JP7565182B2 - Skin condition evaluation method - Google Patents

Description

The present invention relates to a technology for evaluating skin condition.

One of the indices for evaluating skin condition is color difference. For example, this is used as an index when applying a cosmetic such as a liquid cosmetic or foundation with concealing power to skin that has color differences due to the presence or absence of spots or bruises, and comparing the color difference of the skin before and after application to measure the concealing effect of the applied liquid cosmetic or foundation (for example, Patent Document 1).

JP 2007-223976 A

In the case of Patent Document 1, evaluation is only done from the perspective of how much color difference in skin with color differences has been corrected. However, according to the inventors' research, in addition to skin color differences, it is also important to consider whether areas where cosmetics have been applied, such as blemishes or birthmarks, look natural in order to make the skin look beautiful, and the invention of Patent Document 1 does not provide an evaluation from this perspective.

The present invention was made in consideration of the above problems, and relates to a technology that makes it possible to evaluate the aesthetic beauty of skin from a perspective other than the conventional parallax. In this specification, "evaluation of skin condition" means evaluation of the apparent condition of the skin surface for non-medical purposes, and includes evaluations that can be performed even by evaluators who are not experts.

The present invention relates to a skin condition evaluation method including a skin image acquisition step of acquiring a skin image of a skin surface including a predetermined unevenness, an extraction step of extracting from the skin image a first evaluation area including the predetermined unevenness and a second evaluation area including at least an area other than the first evaluation area, a first variation calculation step of calculating a variation in pixel values of the first evaluation area, a second variation calculation step of calculating a variation in pixel values of the second evaluation area, a variation comparison step of comparing a calculation result in the first variation calculation step with a calculation result in the second variation calculation step, and an evaluation step of evaluating an apparent condition of the skin surface based on the calculation result in the variation comparison step.

The present invention also relates to a skin condition evaluation system including a skin image acquisition means for acquiring a skin image of a skin surface including a predetermined unevenness, an extraction means for extracting from the skin image a first evaluation area including the predetermined unevenness and a second evaluation area including at least an area other than the first evaluation area, a first variation calculation means for calculating a variation in pixel values of the first evaluation area, a second variation calculation means for calculating a variation in pixel values of the second evaluation area, a variation comparison means for comparing a calculation result by the first variation calculation means with a calculation result by the second variation calculation means, and an evaluation means for evaluating the apparent condition of the skin surface based on the calculation result by the variation comparison means.

The technology provided by the present invention makes it possible to more appropriately evaluate the aesthetic appearance of skin.

FIG. 13 is a diagram for comparing and explaining skin images, where (a) is an image in which the texture of bare skin has been enhanced, (b) is an image in which the texture of skin to which a certain formulation (Formulation A) has been applied has been enhanced, and (c) is an image in which the texture of skin to which another formulation (Formulation B) has been applied has been enhanced. 1 is a flowchart showing a skin condition evaluation method (present method) of the present embodiment. 13 is a flowchart showing an analysis method of steps S110 and S130. 3 is a graph created by the skin condition evaluation method according to the present embodiment. 3 is a graph created by the skin condition evaluation method according to the present embodiment. FIG. 1 is a block diagram of a skin condition evaluation system 100 according to the present embodiment. 3 is a graph created by the skin condition evaluation method according to the present embodiment. 3 is a graph created by the skin condition evaluation method according to the present embodiment.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
First, an overview of this embodiment will be described.
The skin condition evaluation method of this embodiment (hereinafter, sometimes referred to as the present method) includes a skin image acquisition step, an extraction step, a first variation calculation step, a second variation calculation step, a variation comparison step, and an evaluation step. The skin image acquisition step is a step of acquiring a skin image in which a skin surface including a predetermined unevenness is photographed. The extraction step is a step of extracting a first evaluation area including the predetermined unevenness and a second evaluation area including at least an area other than the first evaluation area from the skin image to be evaluated. The first variation calculation step is a step of calculating the variation of pixel values of the first evaluation area. The second variation calculation step is a step of calculating the variation of pixel values of the second evaluation area. The variation comparison step is a step of comparing the calculation result in the first variation calculation step with the calculation result in the second variation calculation step. The evaluation step is a step of evaluating the apparent state of the skin surface to be evaluated based on the comparison result in the variation comparison step.

The present inventors have found that whether a local area (uneven color area) having a size and color difference from the surrounding skin such as a stain or birthmark looks natural depends not only on the appearance of the uneven color area itself but also on the fact that no discontinuity in appearance occurs between the uneven color area and the surrounding skin. More specifically, the inventors have found that in order to realize a beautiful skin appearance, it is important that the textures caused by the minute unevenness caused by the minute irregularities of pores and skin grooves and the uneven distribution of pigments such as melanin pigments are similar to each other inside the uneven color area and outside the uneven color area. That is, according to the invention of Patent Document 1, although it is evaluated that the global color difference between the inside and outside (surrounding skin) of the uneven color area is reduced by applying a preparation such as foundation to the skin, it is not evaluated whether the minute irregularities inside the uneven color area are similar to the minute irregularities outside the uneven color area, in other words, the degree of the covering effect of the preparation on the fine texture is not evaluated.
Here, Fig. 1(a) to (c) are images close to normal observation conditions including surface reflected light and internal reflected light on the skin surface of the subject. Fig. 1(a) is a bare skin image of a part of the subject's cheek, Fig. 1(b) is a skin image of the same part with a certain preparation (preparation A described later), and Fig. 1(c) is a skin image of the same part with another preparation (preparation B described later). The bare skin image shown in Fig. 1(a) shows a large stain (uneven color area). In both the image of Fig. 1(b) where preparation A is applied and the image of Fig. 1(c) where preparation B is applied, there is no overall color difference between the surrounding skin, but in Fig. 1(b), the unevenness of the pores stands out as black spots, whereas in Fig. 1(c), the black spots are better covered by preparation B and are less noticeable. Thus, when formulation A is applied and when formulation B is applied, it appears that the spots (color unevenness) are covered in the same way at first glance, but when the texture of the skin is emphasized, there is a difference in the degree of covering the unevenness of the pores, and this difference affects the beautiful appearance of the skin. Furthermore, if there is a difference in the degree of covering the unevenness of the pores inside the color unevenness and the degree of covering the unevenness of the pores outside the color unevenness, that is, the uniformity between the inside and outside of the color unevenness, an unnatural impression is given, and as a result, the skin does not look beautiful. According to the present inventors, it has become clear that an evaluation that is well consistent with the impression of the beautiful appearance of the skin can be made by adopting and judging a new index, the degree of uniformity of the inside and outside of the color unevenness area, as an evaluation item. Therefore, in this method, the degree of uniformity of the cover is calculated by a variation comparison process that compares the variation of pixel values of the first evaluation area (e.g., inside the color unevenness area) with the variation of pixel values of the second evaluation area (e.g., outside the surrounding area adjacent to the color unevenness area). In the evaluation step, the apparent condition of the skin surface is evaluated based on the results of these calculations.

This method is explained in more detail below.

The skin image acquired in the skin image acquisition step is an image of at least a part of the subject's body including the evaluation target part of the subject's skin surface. The evaluation target part reflected in the skin image is not particularly limited. Any part of the skin surface that includes a predetermined unevenness can be evaluated, but a face image, particularly a face image that includes at least a part of the cheek, is a preferred example.
Here, "specific unevenness" is not limited to those caused by melanin pigments produced inside the skin being deposited on the skin, such as senile lentigo, melasma, freckles, symmetric dermal melanocytosis, and post-inflammatory pigmentation, but also includes unevenness associated with unevenness on the skin surface, such as pores and wrinkles.
The method of acquiring a skin image is not limited to a newly captured image of the skin, but also includes a method of acquiring an already captured image via a predetermined medium or a network.
Furthermore, the acquired skin image is not limited to an RGB image, but may be a monochrome (grayscale) image, a UV image, a near-infrared image, a spectroscopic image, etc. In addition, when acquiring an image, polarized light may be applied to one or both of the illumination and the photographing device. Furthermore, the acquired skin image is not limited to the image itself acquired in the skin image acquisition process, but may also include images that have been subjected to various image processing such as trimming, applying a frequency filter, and converting to grayscale.

Here, the environment in which the acquired skin image is photographed will be described. In actual situations, the unevenness can be viewed under various lighting conditions (e.g., under sunlight, under a specified lighting light, etc.) and from various directions (e.g., from the front of the unevenness, from an oblique angle, etc.). Accordingly, various evaluation purposes can be considered, such as when it is desired to evaluate the apparent state of the skin surface under a specific lighting condition (e.g., under sunlight), when it is desired to evaluate the apparent state of the skin surface when the unevenness is viewed from a specific direction, and when it is desired to evaluate the overall apparent state of the skin surface under multiple conditions. Therefore, it is necessary to use skin images photographed in an environment corresponding to the situation in which the unevenness is viewed, taking into consideration the evaluation purpose. Note that, if it is desired to evaluate the overall apparent state of the skin surface, it is sufficient to perform evaluation by this method for each skin image photographed in various shooting environments, and perform an overall evaluation based on the evaluation results.

The first evaluation region extracted in the extraction step may be a region that includes the predetermined unevenness, and may be a region that includes only the predetermined unevenness, or a region that includes the predetermined unevenness and a region that does not include the predetermined unevenness.
The second evaluation area extracted in the extraction process need only include at least an area other than the first evaluation area, and may not include the first evaluation area at all, or may include a part of the first evaluation area as the second evaluation area.
Furthermore, the acquired skin image does not necessarily have to be divided into a first evaluation area and a second evaluation area, but may be divided into a first evaluation area, a second evaluation area and other areas (areas not evaluated).
The "extraction" in the extraction process may refer to either automatically extracting each evaluation area using a predetermined algorithm, or an operator visually inspecting the skin image and specifying and extracting the evaluation area using an input device such as a mouse.

The "pixel value" for calculating the variation in the first variation calculation step and the second variation calculation step is the pixel intensity for each minute region constituting the image, and the minute region may be a pixel unit or a region with a set of a plurality of pixels as a unit. The pixel value here may be a calculated value correlated with the pixel intensity, for example, a luminance value. The pixel value to be used can be determined depending on the type of skin image used for evaluation.
In addition, as a method of calculating the variation in the first variation calculation step and the second variation calculation step, the standard deviation is calculated in this embodiment, but the variation may be calculated by calculating the variance. The variation calculated in the variation calculation step may be the standard deviation or variance, or may be skewness, kurtosis, or other index values correlated with these.

Typical examples of "prescribed preparations" include preparations that are applied to the skin, such as concealers, foundations, skin whitening agents, and UV protection agents, but are not limited to these, and may also include oral supplements such as various vitamins that improve the appearance of the skin surface. Therefore, applying a prescribed preparation means bringing about a change in the apparent condition of the skin surface by applying a preparation to the skin surface or taking a supplement.

In this way, this embodiment allows the apparent condition of the skin surface to be objectively evaluated using various indicators.

In this embodiment, a case will be described in which the covering effect of a specified formulation is evaluated based on the analysis results of a skin surface including a specified unevenness and a skin surface to which the specified formulation has been applied. Note that the skin condition evaluation method of this embodiment is not limited to evaluating the covering effect of a specified formulation, but can also evaluate, for example, the appearance of the specified unevenness on bare skin itself. This point will be described later.

<Skin condition evaluation method>
The skin condition evaluation method of this embodiment will be described below with reference to FIGS.
FIG. 2 is a flowchart showing the skin condition evaluation method of this embodiment (the present method).
FIG. 3 is a flowchart showing the analysis method of steps S110 and S130.
FIG. 4 is a graph created by the skin condition evaluation method according to the present embodiment.

The process (step S100) is a process of acquiring a skin image of a subject in a first state. Here, the first state refers to the state of bare skin itself including a predetermined uneven portion and its surrounding area (a state in which a predetermined preparation or the like has not been applied).
The skin image acquired in the process (step S100) is an image of bare skin including a predetermined uneven portion and its surrounding area captured by an imaging device (camera). The imaging device may be a general RGB camera, a monochrome camera, or a spectrum camera, and the performance and specifications of the imaging device are not limited. There are no limitations on the imaging environment conditions such as the number of lights, the lighting angle, the illuminance, and the shooting angle, and any angle of view suitable for evaluating the covering effect of a predetermined formulation may be used. In this embodiment, an RGB image is acquired.
The process (step S100) corresponds to the "skin image acquiring process" of this method for acquiring a skin image of the subject in the first state.

The process (step S110) analyzes the first state skin image acquired in the process (step S100). The analysis content will be explained using the flowchart in Figure 3.

In the step (step S111), a first evaluation area and a second evaluation area are extracted from the skin image in the first state acquired in the step (step S100).
In this embodiment, the first evaluation area designated by the operator is an area containing a certain unevenness, for example, a part of the skin image such as the stain shown in Fig. 1A. The second evaluation area is an area adjacent to the first evaluation area and surrounding the first evaluation area, and the skin image of the evaluation target area can be extracted by trimming the skin image of each area.
The process (step S111) corresponds to the "extraction process" of this method.

The skin image extracted in the process (step S111) is an RGB image, and the process (step S112) converts the skin image into a grayscale image.
In this embodiment, the captured skin image is converted from an RGB image to a grayscale image, but this is not limited to the above. Depending on the type and characteristics of a specified unevenness, the image may be converted to an image calculated from R, G, or B, or a combination of these, or may be converted to a color-converted image such as L*a*b.

In the step (step S113), the luminance value of the skin image in the first evaluation region in the grayscale image converted in the step (step S112) is measured, and the standard deviation of the luminance value is calculated. Note that the standard deviation of the luminance value is calculated by the reciprocal of the standard deviation. This indicates that the skin image in the first evaluation region is more uniform when the standard deviation is smaller (i.e., the variation is smaller), and the reciprocal of the standard deviation is calculated so that the evaluator can easily grasp the result when the result is displayed in a graph, as described later.
The process (step S113) corresponds to the "first variation calculation process" of this method.

In a step (step S114), the luminance values of the skin image in the second evaluation region in the grayscale image converted in the step (step S112) are measured, and a standard deviation of the luminance values is calculated.
The process (step S114) corresponds to the "second variation calculation process" of this method.

In the step (step S115), 1/standard deviation of the luminance value of the skin image of the first evaluation region calculated in the step (step S113) is compared with 1/standard deviation of the luminance value of the skin image of the second evaluation region calculated in the step (step S114). A skin image with less variation in luminance value, i.e., a skin image with a fine and highly uniform skin surface, tends to have a smaller variation in luminance value. As a result, the higher the uniformity, the larger the reciprocal of the standard deviation value, which is an index of variation, tends to be. And, compared to the first evaluation region (uneven color region), the second evaluation region, which is the region outside the first evaluation region, has a finer and more beautiful skin and a smaller variation in luminance value. Therefore, when the ratio of the reciprocal of the standard deviation of the luminance value of the first evaluation region (uneven color region) to the reciprocal of the standard deviation of the luminance value of the second evaluation region (surrounding region) approaches 1, it means that the textures of the first evaluation region (uneven color region) and the second evaluation region (surrounding region) are similar, and unnatural discontinuity in appearance is unlikely to occur at the boundary between these regions. For this reason, in the variation comparison step of this method, the reciprocal of the standard deviation of the luminance values is compared between the first evaluation area and the second evaluation area, and whether or not the first evaluation area (color uneven area) looks natural is evaluated. In this embodiment, in order to grasp the degree of agreement (similarity) between the uniformity of 1/standard deviation of the luminance values of the skin image of the first evaluation area and the uniformity of 1/standard deviation of the luminance values of the skin image of the second evaluation area, the ratio of these two values (the reciprocal of the standard deviation of the luminance values) is calculated, but this is not limited to this, and the difference between these two values may be used, or the ratio or difference between the standard deviations of the luminance values may be calculated instead of the reciprocal. Any method is possible as long as the degree of agreement (similarity) between these two values can be grasped.
It should be noted that this step (step S115) corresponds to a part of the "variation comparison step" of this method.

2, the process (step S120) is a process of acquiring a skin image of the subject in a second state, where the second state refers to a state in which a predetermined preparation is applied to the predetermined uneven portion and its surrounding area.
The skin image acquired in the step (step S120) is a skin image of approximately the same range as the skin image acquired in the step (step S100). Therefore, it is an image of skin with a predetermined preparation applied to a predetermined uneven portion and its surrounding area photographed by an imaging device (camera). In this embodiment, in order to evaluate the covering effect of the predetermined preparation applied to the predetermined uneven portion and its surrounding area, it is preferable to use an image photographed under the same photographing environmental conditions as when the skin image in the first state was acquired. In this embodiment, an RGB image is acquired.
The process (step S120) corresponds to the "skin image acquiring process" of the present method for acquiring the skin image of the subject in the second state.

In step (step S130), the skin image in the second state acquired in step (step S120) is analyzed. The analysis content will be explained using the flowchart in FIG. 3. Note that the analysis content of the skin image in the second state is the same as the analysis content of the skin image in the first state, so only the parts unique to the analysis of the skin image in the second state will be explained.

In the step (step S111), a first evaluation area and a second evaluation area are extracted from the skin image in the second state acquired in the step (step S120).
In this embodiment, the same areas as in the first state are extracted as the first evaluation area and the second evaluation area.
The analysis content of the skin image in the second state (step (step S112) to step (step S115)) is the same as the analysis content of the skin image in the first state, so the description will be omitted.

Returning to Figure 2, the process (step S140) graphically displays the degree of agreement (similarity) of uniformity between the first evaluation area and the second evaluation area calculated in the process (step S110) by comparing the 1/standard deviation of the luminance values of the skin image of the first evaluation area in the first state with the 1/standard deviation of the luminance values of the skin image of the second evaluation area, and the degree of agreement (similarity) of uniformity between the first evaluation area and the second evaluation area calculated in the process (step S130) by comparing the 1/standard deviation of the luminance values of the skin image of the first evaluation area in the second state with the 1/standard deviation of the luminance values of the skin image of the second evaluation area.
Here, the condition of the skin surface (first state (bare skin) and second state (state after a specified formulation has been applied)) is shown on the x-axis, and the degree of agreement (similarity) of uniformity between the first evaluation area and the second evaluation area in the first state and the degree of agreement (similarity) of uniformity between the first evaluation area and the second evaluation area in the second state are shown on the y-axis.
We then present the evaluation results of the cleanliness of the cover, which can be understood from the graph.
The process (step S140) corresponds to the "evaluation process" of this method.

FIG. 4 is a graph created by the skin condition evaluation method according to the present embodiment.
In Fig. 4, three types of patterns are evaluated: bare skin as the first state, and a case in which formulation A is applied and a case in which formulation B is applied as the second state, and in addition to evaluating the cleanliness of the individual covers of formulation A and formulation B, a relative evaluation is also performed. Note that by increasing the number of types of formulations applied and the number of evaluation patterns, it is possible to perform a further relative evaluation.
The x-axis in Fig. 4 represents the condition of the skin surface. "Bare skin" represents the state where nothing is applied to the first evaluation area and the second evaluation area. "Preparation A" represents the state where preparation A is applied to the first evaluation area and the second evaluation area. "Preparation B" represents the state where preparation B, which is different from preparation A, is applied to the first evaluation area and the second evaluation area.
4 is a value calculated by comparing 1/standard deviation of the luminance value of the skin image in the first evaluation area with 1/standard deviation of the luminance value of the skin image in the second evaluation area, and the closer this value is to 1, the more consistent (similar) the uniformity of the first evaluation area is with the uniformity of the second evaluation area. In other words, it indicates that the area is naturally covered.
As is clear from Figure 4, when formulation B is applied, the uniformity of the first evaluation area and the uniformity of the second evaluation area are more consistent than when formulation A is applied, and it can be seen that the specified unevenness is covered more naturally.
In this way, by using the skin condition evaluation method according to the present embodiment for evaluation, it becomes possible to evaluate whether natural coverage is achieved by using the degree of agreement (degree of similarity) between the uniformity of the uneven part and the uniformity of the surrounding part as an evaluation index. In addition, the evaluation result using the skin condition evaluation method according to the present embodiment can also be used as an index for selecting a formulation suitable for the unevenness.
In addition, using the degree of agreement (similarity) between the uniformity of an uneven area and the uniformity of its surrounding area as an evaluation index is an important index for measuring the effect of a preparation such as a low-concealing (low-coverage) makeup cosmetic agent intended to correct the texture of the skin. In particular, uneven areas caused by irregularities such as pores and wrinkles have little color difference with the surrounding areas, but look unclean due to the lack of uniformity, so whether the uniformity between the uneven area and its surrounding areas has been improved by applying the preparation can be expected to be an important index for evaluating the natural covering effect.

In the above-described embodiment, a case has been described in which the covering effect of a specified formulation is evaluated based on the analysis results of a skin surface including a specified unevenness and a skin surface to which the specified formulation has been applied. Below, a case in which the appearance of the specified unevenness on bare skin itself is evaluated will be described.

When evaluating the appearance of a specified unevenness on bare skin itself, since there is no first state and second state (since there is only one state), steps (step S100), (step S110), and (step 140) of the flowchart showing the skin condition evaluation method shown in Figure 2 are carried out.
Furthermore, the graph presented in the process (step S140) will be only the bare skin bars of the graph shown in FIG.
The evaluator is then able to assess how unclean the particular irregularity evaluated is, and how inconsistent it is with the external irregularity.
Here, when there is only one subject, there is little effect in displaying the graph, but when there are multiple subjects, for example, displaying the graph can be used for relative evaluation.
Furthermore, for example, the first state is an evaluation of a first evaluation area containing a specific unevenness (e.g., melasma) and an evaluation of a second evaluation area for a first subject, and the second state is an evaluation of the first evaluation area and the second evaluation area containing the same specific unevenness as the first subject for a second subject, making it possible to evaluate whether the same type of unevenness possessed by different subjects is not beautiful or which unevenness is not beautiful.
In addition, by evaluating the first state for the same subject by evaluating a first evaluation area including a first unevenness and a second evaluation area, and by evaluating the second state for the first evaluation area including a second unevenness different from the first unevenness and an evaluation of the second evaluation area, it becomes possible to evaluate whether the different unevennesses possessed by the same subject are neat or which unevenness is neater.

In this way, the skin condition evaluation method according to this embodiment can also be used to evaluate the appearance of unevenness on bare skin itself, and can be used to give advice when applying makeup, such as what types of unevenness should be particularly covered, or when selecting a product.

In the above-described embodiment, when calculating the variation of each evaluation area in the first variation calculation process and the second variation calculation process, the pixel values of the evaluation area are used. However, it is also possible to extract "pixel values of specific conditions" in the first variation calculation process and the second variation calculation process, and calculate the variation of the extracted unevenness component pixel values.
Here, the "pixel value under a specific condition" extracted in the first variation calculation process and the second variation calculation process is a pixel value extracted using a resolution filter capable of extracting a specified unevenness, and is extracted using a resolution filter corresponding to the specified unevenness.
In this way, by extracting pixel values consisting of frequency components corresponding to a specified unevenness and calculating the variation, it is possible to evaluate the degree of correspondence between the uniformity of a specified unevenness element and the uniformity of its surrounding areas, thereby making it possible to evaluate the cleanliness of the first evaluation area and the second evaluation area as a whole.
In addition, the unevenness component pixel values consisting of frequency components corresponding to a specified unevenness may be divided by a specific threshold value and extracted using frequency characteristics divided into two stages, such as high frequency and low frequency, or multi-resolution analysis may be applied to divide the data into several stages, and an image of a frequency band matching the specified unevenness may be extracted from among them.

Here, the "pixel values under specific conditions" extracted in the first variation calculation step and the second variation calculation step will be further described.
The frequency ranges to be extracted in the first variation calculation step and the second variation calculation step are determined in consideration of the following points.
In the case of a frequency lower than a predetermined low frequency, the global color change of the predetermined unevenness is overlooked by a human, and in the case of a predetermined high frequency, the predetermined unevenness is too fine to be visually recognized by a human. The appropriate frequency range differs depending on the position (distance) from which a person views the skin, the angle from which the person views the skin, the lighting under which the person views the skin, and so on. Therefore, in this embodiment, a frequency filter that is suitable for the size of the unevenness to be analyzed contained in the skin image captured under a shooting environment determined in consideration of the evaluation purpose is used.
Here, when generally observing the skin while facing the face and talking, that is, when observing the skin from a distance of 1 to 2 meters, a frequency filter capable of extracting color unevenness of 0.1 millimeters or more is determined based on the shooting environment. The reason why the size of the color unevenness itself is set to 0.1 millimeters or more is that it is difficult to recognize unevenness smaller than 0.1 millimeters when observing the skin from a distance of 1 to 2 meters. In addition, the upper limit of the size of the color unevenness itself to be extracted may be specified depending on the type of unevenness contained in the acquired skin image. In this way, evaluation can be performed only on the frequency components of the unevenness to be evaluated. However, since the size of the unevenness itself varies depending on the cause of the unevenness (whether it is senile pigmentation or wrinkles, etc.), it is preferable to evaluate it without specifying the upper limit.
On the other hand, when the skin is closely observed with the naked eye, a frequency filter capable of extracting color unevenness of 20 microns to 1 centimeter in size is determined based on the shooting environment. The size of the color unevenness itself is set to 20 microns or more because it is difficult to recognize color unevenness smaller than 20 microns. Also, the size of the color unevenness itself is set to 1 centimeter or less because in the case of large unevenness, the color (brightness) within the unevenness is also uneven, and the size of the color unevenness within this unevenness is about 1 centimeter.

FIG. 5 shows a graph created by extracting unevenness component pixel values consisting of frequency components corresponding to the specified unevenness when calculating the variation in the first evaluation region including the specified unevenness in the first variation calculation process and the second variation calculation process.
The x-axis in Fig. 5 represents the state of the skin surface, as in Fig. 4. "Bare skin" represents the state in which nothing is applied to the first and second evaluation areas. "Preparation A" represents the state in which preparation A is applied to the first and second evaluation areas. "Preparation B" represents the state in which preparation B, which is different from preparation A, is applied to the first and second evaluation areas.
5 is a value calculated by extracting unevenness component pixels consisting of frequency components corresponding to a predetermined unevenness, calculating the standard deviation of the luminance values of the extracted images, and comparing the values calculated as 1/standard deviation when calculating 1/standard deviation of the luminance values of the skin image in the first evaluation region and the skin image in the second evaluation region, and the closer this value is to 1, the more consistent (similar) the uniformity of the first evaluation region and the uniformity of the second evaluation region are. In other words, it indicates that the area is naturally covered.
As is clear from Figure 5, the uniformity of the first evaluation area and the uniformity of the second evaluation area are more consistent (similar) when formulation B is applied than when formulation A is applied, and it is understood that the specified unevenness can be naturally covered. Furthermore, compared to Figure 4, it can be seen that both formulations A and B can uniformly cover the extracted pixel values.

Here, each step of the method may be performed by a person operating a computer. For example, in step (step S100) and step (step S120), a person may operate a camera to capture a skin image of a subject, and operate a computer to input the skin image into the computer. Step (step S110), step (step S130), and step (step S140) can be performed by a person operating commercially available software.
Furthermore, each step of the present method may be automatically executed by a computer (for example, a skin condition evaluation system 100 described below).

<Skin condition evaluation system 100>
A skin condition evaluation system 100 for implementing this method will be described below.
FIG. 6 is a block diagram of a skin condition evaluation system 100 according to this embodiment.
The skin condition evaluation system 100 in this embodiment is an information processing terminal capable of executing various processes. Although not shown in the figure, the skin condition evaluation system 100 is equipped with input devices such as a keyboard and a pointing device, a processing unit, a storage unit, etc.
Skin condition evaluation system 100 includes skin image acquiring means 110, extracting means 120, first variation calculating means 130, second variation calculating means 140, variation comparing means 150, and evaluating means 160. In this embodiment, input means 170, display control means 180, and display means 190 are provided separately from skin condition evaluation system 100 as means for transmitting and receiving control signals to and from skin condition evaluation system 100. Note that input means 170, display control means 180, and display means 190 may be provided in the skin condition evaluation system.

The skin image acquisition means 110 acquires a skin image of the skin surface. Specifically, the skin image acquisition means 110 acquires image data of the subject's skin image captured by an imaging device (camera) from the imaging device, a portable recording medium, or another computer via communication. The skin image acquisition means 110 may also be a camera built into the skin condition evaluation system 100.

The extraction means 120 extracts a first evaluation area and a second evaluation area from the skin image data acquired by the skin image acquisition means 110. Here, the extraction performed by the extraction means 120 may be either a method of automatically extracting each evaluation area using a predetermined algorithm or a method in which an operator visually checks the skin image and specifies and extracts the evaluation area using an input device such as a mouse.
The extraction means 120 extracts a skin image of the evaluation target area by trimming a first evaluation area including a predetermined unevenness and a second evaluation area surrounding the first evaluation area. In this embodiment, the extracted skin image is an RGB image, and is converted to a grayscale image by the extraction means 120.

The first variation calculation means 130 calculates the reciprocal of the standard deviation of the luminance values in the first evaluation region.
The second variation calculation means 140 calculates the reciprocal of the standard deviation of the luminance values in the second evaluation region.
The variation comparison means 150 calculates the degree of consistency (similarity) of the uniformity of the cover from 1/standard deviation of the luminance value of the first evaluation area calculated by the first variation calculation means 130 and 1/standard deviation of the luminance value of the second evaluation area calculated by the second variation calculation means 140.

The evaluation means 160 displays the degree of agreement (similarity) of the cover uniformity calculated by the variation comparison calculation means 150 in a graph.
Here, the condition of the skin surface is represented on the x-axis. "Bare skin" is the state where nothing is applied to the first evaluation area and the second evaluation area. "Preparation A" is the state where preparation A is applied to the first evaluation area and the second evaluation area. "Preparation B" is the state where preparation B, which is different from preparation A, is applied to the first evaluation area and the second evaluation area.
The y-axis represents the degree of agreement of the uniformity of the coverage calculated by comparing 1/standard deviation of the brightness value of the skin image in the first evaluation area with 1/standard deviation of the brightness value of the skin image in the second evaluation area. The closer this value is to 1, the more the uniformity of the first evaluation area and the uniformity of the second evaluation area match (are similar). In other words, it indicates that the coverage is natural.
In this way, by displaying the evaluation results of the skin condition evaluation method according to this embodiment in a graph, it is possible to present the evaluation results to the operator in an easily understandable manner.

The input means 170 is an input device such as a keyboard or a pointing device. In this embodiment, the input means 170 is a means used by an operator to visually inspect the skin image and specify the evaluation area when the extraction means 120 extracts the first evaluation area and the second evaluation area from the skin image data acquired by the skin image acquisition means 110. By providing a means for an operator to specify the evaluation area in this way, for example, a beauty staff member at a store or the like can take an image of the skin of a subject (customer) and specify the unevenness desired by the customer from the captured data, making it easy to evaluate the apparent condition of the skin surface.

The display control means 180 is a control means for displaying the evaluation contents including the graphs presented by the evaluation means 160 on the display means 180.

The display means 190 is a device that displays the evaluation contents including the graphs presented by the evaluation means 160 under the control of the display control means 180 .
The display device may be a display device such as a monitor, or a printing device such as a printer.

The results of the evaluation using the skin condition evaluation system of this embodiment are the same as those shown in Figure 4 above.

<Other skin condition evaluation methods>
The above-mentioned skin condition evaluation method uses the degree of agreement (similarity) of the uniformity of the cover calculated by comparing 1/standard deviation of the luminance value of the skin image in the first evaluation area with 1/standard deviation of the luminance value of the skin image in the second evaluation area as an index for evaluating the apparent condition of the skin condition.
However, some formulations are required to cover unevenness beautifully in addition to naturally covering the unevenness.
Therefore, a method of evaluating skin condition will be described in which a quantitative element, namely, the value of the color difference, is added to the index in addition to the uniformity between the uneven portion and its surrounding portion.

Here, the calculated global color difference between the uneven area and its surrounding area is used as an index to evaluate the apparent condition of the skin surface together with an index showing the degree of uniformity between the uneven area and its surrounding area.
2 (step S110), in addition to calculating the reciprocal of the standard deviation of the luminance values of the first evaluation area in the first state and the reciprocal of the standard deviation of the luminance values of the second evaluation area, the average value of the luminance values of the first evaluation area in the first state and the average value of the luminance values of the second evaluation area are also calculated. Then, the contrast between the calculated average value of the luminance values of the first evaluation area and the average value of the luminance values of the second evaluation area is calculated.
2 (step S130), in addition to calculating the reciprocal of the standard deviation of the luminance values of the first evaluation area in the second state and the reciprocal of the standard deviation of the luminance values of the second evaluation area, the average value of the luminance values of the first evaluation area in the second state and the average value of the luminance values of the second evaluation area are calculated. Then, the contrast between the calculated average value of the luminance values of the first evaluation area and the average value of the luminance values of the second evaluation area is calculated.
In this embodiment, in order to grasp the state difference between the average brightness value of the skin image in the first evaluation area and the average brightness value of the skin image in the second evaluation area, the contrast between these two values is calculated, but this is not limited to this, and the difference between these two values may also be used, and any method is acceptable as long as it is possible to grasp the state difference between these two values.
Note that calculating the contrast between the average luminance value of the first evaluation region and the average luminance value of the second evaluation region corresponds to the "comparison calculation step" of this method.

Next, the process (step S140) shown in FIG. 2 graphically displays the degree of agreement (similarity) of uniformity between the first evaluation area and the second evaluation area calculated in the process (step S110) by comparing the 1/standard deviation of the luminance values of the skin image of the first evaluation area in the first state with the 1/standard deviation of the luminance values of the skin image of the second evaluation area, the contrast between the average value of the luminance values of the skin image of the first evaluation area in the first state and the average value of the luminance values of the skin image of the second evaluation area, the degree of agreement (similarity) of uniformity between the first evaluation area and the second evaluation area calculated in the process (step S130) by comparing the 1/standard deviation of the luminance values of the skin image of the first evaluation area in the second state with the 1/standard deviation of the luminance values of the skin image of the second evaluation area, and the contrast between the average value of the luminance values of the skin image of the first evaluation area in the second state and the average value of the luminance values of the skin image of the second evaluation area.
Here, the contrast between the average brightness value of the skin image of the first evaluation area in the first state and the average brightness value of the skin image of the second evaluation area is "x_0", the comparison result between 1/standard deviation of the brightness value of the skin image of the first evaluation area in the first state and 1/standard deviation of the brightness value of the skin image of the second evaluation area in the first state is "y_0", the contrast between the average brightness value of the skin image of the first evaluation area in the second state and the average brightness value of the skin image of the second evaluation area is "x_1", and the comparison result between 1/standard deviation of the brightness value of the skin image of the first evaluation area in the second state and 1/standard deviation of the brightness value of the skin image of the second evaluation area in the second state is "y_1".
In addition, the first state (x_0, y_0) and the second state (x_1, y_1) are graphed, with the x-axis representing the contrast value between the average brightness value of the skin image in the first evaluation area and the average brightness value of the skin image in the second evaluation area, and the y-axis representing the comparison value between 1/standard deviation of the brightness value of the skin image in the first evaluation area and 1/standard deviation of the brightness value of the skin image in the second evaluation area.

FIG. 7 is a graph created by the skin condition evaluation method according to the present embodiment.
7, two types of patterns are evaluated, one in which bare skin is used as the first state and formulation A is applied as the second state, and the other in which bare skin is used as the first state and formulation B is applied as the second state, to perform a relative evaluation in addition to the evaluation of the individual covering effects of formulations A and B. Note that by increasing the number of types of formulations applied and the number of evaluation patterns, it becomes possible to perform a further relative evaluation.
The value on the x-axis in Fig. 7 is the contrast between the average brightness value of the skin image in the first evaluation region and the average brightness value of the skin image in the second evaluation region (brightness value inside the unevenness/brightness value around the unevenness). Here, the smaller the difference between the average brightness value of the skin image in the first evaluation region and the average brightness value of the skin image in the second evaluation region, in other words, the closer the contrast is to 1, the less overall color difference there is between the skin image in the first evaluation region and the skin image in the second evaluation region, and therefore the more cleanly the specified unevenness is covered. Therefore, the larger the value on the x-axis, the higher the covering power.
7 is the degree of agreement (similarity) of uniformity calculated by comparing 1/standard deviation of the luminance value of the skin image in the first evaluation area with 1/standard deviation of the luminance value of the skin image in the second evaluation area, and the closer this value is to 1, the more the uniformity in the first evaluation area and the uniformity in the second evaluation area match (are similar).
Furthermore, when there is no overall color difference between the first evaluation area and the second evaluation area and the first evaluation area and the second evaluation area are consistent in uniformity, it can be said that the cover is clean and natural.
As is clear from Figure 7, when formulation A is applied, the covering power to cover the specified unevenness is high, but the uniformity of the first evaluation area including the specified unevenness and the uniformity of the second evaluation area do not match very well, resulting in unnatural covering. When formulation B is applied, in addition to the covering power to cover the specified unevenness is high, the uniformity of the first evaluation area including the specified unevenness and the uniformity of the second evaluation area match at values close to 1, resulting in natural covering.
In this way, by using the skin condition evaluation method according to the present embodiment for evaluation, in addition to the color difference between the uneven part and its surrounding area as in the past, the degree of uniformity between the uneven part and its surrounding area can also be used as an evaluation index, and as a result, it becomes possible to evaluate whether the unevenness is covered cleanly and naturally. In addition, the evaluation result using the skin condition evaluation method according to the present embodiment can also be used as an index for selecting a formulation suitable for the unevenness. Furthermore, when selecting a formulation, whether to emphasize clean coverage of the unevenness, natural coverage, or clean and natural coverage may differ depending on the subject (customer). In that case, by using this method, it is also possible to propose a formulation with a high covering effect that the subject emphasizes based on the x-axis value and the y-axis value.

8 is a graph created by extracting unevenness component pixel values consisting of frequency components corresponding to the predetermined unevenness when calculating the variation of the first evaluation region including the predetermined unevenness in the first variation calculation step and the second variation calculation step. Note that the method of extracting the unevenness component pixel values consisting of frequency components corresponding to the predetermined unevenness is as described above.
The value on the x-axis in Fig. 8 is the contrast (brightness value inside unevenness/brightness value around the outside of unevenness) between the average brightness value of the skin image in the first evaluation region and the average brightness value of the skin image in the second evaluation region, as in Fig. 7. Here, the smaller the difference between the average brightness value of the skin image in the first evaluation region and the average brightness value of the skin image in the second evaluation region, in other words, the closer the contrast is to 1, the less overall color difference there is between the skin image in the first evaluation region and the skin image in the second evaluation region, and therefore the more cleanly the specified unevenness is covered. Therefore, the larger the value on the x-axis, the higher the covering power.
In addition, the value on the y-axis in Figure 8 is a value calculated when calculating 1/standard deviation of the luminance values of the skin image in the first evaluation area and 1/standard deviation of the luminance values of the skin image in the second evaluation area as described above, by extracting unevenness component pixels consisting of frequency components corresponding to a predetermined unevenness, then determining the standard deviation of the luminance values of the images consisting of the extracted images, and comparing the values calculated as 1/standard deviation.The closer this value is to 1, the more consistent (similar) the uniformity of the first evaluation area and the uniformity of the second evaluation area are.
As is clear from Figure 8, when formulation A is applied, the covering power to cover the predetermined unevenness is high, but the uniformity of the first evaluation area including the predetermined unevenness and the uniformity of the second evaluation area do not match very well, resulting in unnatural covering. In addition, when formulation B is applied, in addition to the covering power to cover the predetermined unevenness, the uniformity of the first evaluation area including the predetermined unevenness and the uniformity of the second evaluation area match at a value close to 1, so it can be seen that natural covering is also performed. Furthermore, it can be seen that both formulation A and formulation B can provide natural covering for the extracted pixel values, although only slightly, compared to Figure 7.

The present invention is not limited to the above-described embodiment, and various modifications and improvements are possible.

In the above embodiment, the first evaluation area is described as being entirely reflected in the skin image. However, even if the first evaluation area is not entirely reflected in the skin image, for example, the first evaluation area is the left half of the skin image and the second evaluation area is the right half of the skin image, it is possible to evaluate the apparent condition of the skin surface in a similar manner. In this way, since the first evaluation area does not necessarily have to be entirely reflected in the acquired skin image, it is also possible to evaluate unevenness near the hairline, for example, and it becomes possible to evaluate the apparent condition of the skin surface without being limited by the location of the unevenness.

In the above embodiment, the "predetermined unevenness" is described as an example of a "lump of blemish" and the "lump of blemish" is described as the first evaluation region. However, if the "predetermined unevenness" is a collection of small color unevenness such as freckles, the collection can be used as the first evaluation region to evaluate the apparent state of the skin surface. In this way, it becomes possible to evaluate the apparent state of the skin surface without being limited by the size of the color unevenness or the characteristics of the color unevenness (color unevenness in which small color unevennesses are gathered together and recognized as a single unevenness).

In the above embodiment, if the position of the first evaluation area in the skin image in the first state specified in the step (step S111) in the step (step S110) is misaligned with the position of the first evaluation area in the skin image in the second state specified in the step (step S111) in the step (step S130), an accurate evaluation result may not be obtained. Therefore, in order to prevent the position of the first evaluation area in the skin image in the first state from being misaligned with the position of the first evaluation area in the skin image in the second state, it is preferable to align and size the two skin images using facial elements such as hair and eyes, so that the position where the first evaluation area is specified in the skin image in the first state can automatically specify the first evaluation area in the skin image in the second state. Note that, when aligning the two skin images, it is desirable that prominent facial elements such as eyes, nose, mouth, hair, and nasolabial folds are reflected in the two skin images.

In the above embodiment, the first evaluation area and the second evaluation area are mutually exclusive areas in the skin image. However, the entire skin image may be the second evaluation area, and the first evaluation area may be included in the second evaluation area. The skin image may also include an area that does not belong to either the first evaluation area or the second evaluation area (an area that is not subject to evaluation). In this way, as long as the condition that the first evaluation area includes a predetermined unevenness is satisfied, the first evaluation area and the second evaluation area may be extracted in any way from the skin image, so that it is possible to reduce restrictions when capturing the skin image, and also to reduce restrictions on the position where the predetermined unevenness is present. Therefore, it is possible to evaluate the apparent state of the skin surface for the predetermined unevenness at various positions.

The skin condition evaluation method of the present invention is not limited to displaying the evaluation results in a graph as described above, or displaying the calculated values as they are. For example, when multiple types of formulations are applied to the skin and the appearance of the skin surface is evaluated, the type of formulation that provides the best coverage may be recommended without outputting the graph display or calculated values. In this way, even general users who have difficulty understanding the meaning of the calculated values or displayed graphs can effectively use the evaluation results. Note that both the graph display and the recommendation of the type of formulation may be performed, and it may be possible to select what information to display.

The above-described skin condition evaluation system uses the degree of agreement (similarity) of the uniformity of the cover calculated by comparing 1/standard deviation of the luminance value of the skin image in the first evaluation area with 1/standard deviation of the luminance value of the skin image in the second evaluation area. In addition, the average value of the luminance value of the first evaluation area and the average value of the luminance value of the second evaluation area in the first state may be calculated, and the contrast between the calculated average value of the luminance value of the first evaluation area and the average value of the luminance value of the second evaluation area may be calculated, and the overall color difference between the uneven part and its surrounding area may be added to the index. In this way, by using the degree of agreement of the uniformity between the uneven part and its surrounding area as an evaluation index in addition to the color difference between the uneven part and its surrounding area, it is possible to evaluate whether the unevenness is covered cleanly and naturally.
In this case, a "comparison calculation means" may be added to the configuration of the skin condition evaluation system shown in Fig. 6. The comparison calculation means may calculate the average value of the luminance values of the first evaluation region and the average value of the luminance values of the second evaluation region in the first state, and calculate the contrast between the calculated average value of the luminance values of the first evaluation region and the average value of the luminance values of the second evaluation region. Then, the evaluation means 160 may evaluate the contrast between the calculated average value of the luminance values of the first evaluation region and the average value of the luminance values of the second evaluation region by adding it to the evaluation index. Furthermore, the evaluation means 160 graphically displays the degree of agreement (similarity) of uniformity between the first evaluation area and the second evaluation area calculated by comparing 1/standard deviation of the luminance values of the skin image of the first evaluation area in the first state with 1/standard deviation of the luminance values of the skin image of the second evaluation area, the contrast between the average value of the luminance values of the skin image of the first evaluation area in the first state and the average value of the luminance values of the skin image of the second evaluation area, the degree of agreement (similarity) of uniformity between the first evaluation area and the second evaluation area calculated by comparing 1/standard deviation of the luminance values of the skin image of the first evaluation area in the second state with 1/standard deviation of the luminance values of the skin image of the second evaluation area, and the contrast between the average value of the luminance values of the skin image of the first evaluation area in the second state and the average value of the luminance values of the skin image of the second evaluation area. Specifically, the contrast between the average brightness value of the skin image of the first evaluation area in the first state and the average brightness value of the skin image of the second evaluation area is "x_0", the comparison result between 1/standard deviation of the brightness value of the skin image of the first evaluation area in the first state and 1/standard deviation of the brightness value of the skin image of the second evaluation area in the first state is "y_0", the contrast between the average brightness value of the skin image of the first evaluation area in the second state and the average brightness value of the skin image of the second evaluation area is "x_1", and the comparison result between 1/standard deviation of the brightness value of the skin image of the first evaluation area in the second state and 1/standard deviation of the brightness value of the skin image of the second evaluation area in the second state is "y_1".
In addition, the first state (x_0, y_0) and the second state (x_1, y_1) are graphed, with the x-axis representing the contrast value between the average brightness value of the skin image in the first evaluation area and the average brightness value of the skin image in the second evaluation area, and the y-axis representing the comparison value between 1/standard deviation of the brightness value of the skin image in the first evaluation area and 1/standard deviation of the brightness value of the skin image in the second evaluation area.
The results of such evaluation can be presented as a graph similar to that of FIG.

REFERENCE SIGNS LIST 100 Skin condition evaluation system 110 Skin image acquisition means 120 Extraction means 130 First variation calculation means 140 Second variation calculation means 150 Variation comparison means 160 Evaluation means 170 Input means 180 Display control means 190 Display means

Claims (8)

a skin image acquiring step of acquiring a skin image of a skin surface including a predetermined unevenness;
an extraction step of extracting from the skin image a first evaluation area including the predetermined unevenness and a second evaluation area which is a partial area of the skin image and includes at least a part or all of an area other than the first evaluation area and is adjacent to the first evaluation area;
converting the skin image to a grayscale image;
a first variation calculation step of calculating a variation in pixel values of the first evaluation area in the grayscale image ;
a second variation calculation step of calculating a variation in pixel values of the second evaluation area in the grayscale image ;
a variation comparison step of comparing a calculation result in the first variation calculation step with a calculation result in the second variation calculation step using division;
an evaluation step of evaluating the appearance of the skin surface based on the calculation result in the variation comparison step;
A method for evaluating a skin condition comprising: The first variation calculation step includes a step of extracting pixel values of a specific condition in the first evaluation area and calculating a variation of the pixel values of the extracted specific condition,
The skin condition evaluation method according to claim 1 , wherein the second variation calculation step includes a step of extracting pixel values of specific conditions in the second evaluation area and calculating the variation of the pixel values of the extracted specific conditions. The first variation calculation step and the second variation calculation step include:
3. The skin condition evaluation method according to claim 2, further comprising the step of extracting, as the specific condition, an unevenness component pixel value having a frequency component corresponding to the predetermined unevenness. providing a first state in which a predetermined formulation is not applied to the skin surface corresponding to the skin image of the first evaluation area, and a second state in which the predetermined formulation is applied to the skin surface corresponding to the skin image of the first evaluation area, and performing the skin image acquiring step, the extracting step, the first variation calculating step, the second variation calculating step, and the variation comparing step;
The skin condition evaluation method according to any one of claims 1 to 3, characterized in that in the evaluation step, a change in the apparent condition of the skin surface due to application of the specified formulation is evaluated based on the calculation result in the variation comparison step in the first state and the calculation result in the variation comparison step in the second state. a comparison and calculation step of comparing a pixel value of the first evaluation area with a pixel value of the second evaluation area,
5. The skin condition evaluation method according to claim 1, wherein in the evaluation step, the apparent condition of the skin surface is evaluated based on the calculation result in the variation comparison step and the comparison result in the comparison calculation step. A skin image acquisition means for acquiring a skin image of a skin surface including a predetermined unevenness;
an extraction means for extracting from the skin image a first evaluation area including the predetermined unevenness, and a second evaluation area which is a partial area of the skin image and includes at least a part or all of an area other than the first evaluation area and is adjacent to the first evaluation area;
means for converting the skin image into a grayscale image;
a first variation calculation means for calculating a variation in pixel values of the first evaluation area in the grayscale image ;
a second variation calculation means for calculating a variation in pixel values of the second evaluation area in the grayscale image ;
a variation comparison means for comparing a calculation result by the first variation calculation means with a calculation result by the second variation calculation means using division;
an evaluation means for evaluating the appearance of the skin surface based on the calculation result by the variation comparison means;
A skin condition assessment system including: An input means operable by an operator;
A display means for displaying the skin image;
A display control unit that controls the display unit,
The operator is allowed to use the input means to specify the first evaluation area and the second evaluation area for the skin image displayed on the display means;
The skin condition evaluation system according to claim 6 , wherein the extraction means extracts the first evaluation area and the second evaluation area designated by the operator. A comparison calculation means for comparing a pixel value of the first evaluation area with a pixel value of the second evaluation area is further provided,
performing processing by the skin image acquisition means, the extraction means, the first variation calculation means, the second variation calculation means, the variation comparison means, and the comparison calculation means for a first state in which a predetermined formulation is not applied to the skin surface corresponding to the skin image of the first evaluation area and a second state in which the predetermined formulation is applied to the skin surface corresponding to the skin image of the first evaluation area;
the evaluation means has a function of generating graph data in which the calculation results calculated by the comparison calculation means and the calculation results calculated by the variation comparison means for the skin image in the first state, and the calculation results calculated by the comparison calculation means and the calculation results calculated by the variation comparison means for the skin image in the second state are plotted in a coordinate system including a first axis indicating the calculation results calculated by the comparison calculation means and a second axis indicating the calculation results calculated by the variation comparison means,
The skin condition evaluation system according to claim 7 , wherein the display control means controls the display of the graph data generated by the evaluation means on the display means.

Images