JP2005296280A - How to create a photo scale - Google Patents

Abstract

PROBLEM TO BE SOLVED: To create a photoscale that obtains evaluation results with good reproducibility.
SOLUTION: A method for creating a photoscale comprising a skin photograph used as an evaluation index at each stage when the specific characteristics of the skin are divided into a plurality of stages is classified into two or three groups according to the specific characteristics of the skin. Categorizing at least one group into two or three groups according to the skin properties, repeating the same categorization if necessary, extracting representative photographs from each stage group, It consists of the process of making a photo scale reference photo.
[Selection] Figure 1A

Description

  The present invention relates to a method for creating a photoscale used as an evaluation index for specific properties of skin, and a skin evaluation method using the method.

  A photoscale is used as an evaluation index of specific properties such as skin spots and wrinkles (Non-patent Document 1, Patent Document 1). The photoscale is composed of skin photographs that serve as evaluation indices for each stage when the specific properties of the skin are divided into a plurality of stages.

  In addition, as an evaluation index for skin spots, etc., the area of a spot portion extracted from a skin photograph by image processing is also used. Since the stain position information is not included, for example, a stain located in a conspicuous part of the face and a spot located in a conspicuous part receive the same evaluation. Therefore, the evaluation result by image processing and the impression of appearance may be different. On the other hand, according to the photoscale, since the stain is evaluated from the impression of the entire face photograph, such a problem does not occur.

  Conventionally, as a method of creating a photoscale for specific skin properties, the photoscale creator first sets the words that are the criteria for each stage when the skin specific properties are classified into multiple stages, and the criteria A skin photograph that applies to the words to be applied is subjectively selected from a large number of skin photographs, or a skin photograph that applies to the words to be used as a criterion is synthesized by image processing.

British Journal of Dermatology, 130,167-173 (1994) Japanese Patent Laid-Open No. 2000-212034, claim 3

  However, if a skin photo that meets the criteria set in advance in words is selected from a number of skin photos, or a photo scale created by combining images is used to evaluate the subject's specific skin characteristics, The reproducibility of the skin evaluation results obtained by is poor, the evaluation of the same skin photo varies depending on the evaluator, and even when the same evaluator evaluates the same skin photo, if the evaluation is repeated multiple times, There was a problem that the evaluation result of each time was not constant.

  In contrast, an object of the present invention is to obtain a photoscale that can obtain an evaluation result with good reproducibility.

  In creating the photoscale, the present inventor first observes, for example, about 100 skin photographs as a whole by the photoscale creator, and then determines the skin photograph according to the specific property of the skin of interest. Dividing into 3 groups, dividing the divided at least one group into 2 or 3 groups according to the specific properties, and creating a photo scale by selecting a representative skin photo from each evaluation stage group, It has been found that when the skin property of an arbitrary subject is evaluated using the obtained photoscale, a reproducible evaluation result can be obtained.

That is, the present invention is a method for creating a photoscale consisting of a skin photograph used as an evaluation index for each stage when the specific property of the skin is divided into a plurality of stages,
A step of dividing a plurality of skin photographs into 2 or 3 groups according to the specific properties of the skin, a step of further dividing at least one divided group into 2 or 3 groups according to the skin properties, and a group of each stage The present invention provides a method for creating a photoscale, which includes a step of extracting representative photographs from the image and using them as reference photographs of the photoscale.

  The present invention also provides a photoscale created by this method, and further provides a skin evaluation method for evaluating the skin of a subject by comparing the photoscale with a photograph of the subject's skin.

  When the specific property of the subject's skin is evaluated from the subject's skin photograph, the reproducibility of the evaluation result can be improved by using the photoscale created by the method of the present invention. That is, even when the evaluator is different, a specific evaluation result can be obtained with high reproducibility. In particular, a reproducible evaluation result can be obtained not only when the expert judges the skin property of the subject but also when the general evaluator judges.

  Further, even when the same evaluator repeats the evaluation for the same skin photograph a plurality of times, a specific evaluation result can be obtained with high reproducibility.

  Therefore, for example, it is possible to evaluate the effect of the whitening agent against the stain by evaluating the skin stain state when the whitening agent is continuously applied to the skin using the photoscale of the present invention.

  Moreover, it becomes possible to investigate the seasonal change of the appearance of a spot by evaluating the skin spot condition for every season using the photo scale of the present invention.

  Hereinafter, the present invention will be described in detail with reference to the drawings. In addition, the same code | symbol represents the same or equivalent component.

  The photoscale of the present invention is used for evaluating the characteristic properties of skin such as skin spots, freckles, bears, post-inflammation pigmentation, etc., texture, wrinkles, sagging, pores, etc. It consists of a skin photograph that is an evaluation index for each stage when the specific property is divided into a plurality of stages (usually 2 to 11 stages).

  1A to 1C are explanatory diagrams of a method for creating a photoscale of one embodiment of the present invention. In this method, a plurality of skin photographs of a specific part are collected from a plurality of randomly extracted photograph providers.

  The part where the skin photograph is taken is appropriately determined according to the property of the target skin. For example, when creating a photoscale such as a stain, wrinkle, sagging, etc., as shown in FIG. 2, it may be selected from the face 1, the chest 2, the left and right arms 3, the left and right hand backs 4, and the like.

  In addition, how to take a picture is determined as necessary. For example, when creating a photoscale for evaluating the specific properties of the skin with an impression by normal visual observation, it is sufficient to take a skin photograph using normal light as illumination light, and a preliminary image that cannot be observed by normal visual observation. In the case of evaluation including a general stain or the like, a skin photograph may be taken using UV light or polarized light as illumination light.

  The total number of skin photos to be collected is such that the photo scale creator can visually grasp the total number of individual skin photos with respect to the specific properties by visually observing the total number. When it is classified into 2 or 3 according to the specific property, it is preferable that the individual skin photograph can be grasped to which category it belongs. If the number is too large, it will be difficult for a single photoscale creator to grasp the whole, and if the number is too small, it will be difficult to create a photoscale that can obtain evaluation results with good reproducibility. Therefore, the total number of skin photographs is preferably 30 to 200, and more preferably 50 to 100.

  In the method for creating a photoscale of the present invention, a photoscale creator, preferably a photoscale creator who is an expert in evaluating skin properties, observes the above-mentioned plurality of skin photographs, and first creates the specific property. The process of dividing into 2 or 3 groups is performed accordingly.

  For example, when the whole of the number of spots is divided into two groups, it is divided into two groups: a group with many spots and a group with few spots. Further, when dividing the entire number of spots into three groups, the lowest rank group with the least spots and the highest rank group with the most spots are extracted, and the rest are set as the intermediate rank groups (FIG. 1A). Thus, by setting the number of divisions as 2 or 3, it is possible to create a photoscale that can obtain evaluation results with good reproducibility.

  Next, at least one group divided into 2 or 3 groups is further divided into 2 or 3 groups according to the skin properties. Preferably, when first divided into two groups, each group is divided into two or three groups as described above. Or when dividing into 3 groups first, the group of an intermediate rank is divided into 2 or 3 groups in the above-mentioned way (Drawing 1B).

  If the properties in each group divided at this stage do not vary greatly, the operation of dividing the original skin photograph may be finished here, and the skin properties vary widely in each group If there is, the operation of dividing the group into two or three groups may be repeated (FIG. 1C). It is preferable that the number of stages of skin properties to be divided in this way is 2 to 11 stages as a whole.

  Next, the photoscale creator extracts one representative photo that is considered to be representative of the group from the group of each stage divided as described above. Then, from the representative photographs of each group, for example, a photo scale is constructed as shown in FIGS. 3A, 3B, 3C, and 3D.

  The photoscale of the present invention can be used in the same manner as a known photoscale. Therefore, it can be used in a method for evaluating the specific properties of the subject's skin using this photoscale. In this case, the subject's skin photo is taken at the same site as the photoscale and the same as the photoscale. By comparing the skin photograph obtained under the conditions and the photo scale, it is determined where the skin property of the subject belongs to the evaluation rank divided into a plurality of stages. As a result, reproducible evaluation results can be obtained not only when the expert judges the skin properties of the subject but also when the general evaluator judges.

Examples 1 to 10
(1) Creation of a photo scale Skin photos of 85 white women in their 20s to 60s (16-21 people of each age) randomly selected from August 20-30, living in Cincinnati, Ohio, USA Was taken as follows. In this case, as shown in FIG. 2, the imaging part is a part of the face (front), chest, both arms, and both hands, and UV light, polarized light, or normal light (strobe light) is taken as illumination light. Output as an image. Table 1 shows the number and average age for each of these 85 people.

[Table 1]

  The obtained photograph is spread on a desk surface under constant illumination for each photographing part and each kind of illumination light, and a professional evaluator for skin spots has the least spots from each group as shown in FIG. 1A. Select the group with the most spots and then select the group with the least spots and the group with the most spots from the remaining photographs as shown in FIG. 1B. It was divided into. Then, as shown in FIG. 1C, the remaining group was divided into two groups, a group with few spots and a group with many spots, and was divided into a total of six groups. The 6 groups thus obtained were assigned a score of 0 to 5 from the least spot.

  From each group, one photo that is considered to be representative of the level of spots was selected, a photo scale of a face photo illuminated with UV light (Example 1), and a photo scale of an arm photo illuminated with UV light. (Example 2), photo scale of a face photograph illuminated with polarized light (Example 3), photo scale of a face photograph illuminated with normal light (Example 4), photo scale of a chest photograph illuminated with polarized light (Example 5) ), Photo scale of chest photo illuminated with normal light (Example 6), photo scale of arm photo illuminated with polarized light (Example 7), photo scale of arm photo illuminated with normal light (Example 8), polarized light A photo scale (Example 9) of the back of the hand illuminated with a light and a photo scale (Example 10) of the back of the hand illuminated with normal light were prepared.

(2) Photoscale evaluation (2-1) Score by expert evaluator and κ value of score by general evaluator Using the photoscale of Example 1 for five general evaluators, use it to create the photoscale The skin photograph was evaluated in 6 levels (score 0 to score 5). Then, the score obtained from the general evaluator and the score given by the professional evaluator at the time of creating the photoscale are summarized as shown in Table 2, and the κ value is used as an index indicating how much both scores match. It calculated | required like the Formula 1 of. Table 2 shows the relationship between a score by one general evaluator and a score by a professional evaluator.

[Table 2]

[Equation 1]
κ = (f ₀ −f _c ) / (N _t −f _c )
= (45-12.89) / (65-12.89)
= 0.616 (Formula 1)
Where
N _t : Total number of photos = 65
f ₀ : Number of photographs in which the score by the professional evaluator matches the score by the general evaluator f ₀ = 1 + 7 + 9 + 9 + 13 + 6 = 45
f _c : The overall ratio at which the score by the professional evaluator and the score by the general evaluator are expected to coincide by chance:
f _c = ((1 × 4) + (10 × 8) + (13 × 13) + (16 × 15) + (19 × 15) + (6 × 10)) / 65
= 12.89

  Other general evaluators similarly evaluated skin photographs, obtained κ values for each, and determined their average values. The results are shown in FIG.

  Similarly, the κ value of the score by the general evaluator and the score by the professional evaluator when evaluating the skin photograph used to create the photo scale using the photo scale of Example 2 to Example 10 is the same. Asked. The results are shown in FIG.

The interpretation of the κ value is generally as shown in Table 3.
[Table 3]

From FIG. 4, it was obtained using the photoscale of the present example except for the photoscale of the face photograph illuminated with normal light (Example 4) and the photoscale of the chest photograph illuminated with polarized light (Example 5). The κ values of the score by the general evaluator and the score by the professional evaluator were both 0.4 or more, and it was confirmed that the evaluation result when the photoscale of this example was used was reproducible.

(2-2) Relationship between the score of the stain and the age Plotting the relationship between the score of the stain and the age assigned by the expert evaluator to each skin photo in Examples 1 to 10 for each type of illumination light, 5 to 7 were obtained.

  From these figures, it can be seen that in skin photographs taken with normal light illumination, the spots tend to increase with age.

  Also, skin photographs taken with UV light or polarized light have higher scores for younger generation spots than skin photographs taken with normal light. In general, it is considered that a stain that is difficult to see visually is easily seen by illuminating with UV light or polarized light.

Example 11 and Comparative Example 1
A face photo was taken of 55 Japanese women using polarized light as illumination light, and a professional evaluator produced a photoscale from this face photo in the same manner as in Example 1.

  Using this photo scale, a general evaluator scores 55 face photographs. If the general evaluator gives it, the score and the score given by the professional evaluator at the time of creating the photo scale are used in the same manner as in the first embodiment. A κ value was obtained, and the degree of coincidence of both scores was examined (Example 11).

On the other hand, a professional evaluator selected a photo that seems to best fit the score words in the following Table 4 from the 55 face photographs used in Example 11, and created a photo scale of the comparative example (Comparative Example) 1).
[Table 4]

  Using the photo scale of this comparative example, each of the professional evaluator and the general evaluator was asked to score the 55 face photographs, and the κ value was determined in the same manner as in Example 1 from the score given by both, We examined how well both scores matched. The results are shown in FIG.

  When the photo scale of Example 11 is used, the κ value is 0.68, whereas when the photo scale of Comparative Example 1 is used, the κ value is 0.51, and the photo scale of Example 11 is It turns out that the reproducibility of the evaluation of the stain is higher when it is used.

Example 12
Randomly extracted 61 white women living in Cincinnati, Ohio, USA twice a year in summer and winter, using normal light (strobe light) for the face (front), chest, arms, and hands. A skin photograph was taken, and the resulting skin photograph was scored with a spot using the photocales prepared in Examples 4, 6, 8, and 10. For each measurement site between the same subjects, the summer score was subtracted from the winter score, and the average value of all subjects was obtained.

  The result is shown in FIG. From FIG. 9, it can be seen that using the photo scale of the present invention, it can be confirmed that the score of the spot is significantly reduced from summer to winter.

  The photoscale creation method of the present invention and the photoscale obtained by this method are used in various fields for evaluating skin properties, for example, development of beauty methods and cosmetics, advice to these customers, various effects on the skin. This is useful in the field of research on factors.

It is explanatory drawing of the production method of a photo scale. It is explanatory drawing of the production method of a photo scale. It is explanatory drawing of the production method of a photo scale. It is explanatory drawing of an imaging | photography site | part. It is a schematic diagram of a photo scale. It is a schematic diagram of a photo scale. It is a schematic diagram of a photo scale. It is a schematic diagram of a photo scale. It is the figure showing (kappa) value of the score by a professional evaluator, and the score by a general evaluator at the time of using the photo scale of each Example. It is a relationship figure of the score of skin photograph taken with UV light illumination, and age. It is a relationship figure of the score and age of a skin photograph taken with polarized light. It is a relationship figure of the score and age of a skin photograph taken with normal light illumination. It is a figure showing κ value of a score by a professional evaluator and a score by a general evaluator when the photo scale of Example 11 and the photo scale of Comparative Example 1 are used. It is a figure showing the difference of the score of a winter and summer spot.

Explanation of symbols

1 face 2 chest 3 arms 4 back of hand

Claims (6)

A method for creating a photoscale consisting of a skin photograph used as an evaluation index for each stage when the specific properties of the skin are divided into a plurality of stages,
A step of dividing a plurality of skin photographs into two or three groups according to the specific properties of the skin;
A photo comprising the steps of further dividing at least one group into two or three groups according to the skin properties, and extracting representative photographs from the groups at each stage and using them as photo scale reference photographs How to create a scale. The method for producing a photoscale according to claim 1, wherein the specific property is a stain. Dividing the plurality of skin photographs into three groups by extracting a photograph belonging to the lowest rank in the evaluation stage and a photograph belonging to the highest rank from the plurality of skin photographs;
A step of dividing the intermediate rank skin photos into three groups by extracting from the intermediate rank photos the skin photos belonging to the lowest rank and the photos belonging to the highest rank among the intermediate rank photos;
The photoscale according to claim 1 or 2, further comprising a step of distinguishing, from the remaining intermediate rank skin photographs, a photograph belonging to the lower rank and a photograph belonging to the upper rank among the skin photographs belonging to the intermediate rank. How to make. The photoscale creation method according to any one of claims 1 to 3, wherein a photoscale is created from a skin photograph taken using UV light or polarized light as an irradiation light source. A photoscale created by the method according to claim 1. A skin evaluation method for evaluating a specific property of a subject's skin by comparing the photoscale according to claim 5 with a skin photograph of the subject.

Images