pale type and the purple type, and between the purple type and the brown type. Figure 2(c) shows the distribution of h for the subjects' lips by color type. The horizontal axis represents color type, and the vertical axis represents hue. The redder the lip color, the lower the h value. As is clear from Figure 2(c), the pink and purple types have low h values, while the pale and brown types have high h values. There are significant differences between the pink and pale types, and between the pink and brown types.

[0017] Figure 3 shows the distribution of color types among subjects by age group. The horizontal axis shows age, and the vertical axis shows the proportion of subjects with each color type. For example, the proportion of pink types is high among people in their 20s, and the proportion of purple types is high among people in their 70s. Furthermore, the proportion of pink types decreases as the age group increases. In this way, the proportion of subjects with each color type varies depending on age group.

[0018] Figure 4 shows the distribution of melanin index (MI) of the subjects' lips by color type. The horizontal axis represents color type, and the vertical axis represents MI. Of the four types, the pink, pale, and purple types have low MI values, while the brown type has high MI values. There are significant differences between the brown type and the pink, pale, and purple types.

[0019] Figure 5 shows the ATI (peak blood flow position) by color type. The lower the ATI, the earlier the peak is reached. The pink type has blood flow that reaches its peak quickly, while the pale and purple types have slower peaks. Table 1 summarizes the details of Figures 1 to 5.

[0020] [Table 1]

[0021] As shown in Table 1, the factors belonging to the color information category (L * , C * By classifying lip color into types using factors belonging to the physiological information category (MI, ATI) and the features estimated from them, it is possible to make suggestions based on the classification results, taking into account the physiological information factors that contribute to lip color information, such as makeup techniques, makeup products, skin care methods, and skin care products.

[0022] Next, each factor used in this lip evaluation method and each category to which each factor belongs will be described. Table 2 shows the categories of each factor in this embodiment.

[0023] [Table 2]

[0024] As shown in Table 2, in this embodiment, the categories are comprised of a "physiological information category" based on physiological characteristics of the subject's lips or body as factors, a "sensory information category" based on sensory characteristics of the subject's lips or body as factors, and a "color information category" based on color information of the subject's lips or body as factors. The "physiological information category," "sensory information category," and "color information category" each include lip-related factors and factors other than lip-related factors. In this embodiment, the major categories are the "physiological information category," the "sensory information category," and the "color information category." The intermediate categories are the "pigment category," the "blood flow category," the "blood vessel category," the "shape category," the "surface shape index category," the "color type category," the "body category," the "age category," and the "gender category" that belong to the "physiological information category," the "lip visual category," the "lip tactile category," the "lip somatosensory category," the "body visual category," the "body tactile category," and the "body somatosensory category," and the "lip color category" and the "body color category" that belong to the "color information category."

[0025] The "physiological information category" is a category to which factors of the physiological characteristics of the subject's lips and the subject's body belong. In this embodiment, categories are further defined for each physiological characteristic factor. The categories include a "pigment category" using lip pigment information as a factor, a "blood flow category" using lip blood flow information as a factor, a "blood vessel category" using lip blood vessel information as a factor, a "shape category" using lip shape information as a factor, a "surface shape index category" using lip surface shape index information as a factor, a "color type category" using lip color group information as a factor, a "body category" using the physiological characteristics of the subject's body as a factor, an "age category" using the subject's age information as a factor, and a "gender category" using the subject's gender information as a factor.

[0026] The "pigment category" includes factors that are indicators (pigment information) related to the type and amount of pigment in the lips. The apparent color of the lips changes depending on the type and amount of pigment, so the factors in the lip color category (L * However, since the amount of pigment is a physiological characteristic, the pigment category belongs to the "physiological information category." In this embodiment, MI (melanin index) and EI (erism index) belong to the pigment category. MI is an index that roughly corresponds to the amount of melanin. EI is an index that roughly corresponds to hemoglobin content. For example, at low MI, lip color will be more dependent on hemoglobin content characteristics. For each factor belonging to the pigment category, in addition to the measured value itself, a numerical range is set for each factor, and each range is configured as an element.

[0027] The "blood flow category" includes MBR (Mean Blur Rate): blood flow rate, Fluctuation: rate of change in blood flow rate, Skew: bias of blood flow waveform, BOS (Blowout Score): steady blood flow, BOT (Blowout Time): sustained high blood flow, Rising rate: time variation in the rate of blood flow increase, Falling rate: time variation in the rate of blood flow decrease, FAI: maximum acceleration of blood flow increase, ATI: position of peak blood flow, and RI (Resistivity Index): peripheral vascular resistance. Each factor in the blood flow category is composed of not only the measured value itself but also a numerical range set for each factor, and each range is used as an element.

[0028] MBR is the relative blood flow rate, and is a value obtained by applying the average MBR value to the linear regression equation: blood flow rate mL / (min·100 g)=41.7×MBR value−15.1. Fluctuation is an index that indicates the coefficient of variation of MBR. Skew is a value that indicates the deviation of the blood flow waveform. BOS is a value that indicates steady state blood flow. BOT is an index showing the sustainability of a high MBR value, and the BOT value is expressed by the following equation using the half-width (time) (W) and one-beat width (time) (F) of the MBR value shown in Figure 6, and a proportionality constant "C". (BOT)=C·(W) / (F) Rising rate is an index that indicates the time variation of the rate of blood flow increase. The falling rate is an index that indicates the time variation of the rate of decline from the maximum MBR value in the time waveform of the MBR value per heartbeat. Using (Sall) and (S2) shown in Figure 6, the falling rate can be expressed by the following equation. In the following, "C" is a proportionality constant. (Falling rate)=C·(S2) / (Sall) FAI indicates the instantaneous maximum blood flow rate when blood flow increases. The ATI is the proportion of the time during which the MBR time waveform reaches the maximum MBR value (MBRmax in FIG. 6) in one heartbeat. RI is an index showing peripheral vascular resistance, and the RI value is obtained by dividing the difference between the maximum MBR value (MBRmax in FIG. 6) and the minimum MBR value (MBRmin in FIG. 6) by the maximum MBR value.

[0029] The "vascular category" includes information about the blood vessels of the lips, such as area, width, length, density, and branching. Each factor of area, width, length, density, and branching is composed of not only the measured value itself measured from image data of the lips, but also a numerical range set for each factor, and each range is configured as an element.

[0030] The "shape category" includes the width of the upper lip velum, the length of the upper lip velum, the width of the lower lip vermilion, and the vertical and horizontal widths of the vermilion. Figure 7 shows the length, width, and other parts of each factor in the shape category. Factors in the shape category are configured by setting a numerical range for each factor in addition to the values ​​of each part shown in Figure 7, and each range is used as an element.

[0031] The "surface shape index category" includes wrinkle-derived and peel-derived factors. Wrinkle-derived factors are values ​​obtained from the depth, length, width, etc., obtained from surface irregularity information. Peel-derived factors are values ​​obtained from the unevenness of minute irregularities on the surface. Factors that belong to the surface shape index category are configured by setting a numerical range for each factor in addition to the obtained value itself, and each range is used as an element.

[0032] The "color type category" includes the color group A, which is composed of the above-mentioned "pink type" element, "pale type" element, "purple type" element, and "brown type" element, as well as the color pink group, which is composed of dark pink type element and light pink type element. In this embodiment, the elements that compose each color group, which is each factor, are determined by, for example, measuring the lip color of the measured L that belongs to the lip color category. * , a * (The larger the positive value, the redder the color, and the larger the negative value, the greener the color. The larger the absolute value, the more vivid the color.) * (The larger the positive value, the more yellow the color, and the larger the negative value, the more blue the color. The larger the absolute value, the more vivid the color becomes.) *This is done by cluster classification using the values ​​of h and Age, which belongs to the age category. As mentioned above, the lip color becomes less red and more brown with age, so L * , a * , b * , C * Factors belonging to the lip color category such as , h (L * It is preferable to determine the color type in consideration of factors belonging to the age category (Age) rather than just factors belonging to the age category. In this way, factors of the age category included in the physiological information category are used to determine the elements of each color group in the color type category, so in this embodiment, the "color type category" is included in the physiological information category. Note that factors of the age category are not used to determine the elements of each color group, and only the color of appearance (factors of the lip color category (L)) is used. * If only lip color (e.g., lip gloss) is used, Color Group A may be included in the lip color category.

[0033] "Physical category" is a category to which factors related to physical information such as height, weight, BMI (Body Mass Index), blood pressure, heart rate, etc. In addition to the obtained values ​​themselves, a numerical range is set for each factor, and each range is configured as an element. "Age category" refers to age itself or a generation (such as 20s, 30s, etc.), and when generation is a factor, for example, 20s, 30s, 40s, etc. are the elements that make up the generation." The "gender category" is a factor of gender, and elements constituting gender may include not only male and female, but also elements such as masculine and feminine.

[0034] The "sensory information category" is a category to which sensory feature factors related to the subject's lips and the subject's body belong. In this embodiment, the sensory feature factors are further divided into categories for each feature. There are provided a "lip visual category" that uses visual information for the lips as a factor, a "lip tactile category" that uses tactile information for the lips as a factor, a "lip somatosensory category" that uses somatosensory information for the lips as a factor, a "body visual category" that uses visual information for the body, particularly the face, neck, and décolleté close to the lips, as a factor, a "body tactile category" that uses tactile information for the body, particularly the face, neck, and décolleté close to the lips, as a factor, and a "body somatosensory category" that uses somatosensory information for the body, particularly the face, neck, and décolleté close to the lips, as a factor.

[0035] Here, the results of an experiment conducted on the influence of lip color on body impression (facial impression) will be described. <Experimental Method> 31 people Evaluators (8 in their 20s, 10 in their 30s, 6 in their 40s, 7 in their 50s: 17 men, 14 women) Method: Stimulus pairs created from average faces were presented on a PC monitor and evaluated using Scheffe's paired comparison method (Ura's modified method). Evaluation items: 1) Bright complexion 2) Good complexion 3) Beautiful skin 4) Attractiveness rated on a 5-point scale Stimulus: The average face of a person in their 40s (original image) with only the lip color removed. * , a * , b * Seven stimulus images, each varied by ±5 based on the color system. Note that the facial colors in images A to G are all the same, with only the lip color varied. Figure 8 shows the experimental results.

[0036] From Figure 8, for example, the L of the lips * Increasing the value makes the face appear "brighter" and the L of the lips * If the value is reduced, the complexion does not appear brighter, and the L of the lips * It can be said that there is a relationship between the value of and whether the face is perceived as bright or not. * Increasing the value makes the face appear "brighter" and the lips appear "brighter."* If the value is reduced, the complexion does not appear brighter, and the a * It can be said that there is a relationship between the value and whether or not one's complexion appears bright. Similarly, for example, the lip L * When the value is reduced, it is clear that the complexion does not "feel good" and the L of the lips * It can be said that there is a relationship between the value and whether the complexion feels good or not. * Higher values ​​make the complexion "feel" better, and the lip a * When the value is reduced, it is clear that the complexion is "not good" and the lip a * It can be said that there is a relationship between the value and whether or not one's complexion looks good. Similarly, for example, the lip L * It is clear that the skin does not "feel beautiful" when the value is reduced, and the L * It can be said that there is a relationship between the value and whether the skin feels beautiful or not. * Higher values ​​make the skin look "cleaner" and the lips look * It is clear that the skin does not "feel beautiful" when the value is reduced, and the a * It can be said that there is a relationship between the value and whether or not your skin feels beautiful. Similarly, for example, the lip L * It is clear that a smaller value makes the subject appear less attractive, and the L * Similarly, for example, the value of the lip a * A higher value makes the lips appear more attractive, * It is clear that a smaller value makes the subject less attractive, and the lips * It can be said that there is a relationship between value and whether or not something is perceived as attractive. In this way, it is clear that there is a relationship between lip color (lip color category) and body visual perception category, and it is also clear that the visual perception of the face changes just by changing the lip color without changing the facial color.

[0037] The "lip visual impression category" includes factors such as complexion, transparency, bright color, good color, beautiful, and attractive. Complexion refers to how the subject or other subjects feel about the complexion when looking at the lips, and is composed of, for example, elements such as "the complexion looks good (applies)" or "doesn't look good (doesn't look good (doesn't look good)"). Transparency refers to how the subject or other subjects feel about the transparency when looking at the lips, and is composed of, for example, elements such as "the complexion looks clear (applies)" or "doesn't look clear (doesn't look good ... Beauty refers to how the subject or another subject feels about beauty when looking at the lips, and is composed of, for example, elements of "looks beautiful" / "doesn't look beautiful (doesn't apply)." Attractiveness refers to how the subject or another subject feels about attractiveness when looking at the lips, and is composed of, for example, elements of "looks attractive" (does apply) / "doesn't look attractive (doesn't apply). In this embodiment, all factors included in the visual category are composed of two elements, "applies" / "doesn't apply," but this is not a limitation, and all factors may be composed of three elements, "very applicable," "applies," and "doesn't apply." Furthermore, some factors may be composed of two elements and other factors may be composed of three elements.

[0038] The "lip texture category" includes factors such as roughness and elasticity. Roughness is comprised of the elements of whether the subject or another subject felt it was rough when touching their lips (applies) / did not feel it was rough (does not apply). Elasticity is comprised of the elements of whether the subject or another subject felt it was elastic when touching their lips (applies) / did not feel it was elastic (does not apply). In this embodiment, all factors included in the texture category are comprised of two elements, "applies" / "does not apply," but this is not a limitation, and all factors may be comprised of three elements, "very applicable," "applies," and "does not apply." Furthermore, some factors may be comprised of two elements and other factors of three elements.

[0039] The "lip somatic sensation category" includes factors such as a feeling of tightness and a feeling of dryness. The feeling of tightness is comprised of the elements of whether the subject felt that their lips were tight (applies) or not (does not apply). The feeling of dryness is comprised of the elements of whether the subject felt that their lips were dry (applies) or not (does not apply). In this embodiment, all factors included in the somatic sensation category are comprised of two elements, "applies" or "does not apply," but this is not limited to this, and all factors may be comprised of three elements, "very applicable," "applies," and "does not apply." Furthermore, some factors may be comprised of two elements and other factors of three elements.

[0040] The "body visual perception category" includes factors such as complexion, transparency, bright color, good color, beautiful, and attractive. Complexion refers to how the subject or other subjects perceive the complexion when looking at the body, and is composed of, for example, elements such as "the complexion looks good (applies)" or "the complexion looks bad (does not apply)." Transparency refers to how the subject or other subjects perceive the transparency when looking at the body, and is composed of, for example, elements such as "the complexion looks transparent (applies)" or "it does not look transparent (does not apply)." Bright color refers to how the subject or other subjects perceive the brightness of the color when looking at the body, and is composed of, for example, elements such as "the color looks bright (applies)" or "it does not look bright." Good color refers to how the subject or other subjects perceive the good color when looking at the body, and is composed of, for example, elements such as "it applies" or "it does not look good (does not apply)." Beauty is how the subject or another subject feels about beauty when looking at the lips, and is composed of, for example, elements such as "looks beautiful" / "doesn't look beautiful" (doesn't apply). Attractiveness is how the subject or another subject feels about attractiveness when looking at the body, and is composed of, for example, elements such as "looks attractive" (does apply) / "doesn't look attractive" (doesn't apply). In this embodiment, all factors included in the body visual perception category are composed of two elements, "applies" / "doesn't apply," but this is not a limitation, and all factors may be composed of three elements, "very applicable," "applies," and "doesn't apply." Furthermore, some factors may be composed of two elements and other factors may be composed of three elements.

[0041] The "body texture category" includes factors such as roughness and elasticity. Roughness is comprised of the elements of whether the subject or another subject felt it was rough when touching the body (applies) / whether it was not rough (does not apply). Elasticity is comprised of the elements of whether the subject or another subject felt it was elastic when touching the body (applies) / whether it was not elastic (does not apply). In this embodiment, all factors in the body texture category are comprised of two elements, "applies" / "does not apply," but this is not a limitation; all factors may be comprised of three elements, "very applicable," "applies," and "does not apply." Furthermore, some factors may be comprised of two elements and other factors of three elements.

[0042] The "body somatic sensation category" includes factors such as a feeling of tightness and a feeling of dryness. The feeling of tightness is comprised of the elements of whether the subject felt that their body was tight (applies) or not (does not apply). The feeling of dryness is comprised of the elements of whether the subject felt that their body was dry (applies) or not (does not apply). In this embodiment, all factors included in the body somatic sensation category are comprised of two elements, apply / do not apply, but this is not limited to this, and all factors may be comprised of three elements, very applicable, apply, and do not apply. Furthermore, some factors may be comprised of two elements and other factors of three elements.

[0043] The "color information category" is a category to which the colorimetric values ​​of the subject's lips and body or information factors for identifying the type of color based on the colorimetric values ​​belong, and further categories are established for each part of the subject, such as a "lip color category" in which color information related to the lips is a factor, and a "body color category" in which color information related to the body, particularly the face, neck, and décolleté close to the lips, is a factor.

[0044] The lip color category is a category to which factors of information for identifying the color type based on the color measurement values ​​of the lips or the color measurement values ​​belong. In this embodiment, the following factors belong to the lip color category. L* is the brightness. a * The larger the positive value, the more red the color tends to be, and the larger the negative value, the more green the color tends to be. The larger the absolute value, the more vivid the color becomes. b * The larger the positive value, the more yellow the color tends to be, and the larger the negative value, the more blue the color tends to be. The larger the absolute value, the more vivid the color becomes. C * is the saturation. h is the hue. R is a value related to the intensity of long wavelengths (red region) in RGB, a type of color expression. G is a value related to the intensity of the medium wavelength (green region) in RGB, which is a type of color expression. B is a value related to the intensity of short wavelengths (blue region) in RGB, a type of color expression. The optical spectrum expresses the relative proportion of each single color. A color swatch is a list of color samples created according to a predetermined standard. The color difference is a value that represents the difference in color, for example, a value that represents the difference between the red lip portion and the white lip portion. Color unevenness indicates the spatial non-uniformity of color, and is a value obtained from frequency, histogram shape (variance, skewness, kurtosis, etc.), statistics, etc.

[0045] The body color category is a category to which factors of information for identifying the color type based on the measured color values ​​of the body belong. In this embodiment, the factors belonging to the body color category are similar to those in the lip color category, and therefore a description thereof will be omitted.

[0046] The factors listed in Table 2 are those in this embodiment, and other factors may be included.

[0047] This method will now be described. A "physiological information category with factors being physiological characteristics of the subject's lips or body" is a category to which factors of the physiological characteristics of the subject's lips and body belong, and includes a "pigment category" with lip pigment information as a factor, a "blood flow category" with lip blood flow information as a factor, a "blood vessel category" with lip blood vessel information as a factor, a "shape category" with lip shape information as a factor, a "surface shape index category" with lip surface shape index information as a factor, a "color type category" with lip color group information as a factor, a "body category" with factors being physiological characteristics of the subject's body, an "age category" with subject age information as a factor, or a "gender category" with subject gender information as a factor. A "sensory information category with sensory characteristics related to the subject's lips or body as factors" is a category to which factors of sensory characteristics related to the subject's lips and body belong, and includes a "lip visual category" with visual information for the lips as a factor, a "lip tactile category" with tactile information for the lips as a factor, a "lip somatosensory category" with somatosensory information for the lips as a factor, a "body visual category" with visual information for the body, particularly the face, neck, and décolleté close to the lips as a factor, a "body tactile category" with tactile information for the body, particularly the face, neck, and décolleté close to the lips as a factor, and a "body somatosensory category" with somatosensory information for the body, particularly the face, neck, and décolleté close to the lips as a factor. A "color information category that uses color information about the subject's lips or body as a factor" is a category to which factors of information for identifying the type of color based on colorimetric values ​​or colorimetric values ​​of the subject's lips and body belong, and includes a "lip color category" that uses color information about the lips as a factor and a "body color category" that uses color information about the body, particularly the face, neck, and décolleté that are close to the lips, as factors. The "category group" consists of the following: "Pigment category," "Blood flow category," "Blood vessel category," "Shape category," "Surface shape index category," "Color type category," "Body category," "Age category," and "Gender category" in the physiological information category; "Lip visual category," "Lip tactile category," "Lip somatosensory category," "Body visual category," "Body tactile category," and "Body somatosensory category" in the sensory information category; and "Lip color category" and "Body color category" in the color information category. "One or more factors selected from a group of categories are used as input factors, and one or more factors different from the input factors are used as output factors" means that one or more factors selected from a group of categories are used as input factors, and one or more factors different from the input factors are used as output factors, so that multiple factors in the same category may be used as input factors, or factors in different categories may be used as input factors. Also, as long as the output factors are different from the input factors, multiple factors in the same category may be used as output factors, or multiple factors in different categories may be used as output factors.

[0048] "One or more of the input factors or output factors is a factor related to the lips" means that either the input factor or the output factor is a factor that belongs to any of the "pigment category," "blood flow category," "blood vessel category," "shape category," "surface shape index category," "color type category," "body category," "lip visual category," "lip tactile category," "lip somatosensory category," and "lip color category," and either the input factor or the output factor may include a factor related to the lips, or both may include a factor. "Each factor is composed of multiple elements that indicate the degree of the factor" means, for example, in the case of each factor belonging to the pigment category, the numerical range set for each factor corresponds to multiple elements that indicate the degree of the factor, and also, for example, in the case of each factor belonging to the lip visual category, the stages of "applies" / "does not apply" set for each factor correspond to multiple elements that indicate the degree of the factor. "Each factor is composed of multiple elements that classify the factor" means, for example, in the case of each factor belonging to the color type category, the elements determined by visual evaluation correspond to multiple elements that classify the factor. A "mathematical model that is obtained in advance from a population and shows the correlation between input factors and output factors" is a mathematical model that shows the relationship between multiple input factors and output factors, created from information about each factor obtained in advance from a population consisting of multiple subjects. The mathematical model corresponds to a mathematical model that outputs the ratio of multiple elements, and examples of this include a naive Bayes classifier, gradient boosting, support vector machine, neural network, and deep learning. The "input factors obtained from the subject" refer to the above-mentioned input factors, such as values ​​obtained by measuring the color of a predetermined part of the subject, values ​​obtained by measuring a predetermined value of the subject, and the like. "Calculating the ratio of the multiple elements that make up the output factor to which each applies" means calculating the probability that each of the multiple elements that make up the output factor will apply as a ratio, or finding it as a percentage. "Evaluating the characteristics of the subject's lips or body based on the calculation results" means evaluating the subject's lips or body characteristics, such as external aspects such as color and appearance, or internal aspects such as blood flow, based on the calculated output factors.

[0049] Next, an outline of the processing flow of this method will be explained with reference to FIG.

[0050] The step (step S100) is a step of determining input factors and output factors. In this method, one or more factors selected from a category group consisting of physiological information categories, sensory information categories, and color information categories are used as input factors, and factors different from the input factors are used as output factors, and the proportions of each of the multiple elements constituting the output factor can be calculated. Because at least one of the input factors or output factors must be lip-related, step S100 preferably includes a function (determination function) for checking that at least one lip-related factor is included when determining the input and output factors. If this checking function is not included, it is preferable that an error be generated in step S110 (described below) and no results be calculated for the determined output factors. As long as at least one lip-related factor is included in the input and output factors, any type of input and output factor can be determined, but the accuracy of the output factor obtained varies depending on the input factor. Therefore, by determining the desired output factor, it is preferable that a combination and type of input factors with high accuracy be automatically or recommended. Examples of preferred combinations of input and output factors are shown below. At least one of the input factors and the output factor must be in the physiological information category. There are multiple input factors, and some of the input factors and the output factor are in the same category, while other parts of the input factors and the output factor are in different categories. At least some of the input factors and output factors have the same major classification but different intermediate classifications. - There are multiple input factors, and the categories are divided into different major categories, and the output factor is one of the categories of the input factors. There are multiple input factors, and at least one of the input factors is a color information category, and the other input factors and output factors are physiological information categories.

[0051] In step S110, the ratios of multiple elements that make up the output factors are calculated using the input factors determined in step S100 and a mathematical model that calculates multiple ratios that make up the output factors. The mathematical model shows the relationship between the input factors and the output factors from the population.

[0052] The step (step S120) is a step of suggesting a care method, a makeup product, etc. based on one of the elements constituting the output factor obtained in step S110 or the ratio of the elements constituting the output factor.

[0053] Next, lip evaluation performed using this method will be described. Note that the implementation patterns shown below are examples. Implementation patterns 1 to 8 are examples in which a Gaussian Naive Bayes model is used as a mathematical model showing the correlation between multiple input factors and the output factors to determine the ratio of each of the multiple elements that make up an output factor. Furthermore, each mathematical model was created in advance using the measured values ​​of the factors shown in Table 2 for the population of 85 people described in Figure 1 and the evaluation results of 31 people described in Figure 8. Furthermore, when factors in the color information category are used as input factors, the measured values ​​of the user's bare lips are used (note that L * , a * , b * , C * , h was set to D65). This allows for an accurate evaluation of the user's lips. When the color type is used as an input factor, the color type can be calculated in advance from the measurement values ​​of the factors in the lip color category and the measurement values ​​of the factors included in the physiological information category. For factors in the shape category, an image of the user's lips can be acquired in advance and the size of each part of the lip image can be measured and used.

[0054] <Implementation pattern 1> Figure 10 shows an overview of implementation pattern 1. In the implementation pattern 1, based on the Gaussian Naive Bayes model created using the measurement values ​​of the population (85 people in this embodiment), the ratio of the likelihood of each of the multiple elements (pink type, pale type, purple type, and brown type) that make up the color group A factor belonging to the color type category was calculated. * The input factors were the user's measurements (three variables) of MI belonging to the pigment category of the physiological information category, and h. That is, at least one of the input or output factors (corresponding to the input and output factors in this implementation pattern) was a lip-related factor (corresponding to the pigment category, lip color category, and color type category in this implementation pattern), and a factor different from the input factor (corresponding to the pigment category and lip color category in this implementation pattern) was used as the output factor (corresponding to the color type category in this implementation pattern), and the proportion of each of the multiple elements (pink type, pale type, purple type, and brown type in this implementation pattern) that make up the output factor, Color Group A, was calculated. As shown in Figure 10, the generalization performance of this model for the test data was high, with a correct answer rate of 0.88 (88%). In addition, the classification evaluation performed with this model also showed the highest rate of classification into the desired elements. Then, the user's measurement value (C *Using a mathematical model (H:20, h:40, MI:45) to determine the ratios of the multiple elements that make up color group A, it was determined that the likelihood of the color group A being included was pink, brown, pale, and purple, in that order. Using this result, for example, when recommending recommended makeup products (lipstick), it is possible to recommend not only makeup products suitable for pink-type users but also makeup products suitable for brown-type users. By making such suggestions, even if a user prefers makeup products (favorite colors) suitable for brown-type users over makeup products suitable for pink-type users, the user can more easily select makeup products suitable for brown-type users, thereby making it possible to make appropriate suggestions. Furthermore, in the case of implementation pattern 1, the likelihood that the user's color type is purple is the lowest, so it is also possible to suggest that it is better not to select makeup products suitable for purple types. As mentioned above, there is no clear boundary between lip color types. Therefore, by determining the most likely color types as in implementation pattern 1, it is possible to make suggestions suitable for lip color types that do not have clear boundaries, compared to, for example, determining only one specific color type and suggesting makeup products and care methods according to that color type. In this way, using a mathematical model, multiple elements that make up a factor that belongs to one of the categories based on the input factors are classified and output according to the corresponding ratio.

[0055] <Implementation pattern 2> Figure 11 shows an overview of implementation pattern 2. In implementation pattern 2, based on a Gaussian Naive Bayes model created using measurements from the population (85 people in this embodiment), the ratio of the likelihood of each of the multiple elements (pink type, pale type, purple type, and brown type) that make up the color group A factor belonging to the color type category was calculated. Implementation pattern 2 is L, which belongs to the lip color category of the color information category. * , a * , b * , C *The input factors were the user's measurements (eight variables) of Age (which belongs to the age category of the physiological information category) and MI and EI (which belong to the pigment category). That is, one or more of the input or output factors (corresponding to the input and output factors in this implementation pattern) were lip-related factors (corresponding to the pigment category, lip color category, and color type category in this implementation pattern), and factors different from the input factors (corresponding to the pigment category and lip color category in this implementation pattern) were used as output factors (corresponding to the color type category in this implementation pattern), and the proportions of each of the multiple elements (pink type, pale type, purple type, and brown type in this implementation pattern) that make up the output factor, Color Group A, were calculated. As shown in Figure 11, the generalization performance of this model for the test data was high, with a correct answer rate of 0.76 (76%). In addition, the classification evaluation performed with this model also showed the highest rate of classification into the desired elements. And this input factor (L * :43, a * :14, b * :15, C *Using the variance function (e.g., :20, h:40, Age:20, MI):45, EI:73) and a mathematical model, the ratios of the various elements that make up color group A were calculated, and it was determined that the likelihood of the elements being included was pale, pink, brown, and purple, in that order. Using this result, for example, when recommending recommended makeup products (lipstick), it is possible to recommend not only makeup products suitable for pale-type users, but also makeup products suitable for pink-type users. By making such suggestions, even if a user prefers makeup products suitable for pink-type users (favorite colors) over makeup products suitable for pale-type users, the user will be able to more easily select makeup products suitable for pink-type users, making it possible to make appropriate suggestions. Furthermore, in the case of implementation pattern 2, since it is highly unlikely that the user's color type is purple, it is also possible to suggest that it would be better not to select makeup products suitable for purple-type users. As mentioned above, there is no clear boundary between lip color types, so by finding the most likely color type, as in Implementation Pattern 2, it is possible to make suggestions that are appropriate for lip color types that do not have clear boundaries, compared to, for example, finding only one specific color type and suggesting makeup products or care methods that correspond to that color type. In this way, using a mathematical model, multiple elements that make up a factor that belongs to one of the categories based on the input factors are classified and output according to the corresponding ratio.

[0056] <Implementation pattern 3> FIG. 12 shows an overview of implementation pattern 3. In implementation pattern 3, based on a Gaussian Naive Bayes model created using measurements from the population (85 people in this embodiment), the ratio of the likelihood of each of the multiple elements (20s, 30s, 40s, 50s, 60s, 70s) that make up the age factor belonging to the age category of the physiological information category was calculated. Implementation pattern 3 is L, which belongs to the lip color category of the color information category. * and C *The input factors were the user's measured values ​​(eight variables) of MBR, Skew, ATI, and RI, which belong to the blood flow category of physiological information, and MI and EI, which belong to the pigment category. In other words, at least one of the input or output factors (corresponding to the input factors in this implementation pattern) was a lip-related factor (corresponding to the lip color category, blood flow category, and pigment category in this implementation pattern), and a factor different from the input factors (corresponding to the lip color category, blood flow category, and pigment category in this implementation pattern) was used as the output factor (corresponding to the age category in this implementation pattern), and the proportion of each of the multiple elements (in this implementation pattern, 20s, 30s, 40s, 50s, 60s, and 70s) that make up the age factor, which is the output factor, was calculated. As shown in Figure 12, the generalization performance of this model for the test data was 0.47 (47%). In addition, the classification evaluation performed by this model showed that the percentage of classifications into the desired elements was the highest. * :44.824, and C * Using the mathematical model (Figure 1), we calculated the output factors, revealing a significantly higher ratio for users in their 40s and a lower ratio for users in their 30s and 20s. Since the user's actual age is 42, this result is highly accurate as a ratio of multiple factors that make up the output factors. Using this result, for example, when recommending makeup products (lipstick), we can recommend not only makeup products suitable for users in their 40s, but also makeup products suitable for users in their 30s and 20s. By making such suggestions, even if a user prefers makeup products (preferred colors) for users in their 30s and 20s over makeup products suitable for users in their 40s, the user will be able to more easily select makeup products suitable for users in their 30s and 20s, thereby enabling more appropriate suggestions. In addition, in the case of implementation pattern 3, since the user's age group is unlikely to be in their 50s, 60s, or 70s, it is possible to suggest that it would be better not to select makeup products suitable for people in their 50s, 60s, and 70s. In this way, using a mathematical model, multiple elements that make up a factor that belongs to one of the categories based on the input factors are classified and output according to the corresponding ratio.

[0057] <Implementation pattern 4> FIG. 13 shows an overview of implementation pattern 4. In implementation pattern 4, based on a Gaussian Naive Bayes model created using measurements from the population (85 people in this embodiment), the ratio of the likelihood of each of the multiple elements (pink type, pale type, purple type, and brown type) that make up the color group A factor belonging to the color type category was calculated. In implementation pattern 4, R, G, and B belonging to the lip color category in the color information category were used as input factors, and the proportions of multiple elements (corresponding to pink type, pale type, purple type, and brown type) constituting color group A belonging to the color type category in the physiological information category were calculated. That is, one or more of the input or output factors (corresponding to the input factors in this implementation pattern) were lip-related factors (corresponding to the lip color category and color type category in this implementation pattern), and a factor different from the input factor (corresponding to the lip color category in this implementation pattern) was used as the output factor (corresponding to the color type category in this implementation pattern), and the proportions of each of the multiple elements (in this implementation pattern, pink type, pale type, purple type, and brown type) constituting the color group A factor, which is the output factor, were calculated. As shown in Figure 13, the generalization performance of this model for the test data was high, with a correct answer rate of 0.71 (71%). In addition, the evaluation of classification performed with this model also showed the highest rate of classification into the desired elements. Using these input factors (R: 125, G: 76, B: 70) and a mathematical model, the ratios of the multiple elements that make up the color group A factor were calculated, and the results showed that the likelihood of belonging to the pink type was highest, followed by the pale type, purple type, and brown type. Using this result, for example, when recommending recommended makeup products (lipstick), it is possible to recommend not only makeup products suitable for pink-type users but also makeup products suitable for pale-type users. By making such suggestions, even if a user prefers makeup products suitable for pale-type users (preferred colors) over makeup products suitable for pink-type users, the user can more easily select makeup products suitable for pale-type users, making it possible to make appropriate suggestions. Furthermore, in the case of implementation pattern 4, since the user's color type is unlikely to be brown, it is also possible to suggest that it would be better not to select makeup products suitable for brown-type users. As mentioned above, there is no clear boundary between lip color types, so by finding the most likely color type, as in Implementation Pattern 4, it is possible to make suggestions that are appropriate for lip color types that do not have clear boundaries, compared to, for example, finding only one specific color type and suggesting makeup products or care methods that correspond to that color type. In this way, a plurality of factors belonging to the lip color category are used as input factors, and the proportions to which each of a plurality of elements constituting a factor belonging to the color type category corresponds are calculated and output.

[0058] <Implementation pattern 5> FIG. 14 shows an overview of implementation pattern 5. In implementation pattern 5, based on a Gaussian Naive Bayes model created using measurements from a population (85 people in this embodiment), the ratio of the likelihood of each of the multiple elements (20s, 30s, 40s, 50s, 60s, 70s) that make up the age factor belonging to the age category of the physiological information category was calculated. Implementation pattern 5 is L, which belongs to the lip color category of the color information category. * , a * , and b *The user's measurements (three variables) were used as input factors, and the ratios of multiple elements constituting the age factor belonging to the age category included in the physiological information category (corresponding to 20s, 30s, 40s, 50s, 60s, and 70s in this implementation pattern) were calculated. In other words, one or more of the input or output factors (corresponding to the input factor in this implementation pattern) was a lip-related factor (corresponding to the lip color category in this implementation pattern), and a factor different from the input factor (corresponding to the lip color category in this implementation pattern) was used as the output factor (corresponding to the age category in this implementation pattern), and the ratios of each of the multiple elements constituting the age factor, which is the output factor (corresponding to 20s, 30s, 40s, 50s, 60s, and 70s in this implementation pattern), were calculated. As shown in Figure 14, the generalization performance of this model for the test data was 0.41 (41%). In addition, the classification evaluation performed by this model showed that the percentage of classifications into the desired elements was the highest. * :43.5, a * :15.1, and b * Using the hypothesis model (13.1) and a mathematical model, the output factors were determined, revealing a high ratio for those in their 40s, followed by those in their 30s, and low ratios for those in their 20s, 60s, and 70s. Since the user's actual age is 38, this result can be considered highly accurate as a value for the multiple elements that make up the age factor. Using this result, for example, when recommending makeup products (lipstick), it is possible to recommend not only makeup products suitable for users in their 40s, but also makeup products suitable for users in their 30s. By making such suggestions, if only the single element with the highest ratio (in this case, 40s) were output, the output would suggest makeup products suitable for users in their 40s. However, by examining the ratios for each of the multiple elements, the next highest ratio would suggest makeup products suitable for users in their 30s. As a result, it is possible to suggest makeup products that match the subject's actual age (38 years old in this case), making appropriate suggestions. In addition, in the case of implementation pattern 5, since it is unlikely that the user is in their 60s or 70s, it is also possible to suggest that it would be better not to select makeup products suitable for people in their 60s and 70s. In this way, a plurality of factors belonging to the lip color category are used as input factors, and the proportions of a plurality of elements constituting the generation factor belonging to the age category are calculated and output.

[0059] <Implementation pattern 6> FIG. 15 shows an overview of implementation pattern 6. In implementation pattern 6, based on a Gaussian Naive Bayes model created using measurements from the population (in this embodiment, evaluation results from 31 people), the ratio of the likelihood of each of the multiple elements (0: positive (applies), 1: negative (does not apply)) constituting the factor of bright colors (bright skin color) belonging to the body visual perception category of the sensory information category was calculated. Implementation pattern 6 is L, which belongs to the lip color category of the color information category. * , a * , and b * The user's measured values ​​(three variables) were used as input factors, and the ratios of multiple elements (corresponding to 0: positive (applies), 1: negative (does not apply)) that make up the factor for bright colors belonging to the body visual category included in the sensory information category were calculated. In other words, at least one of the input factors or output factors (corresponding to the input factors in this implementation pattern) is a factor related to the lips (corresponding to the color information category in this implementation pattern), and a factor different from the input factor (corresponding to the color information category in this implementation pattern) is used as the output factor (corresponding to the body visual category in this implementation pattern), and the ratios of each of the multiple elements (corresponding to 0: positive (applies), 1: negative (does not apply)) that make up the output factor for bright colors (corresponding to 0: positive (applies), 1: negative (does not apply) in this implementation pattern) were calculated. As shown in Figure 15, the generalization performance of this model for the test data was 0.73 (73%). * :49, a * :43, and b *:27) and mathematical models were used to calculate the output factors, and the results showed a high positive ratio and a low negative ratio. In this way, it is possible to suggest to a user who wants to make their skin appear brighter that they can do so by changing the color of their lips without changing their foundation. In this way, a plurality of factors belonging to the lip color category are used as input factors, and the corresponding ratios of a plurality of elements constituting the bright color factor belonging to the body visual perception category are calculated and output.

[0060] <Implementation pattern 7> Figure 16 shows an overview of Implementation Pattern 7. In implementation pattern 7, based on a Gaussian Naive Bayes model created using measurements from the population (85 people in this embodiment), the ratio of the likelihood of each of the multiple elements (pink type, pale type, purple type, and brown type) that make up the color group A factor belonging to the color type category was calculated. In implementation pattern 7, MI and EI, which belong to the pigment category of the physiological information category, were used as input factors, and the proportions of multiple elements (equivalent to pink, pale, purple, and brown types) that make up color group A, which belongs to the color type category, were calculated. That is, at least one of the input or output factors (equivalent to the input and output factors in this implementation pattern) was a factor related to the lips (equivalent to the pigment category and color type category in this implementation pattern), and a factor different from the input factor (equivalent to the pigment category in this implementation pattern) was used as the output factor (equivalent to the color type category in this implementation pattern), and the proportions of each of the multiple elements (equivalent to pink, pale, purple, and brown types in this implementation pattern) that make up the output factor, color group A, were calculated. As shown in Figure 16, the generalization performance of this model for the test data was high, with a correct answer rate of 0.53 (53%). In addition, the classification evaluation performed with this model also showed the highest rate of classification into the desired elements. Using these input factors (MI: 39.9, EI: 79.5) and a mathematical model, the ratios of the multiple elements constituting the color group A factor were calculated, and the results showed that the likelihood of belonging to the pink type was highest, followed by the pale type, purple type, and brown type. Using this result, for example, when recommending recommended makeup products (lipstick), it is possible to recommend not only makeup products suitable for pink-type users but also makeup products suitable for pale-type users. By making such suggestions, even if a user prefers makeup products (favorite colors) suitable for pale-type users over makeup products suitable for pink-type users, the user can easily select makeup products suitable for pale-type users, thereby making it possible to make appropriate suggestions. Furthermore, in the case of Implementation Pattern 7, since it is highly unlikely that the user's color type is brown, it is also possible to suggest that it is better not to select makeup products suitable for brown-type users. As mentioned above, there is no clear boundary between lip color types. Therefore, by determining the most likely color type, as in Implementation Pattern 7, it is possible to make suggestions suitable for lip color types that do not have clear boundaries, compared to, for example, determining only one specific color type and suggesting makeup products and care methods according to that color type. In this way, using a mathematical model, multiple elements that make up a factor that belongs to one of the categories based on the input factors are classified and output according to the corresponding ratio. In this way, a plurality of factors belonging to the pigment category are used as input factors, and the proportions to which each of a plurality of elements constituting a factor belonging to the color type category corresponds are calculated and output.

[0061] <Implementation pattern 8> Figure 17 shows an overview of Implementation Pattern 8. In implementation pattern 8, based on a Gaussian Naive Bayes model created using measurements from the population (85 people in this embodiment), the ratio of the likelihood of each of the multiple elements (pink type, pale type, purple type, and brown type) that make up the color group A factor belonging to the color type category was calculated. Implementation pattern 8 is L, which belongs to the lip color category of the color information category. * , C * Using and h as input factors, the proportions of the multiple elements (corresponding to pink, pale, purple, and brown types) that make up color group A, which belongs to the color type category included in the physiological information category, were calculated. That is, at least one of the input factors or output factors (corresponding to the input and output factors in this implementation pattern) is a factor related to the lips (corresponding to the lip color category and color type category in this implementation pattern), and a factor different from the input factor (corresponding to the lip color category in this implementation pattern) is set as the output factor (corresponding to the color type category in this implementation pattern), and the proportions of the multiple elements (in this implementation pattern, pink, pale, purple, and brown types) that make up the color group A factor, which is the output factor, that apply were calculated. As shown in Figure 17, the generalization performance of this model for the test data was high, with a correct answer rate of 0.53 (53%). In addition, the classification evaluation performed with this model also showed the highest rate of classification into the desired elements. And this input factor (L * :41.1, C *Using a mathematical model (h:17.6 and h:38.7), the ratios of the multiple elements that make up the color group A factor were calculated, and the results showed that the likelihood of belonging to the purple type was highest, followed by the brown type, the pale type, and the pink type. Using these results, for example, when recommending recommended makeup products (lipstick), it is possible to recommend not only makeup products suitable for purple type users but also makeup products suitable for brown type users. By making such suggestions, even if a user prefers makeup products suitable for brown type users (preferred colors) over makeup products suitable for purple type users, the user can more easily select makeup products suitable for brown type users, making it possible to make appropriate suggestions. Furthermore, in the case of implementation pattern 8, since it is highly unlikely that the user's color type is pink or pale, it is also possible to suggest that it would be better not to select makeup products suitable for pink or pale types. As mentioned above, there is no clear boundary between lip color types, so by finding the most likely color type, as in Implementation Pattern 8, it is possible to make suggestions that are appropriate for lip color types that do not have clear boundaries, compared to, for example, finding only one specific color type and suggesting makeup products or care methods that correspond to that color type. In this way, a plurality of factors belonging to the lip color category are used as input factors, and the proportions to which each of a plurality of elements constituting a factor belonging to the color type category corresponds are calculated and output.

[0062] As described above, this method can determine various output factors based on various input factors. As in implementation patterns 1 to 8, by determining the ratio of each of the multiple elements that make up the output factor and determining which element is most likely, or an element that is not the most likely but has a certain degree of possibility, the results can be used to suggest makeup products or skin care methods that are more appropriate than if only one of the multiple elements that make up the output factor were determined. Furthermore, for example, even if the user does not have measured values ​​for all of the factors listed in Table 2, it is possible to determine a specific factor as the output factor.

[0063] <Lip Evaluation System 10> A lip evaluation system 10 that executes the lip evaluation method described above and evaluates the lip features or age of a subject using factors different from the input factors as output factors will be described. Fig. 18 shows a conceptual diagram of the lip evaluation system 10. The lip evaluation system 10 (hereinafter also referred to as the present system) is an information processing terminal capable of executing various processes, and includes a memory unit 11, an input unit 12, a calculation unit 13, an output unit 14, and an evaluation unit 15. Although not shown, the lip evaluation system 10 also includes an input device such as a keyboard and pointing device, an arithmetic processing unit, a memory unit, a display device, etc.

[0064] The memory unit 11 is a storage means for storing a mathematical model that is obtained in advance from a population and indicates correlations between multiple input factors and output factors. In addition to the mathematical model, the memory unit 11 also stores programs and data required to execute the present method. The input unit 12 is an input means for inputting a plurality of input factors acquired from the subject, and corresponds to an input device provided in the information processing terminal. The calculation unit 13 is a calculation means for calculating output factors based on a mathematical model and input factors, and corresponds to an arithmetic processing unit provided in the information processing terminal. The output unit 14 is an output means for outputting the ratios of the multiple elements that make up the calculated output factor. This corresponds to outputting to a display device provided in the information processing terminal, but also includes outputting to a device or system other than the lip evaluation system 10 or outputting to a printer. The evaluation unit 15 is an evaluation means that uses the output factors to suggest makeup products and lip care products, and makeup methods and care methods. It corresponds to a calculation processing device provided in the information processing terminal. Depending on the respective ratios of the multiple elements that make up the output factors calculated by the calculation unit 13, the evaluation unit 15 may, for example, extract and suggest a recommended lip cosmetic from among multiple lip cosmetics with different color patterns, or extract and suggest a recommended lip care method from among multiple lip care methods.

[0065] Here, we will explain the lip cosmetics recommended when the corresponding ratios of the multiple elements constituting the color group A factor are determined as output factors using implementation pattern 1, 2, 4, 7, or 8. The lip cosmetics recommended in this embodiment are lip cosmetics (makeup products) that are applied to the lips, such as lipstick, colored lip balm, lip gloss, lip pencil, lip liner, and powder lip. Table 3 shows the recommended makeup product types for each color type.

[0066] [Table 3]

[0067] Makeup products are intended to be enjoyed by beautifully changing the color of lips, but a common finish is often to impart a reddish hue to improve the facial impression. In this embodiment, we propose a makeup product that is applied to bare lips to impart a reddish hue. The pink type has a bright red lip color, so it doesn't matter what type of makeup product you choose. You can simply suggest that the user choose makeup products based on their preferences. Since the pale type has a lip color that is not reddish, low-coverage makeup products will result in the lip color being the same as the natural lip color, i.e., not reddish, even when applied, so low-coverage makeup products are not recommended. The purple type has a lip color with low saturation, so even if makeup is applied with low saturation or low coverage makeup products, the lip color will end up being the natural lip color, that is, a lip color with low saturation and no reddishness, so low saturation or low coverage makeup products are not recommended. Since the brown type has a dark lip color, high-brightness makeup products, low-saturation makeup products, and low-coverage makeup products will result in the natural lip color, i.e., a dark, dull lip color, even when applied, so high-brightness makeup products, low-saturation makeup products, and low-coverage makeup products are not recommended.

[0068] Next, we will explain the recommended care method when the ratio of each of the multiple elements that make up the color group A factor is obtained as the output factor using implementation pattern 1, 2, 4, 7, or 8. Table 4 shows recommended care methods for each color type.

[0069] [Table 4]

[0070] Generally, bright red lips are considered to be healthy and beautiful. Bright red lips have good blood flow, reach the periphery, and have little melanin pigment. Therefore, in this embodiment, we propose a care method aimed at achieving a bright red lip color. Here, care target: blood flow refers to care that improves blood flow to the lips, and includes massage, moisturizing care, hot compresses, lip packs, carbonated preparations, warming preparations, etc. Care target: melanin refers to care to reduce melanin pigment, and includes care agents that reduce melanin pigment, cleansing agents, UV protection preparations, brightening preparations, etc. Pink types have good blood flow and little melanin pigment, so you can suggest that there is no need for blood flow care or melanin care. However, there is no problem with taking preventative measures such as blood flow care or melanin care, even if they are not necessary. Pale types have less melanin pigment, but have poor blood flow, making it difficult for blood to reach the extremities, so blood flow care should be recommended. The purple type has little melanin pigment, but has poor blood flow and does not reach the extremities, so blood flow care should be recommended. Brown types have a lot of melanin pigment, so melanin pigment care should be recommended.

[0071] Next, Table 5 shows the difference in the content of suggestions between when makeup products and care methods are suggested based on the ratios of multiple elements that make up the output factor (referred to as "when based on ratios") and when one element is identified based on the ratios of multiple elements that make up the output factor and the recommended care method or makeup method is suggested using the identified element (referred to as "when based on one element"). When identifying one element based on the ratios of multiple elements, the component with the highest / lowest ratio can be identified, or it can be identified by a staff member at a store or a user.

[0072] [Table 5]

[0073] When making makeup recommendations based on the ratios in Table 5, if the ratio is 60% or higher, the makeup product types shown in Table 3 are recommended; if it is 30% or higher but less than 60%, they are partially recommended; and if it is less than 30%, they are not recommended. Note that partially recommended means that within that type of makeup, some products are recommended and some are not, depending on the color. In addition, when suggesting skin care methods, if the ratio is 60% or higher, the skin care methods shown in Table 4 are actively recommended; if it is 30% or higher but less than 60%, they are recommended; and if it is less than 30%, they are recommended as a preventative measure. As is clear from Table 5, even for the same user, the makeup products and skin care methods recommended will differ depending on whether they are based on a ratio or a single element. As mentioned above, there are no clear boundaries between lip color types, so when there are multiple possible color types, it is possible to make a wider range of recommendations that are more suited to lip color by recommending makeup products and skin care methods according to the ratio that each of the multiple elements that make up the output factor corresponds to, rather than specifying one of the multiple elements that make up the output factor. Furthermore, if a recommended care method is performed based on the ratio and then performed again based on the ratio after a specified period of time, even if the type of element (color type) of the output factor remains unchanged, the ratio may change. Furthermore, the recommended makeup products and care methods may also change depending on the ratio. In this way, by determining the corresponding ratios for each of multiple elements and proposing recommended makeup products and care methods based on those ratios, it is possible to make appropriate suggestions during the process of lip color change (improvement).

[0074] <Modification> In this embodiment, recommended makeup products are suggested based on the output factors, but recommended makeup methods may also be suggested based on the output factors.Furthermore, while recommended skin care methods are suggested based on the output factors, recommended skin care products may also be suggested based on the output factors.

[0075] It is possible to calculate the ratio of the likelihood of each of the elements (pink type, pale type, purple type, and brown type) that make up color group A belonging to the color type category based on a Gaussian Naive Bayes model created using measurements from a population (85 people in this embodiment) using combinations of input factors other than those used in implementation patterns 1, 2, 4, 7, and 8. Table 6 shows other examples of combinations of input factors.

[0076] [Table 6]

[0077] The lip evaluation system 10 may be provided with a photographing device (not shown) for photographing the user's lips, and values ​​serving as input factors may be measured and input from the image photographed by the photographing device. This makes it possible to evaluate the lip condition at the time of the desired evaluation. Alternatively, a photographed image of the user's lips may be acquired outside the lip evaluation system 10, and values ​​serving as input factors may be measured and input from the acquired image. This allows, for example, a lip image photographed by the user to be sent to the lip evaluation system 10, and evaluation may be performed based on that data, thereby improving convenience. Furthermore, convenience can be further improved by transmitting the evaluation results to an information processing terminal (not shown) owned by the user and providing the evaluation results to the user.

[0078] In the lip evaluation system 10 of this embodiment, the values ​​of the input factors are input by an input means, but this is not limiting. For example, as described above, the values ​​that become the input factors may be measured from an image captured by a photographing device or an acquired image and input into the calculation means, and this input process may also be the input means. [Explanation of symbols]

[0079] 10 Lip Evaluation System 11 Storage section 12 Input section 13 Calculation section 14 Output section 15 Evaluation Section

Claims

1. A method for evaluating lip or body features of a subject using one or more factors selected from a category group consisting of a physiological information category using physiological features related to the lips or body of the subject as factors, a sensory information category using sensory features related to the lips or body of the subject as factors, and a color information category using color information related to the lips or body of the subject as factors as input factors, and one or more factors different from the input factors as output factors, comprising: Executed by at least one information processing terminal, at least one of the input factors or the output factors is a factor related to the lips, Each of the factors is composed of a plurality of elements that indicate the degree of the factor or classify the factor, A lip evaluation method characterized by calculating the ratio of each of the multiple elements that make up the output factor based on a mathematical model that is obtained in advance from a population and shows the correlation between the input factors and the output factors, and the input factors obtained from the subject, and evaluating the lip or physical characteristics of the subject based on the calculation result and data that shows the relationship between the multiple elements that make up the output factor and the evaluation content.

2. The physiological information categories include a pigment category using lip pigment information as a factor, a blood flow category using lip blood flow information as a factor, a blood vessel category using lip blood vessel information as a factor, a shape category using lip shape information as a factor, a surface shape index category using lip surface shape index information as a factor, a color type category using lip color type information as a factor, a body category using the physiological characteristics of the subject's body as a factor, an age category using the subject's age information as a factor, or a sex category using the subject's sex information as a factor, The sensory information categories include a lip visual category using lip visual information as a factor, a lip tactile category using lip tactile information as a factor, a lip somatic category using lip somatic sensation as a factor, a body visual category using body visual information as a factor, a body tactile category using body tactile information as a factor, or a body somatic sensation category using body somatic sensation as a factor, 2. The lip evaluation method according to claim 1, wherein the color information categories include a lip color category using lip color information as a factor or a body color category using body color information as a factor.

3. a plurality of factors belonging to the lip color category are used as the input factors; The lip evaluation method according to claim 2, wherein the ratio of each of the elements constituting the factor belonging to the color type category is calculated, and the lip or physical characteristics of the subject are evaluated based on the calculation results.

4. a plurality of factors belonging to the pigment category are used as the input factors; 4. A lip evaluation method according to claim 2, further comprising the steps of: calculating the ratio of each of a plurality of elements constituting the factor belonging to the color type category; and evaluating the lip or physical characteristics of the subject based on the calculation results.

5. a plurality of factors belonging to the lip color category are used as the input factors; A lip evaluation method according to any one of claims 2 to 4, which calculates the ratio of each of the multiple elements that make up the factor belonging to the age category, and evaluates the lip or physical characteristics of the subject based on the calculation results.

6. a plurality of factors belonging to the lip color category are used as the input factors; A lip evaluation method described in any one of claims 2 to 5, which calculates the corresponding ratios of multiple elements that make up the factor belonging to the body visual category, and evaluates the subject's lip or physical characteristics based on the calculation results.

7. A system for evaluating lip or body features of a subject using one or more factors selected from a category group consisting of a physiological information category using physiological features related to the lips or body of the subject as factors, a sensory information category using sensory features related to the lips or body of the subject as factors, and a color information category using color information related to the lips or body of the subject as factors as input factors, and one or more factors different from the input factors as output factors, at least one of the input factors or the output factors is a factor related to the lips, Each of the factors is composed of a plurality of elements that indicate the degree of the factor or classify the factor, a calculation means for calculating a ratio corresponding to each of the plurality of elements constituting the output factor based on a mathematical model previously obtained from a population and showing a correlation between the input factor and the output factor, and the input factor obtained from the subject; A lip evaluation system characterized by comprising an evaluation means for evaluating the lip or physical characteristics of the subject based on the calculation results and data showing the relationship between the multiple elements constituting the output factor and the evaluation content.

Images