Skin color unevenness display method, skin color unevenness evaluation method, skin color unevenness display device, and skin color unevenness display program

Abstract

To provide a novel technique for expressing and evaluating color unevenness of skin.SOLUTION: A skin color unevenness display method comprises: a calculation step for calculating a color difference between neighboring pixels for each pixel with respect to a skin image of a subject; and a display step for displaying appearance frequency for a calculated value of the color difference in a graph or chart.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a method for displaying skin color unevenness, a method for evaluating skin color unevenness, a device for displaying skin color unevenness, and a program for displaying skin color unevenness.

Background Art

[0002] In recent years, by performing image analysis on an image including skin, evaluation of skin condition, evaluation of the apparent age of a face, etc. have been carried out. Patent Document 1 describes a method of calculating two or more characteristic indexes such as wrinkles from a skin image and determining a sensory evaluation of the skin based on the two or more characteristic amounts.

[0003] Patent Document 2 describes a method for evaluating the skin condition of capturing a skin image of a subject as a digital image, calculating a color difference from the color information between a unit area of this image and an adjacent area, and further obtaining a frequency distribution of the color difference.

[0004] Also, as a method for evaluating skin color unevenness, Patent Document 3 describes a method for evaluating skin color unevenness by acquiring color component values represented in an orthogonal coordinate system of a three-dimensional space from an image of the skin, calculating a plurality of movement vectors for an arbitrary pixel and eight pixels adjacent thereto, and calculating an average value of the movement vectors.

[0005] Patent Document 4 discloses a method for calculating skin color unevenness as a means for evaluating skin dullness. Specifically, a method is described in which color information of the skin is acquired from a skin image, a color unevenness index is acquired from the color information of the skin, and the skin dullness corresponding to the skin image is evaluated based on the color unevenness index.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Patent Document 2

Patent Document 3

[0007] In view of the above prior art, an object of the present invention is to provide a new technique for expressing and evaluating skin color unevenness. [Means for Solving the Problems]

[0008] The present invention for solving the above problems is a skin color unevenness display method including a calculation step of calculating a color difference between adjacent pixels for each pixel of a skin image of a subject, and a display step of displaying the appearance frequency of the calculated color difference value in a graph or chart.

[0009] The skin color unevenness display method of the present invention can objectively express the tendency of macro skin color unevenness of the entire skin based on micro skin color unevenness represented by the color difference per pixel.

[0010] In a preferred embodiment of the present invention, the method further includes a classification step of classifying the color difference calculated in the calculation step into two or more groups based on the calculated value, and the display step displays the composition ratio of the groups classified in the classification step as the appearance frequency of the calculated color difference value. By expressing the appearance frequency of the calculated color difference value as the composition ratio of each group, the tendency of skin color unevenness of the entire skin can be clearly recognized.

[0011] In a preferred embodiment of the present invention, the calculation step includes a step of calculating a first color difference at a first time point and a step of calculating a second color difference at a second time point after the first time point, and the display step displays, in one figure, a first graph or chart obtained from the calculated value of the first color difference and a second graph or chart obtained from the calculated value of the second color difference. By arranging and displaying graphs or the like related to a plurality of time points in this way, it is possible to easily represent the change in skin color unevenness over time.

[0012] In a preferred form of the present invention, the graph or chart is a stacked bar graph. By using a stacked bar graph, the composition ratio of the group can be easily represented.

[0013] In a preferred form of the present invention, in the display step, the skin image is displayed side by side with the graph or chart. By arranging and displaying skin images side by side, it is possible to compare the skin color unevenness sensuously recognized from the skin images with the objective skin color unevenness based on the color difference.

[0014] Further, the present invention relates to a calculation step of calculating a color difference between adjacent pixels for each pixel of a skin image of a subject, and an evaluation step of evaluating skin color unevenness based on the appearance frequency of the calculated value of the color difference, and also relates to a method for evaluating skin color unevenness. By using the method for evaluating skin color unevenness of the present invention, it is possible to objectively evaluate the tendency of macro skin color unevenness of the entire skin based on micro color unevenness represented by the color difference in pixel units.

[0015] In a preferred form of the present invention, the calculation step includes a step of calculating a first color difference at a first time point and a step of calculating a second color difference at a second time point after the first time point, and the evaluation step evaluates the change in skin color unevenness by comparing the appearance frequencies of the calculated values of the first color difference and the second color difference. By comparing the color differences at different time points, it is possible to appropriately evaluate the change in skin color unevenness.

[0016] Further, the present invention relates to a calculation means for calculating a color difference between adjacent pixels for each pixel of a skin image of a subject, and a display means for displaying the appearance frequency of the calculated value of the color difference in a graph or chart. A skin color unevenness display device comprising the same.

[0017] Furthermore, the present invention relates to a computer that For a subject's skin image, a calculation means for calculating the color difference from adjacent pixels for each pixel, A display means for displaying the appearance frequency of the calculated value of the color difference in a graph or chart, A skin color unevenness display program for functioning as.

Advantages of the Invention

[0018] According to the present invention, a new technique for expressing and evaluating skin color unevenness can be provided.

Brief Description of the Drawings

[0019]

Figure 1

Figure 2

Figure 3

Figure 4

Modes for Carrying Out the Invention

[0020] <1>Method for Displaying Color Unevenness Hereinafter, embodiments of the skin color unevenness display method of the present invention (hereinafter also referred to as the "color unevenness display method") will be described in detail. The color unevenness display method of the present embodiment includes an image acquisition step of acquiring a subject's skin image, a calculation step of calculating the color difference from adjacent pixels for each pixel, a classification step of classifying the calculated values of the color difference into two or more groups, and a display step of displaying the appearance frequency of the calculated values of the color difference in a graph or chart.

[0021] Hereinafter, the present invention will be described in detail by taking as an example the color unevenness display method of the present invention and embodiments of the display obtained by the color unevenness display method.

[0022] (1) Image acquisition step In the image acquisition step, a skin image of the subject is acquired. As the skin image, preferably, a skin image of the cheek area is used. Here, as the skin image, an image captured by a general video camera or digital camera may be used, and in the present invention, it is preferable to acquire a color image.

[0023] The skin image preferably has a resolution sufficient to withstand image analysis in subsequent steps. At the time of imaging, the resolution of the image may be appropriately set according to the number of pixels for which the color difference is to be calculated. The resolution of the image is preferably 350,000 pixels or more, more preferably 1,000,000 pixels or more, still more preferably 3,000,000 pixels or more, and even more preferably 5,000,000 pixels or more. Also, an appropriate pixel skin image may be obtained by changing the resolution of the image to an arbitrary value after image acquisition.

[0024] Note that the color unevenness display method of the present invention does not necessarily include the step of imaging a skin image, and the following steps may be performed using a previously acquired skin image.

[0025] (2) Calculation step Subsequently, a method for calculating the color difference for each pixel from the skin image will be described in detail. The color difference for each pixel is preferably calculated by defining an arbitrary region in the acquired skin image. The evaluation region is preferably in the range of 1×1 cm or more, more preferably 3×3 cm or more, and still more preferably 5×5 cm or more as the size of the actual skin region of the subject, and its shape is preferably a rectangle or a square.

[0026] The color difference can be calculated using the respective color component values of the pixels. As the color component values, the R value, G value, and B value in the RGB color system, and the L * a * b * value, a * value, a* Value, b * Values include the lightness, hue, chroma, etc. of the Munsell color system.

[0027] Here, when converting an RGB image obtained from a digital camera or the like into an L * a * b * image, an existing conversion formula can be used. That is, a standard skin color chart of 325 colors (a color chart that covers all skin colors manufactured by the Japan Color Research Institute (L * 45.6 to 85.9, a * 0.5 to 26.1, b * 1.8 to 40.7)) is used, and a conversion formula (the following formula) for converting the luminance of RGB into L * a * b * values is obtained by multiple regression analysis, and this is incorporated into image analysis application software to convert the RGB image into an L * a * b * image. In Equation 1, the correlation with the measured value refers to the correlation between the analysis result by the L * a * b * image and the analysis result by RGB analysis.

[0028]

Equation

[0029] The calculation of the color difference is performed according to the following procedure. First, one pixel is defined as the target pixel, and the eight pixels surrounding the target pixel are used as adjacent pixels (see the left side of FIG. 1). Subsequently, using the color component values of the target pixel and the adjacent pixels, a plurality of color differences are calculated for each target pixel. At this time, in order to avoid duplicate calculation of the color difference, it is preferable to calculate using four adjacent pixels for each target pixel. For example, four consecutive adjacent pixels are fixed as the calculation positions of the color difference (see the right side of FIG. 1), and the color difference is calculated using the adjacent pixels at the same calculation position for all target pixels. By using such a calculation method, duplicate calculation of the color difference can be avoided, and a sufficient amount of color difference values for displaying color unevenness can be obtained.

[0030] The color difference of the target pixel can be calculated using an existing calculation formula. For example, it can be calculated using the following formula (2).

[0031]

Equation

[0032] In the following classification process and display process, the calculated value of the color difference is used. The calculated value of the color difference may be all of the values of a plurality of color differences obtained from the target pixel, or may be a representative value of the calculated value of the color difference obtained for each target pixel. Preferably, a representative value of the calculated value of the color difference is used. The representative value of the calculated value of the color difference can be obtained by calculating the average value, median value, mode value, etc. from the calculated values of the plurality of calculated color differences.

[0033] Also, for calculating the color difference, it is preferable to use all the pixels in the evaluation region as the target pixels and calculate the color difference for each target pixel. However, when there is a region including black spots, scratches, etc. in the evaluation region, the color difference can be calculated excluding the region.

[0034] (3) Classification process Subsequently, the calculated value of the color difference is classified into two or more groups based on the value. Here, the number of groups for classifying the color difference is not particularly limited, but preferably classified into 2 to 8 groups, more preferably 2 to 5 groups, and even more preferably 2 to 3 groups.

[0035] Also, as the calculated value of the color difference used in the classification process, all the calculated values of the color differences calculated in the above (2) may be used, or a predetermined range may be set (hereinafter, the set range is referred to as the analysis range), and only the calculated values of the color differences included in the analysis range may be grouped in the classification process.

[0036] In the grouping based on the calculated value of the color difference, it is preferable to set the boundary line of the group within the range where the color difference ΔE * ab is in the range of 0.3 to 5.0. For example, when classifying into two groups based on the calculated color difference value, the boundary of the color difference ΔE * ab can preferably be set within the range of 0.3 to 5.0, more preferably 0.6 to 2.5, and even more preferably 0.6 to 1.2. By providing a boundary line within the above range for grouping, the evaluation of skin color unevenness can be performed more accurately.

[0037] Also, when classifying the calculated color difference value into three groups, the boundaries of the color difference ΔE * ab are preferably set at two locations in the ranges of 0.3 to 0.9 and 1.0 to 2.5, and more preferably in the ranges of 0.5 to 0.8 and 1.0 to 1.5. Specifically, the boundary between the first group and the second group is set at the position of ΔE * ab = 0.6, and an example where the boundary between the second group and the third group is set at the position of ΔE * ab = 1.2 is preferably cited.

[0038] Also, the grouping of the calculated color difference value may be in a form of dividing the calculated color difference value into an arbitrary number. For example, when classifying into two groups, it may be in a form of dividing the calculated color difference value into a plurality of groups by equally spacing the calculated color difference value with a predetermined value, or in a form of setting the boundary line of the group at the median value of the calculated color difference value, at a value of 1 / 3 from the minimum value of the calculated color difference value, at a value of 1 / 3 from the maximum value of the calculated color difference value, etc., which can also be preferably exemplified. When classifying into three groups, it may be in a form of dividing the calculated color difference value into a plurality of groups by equally spacing the calculated color difference value with a predetermined value, or in a form of setting the boundary line of the group at two locations selected from the median value of the calculated color difference value, at a value of 1 / 3 from the minimum value of the calculated color difference value, and at a value of 1 / 3 from the maximum value of the calculated color difference value, which can also be preferably exemplified.

[0039] In addition, when the calculated value of the color difference obtained in the calculation step is used as the evaluation result as it is, the above classification step can be omitted as appropriate. For example, in the following display step, when the calculated value of the color difference is displayed in a graph of cumulative relative frequencies, the classification step can be omitted.

[0040] (4) Display step In the display step, the appearance frequency of the calculated value of the color difference is diagrammed using a graph, chart, etc. (hereinafter also simply referred to as a graph, etc.). The appearance frequency of the calculated value of the color difference may be diagrammed using the composition ratio of the groups obtained in the classification step, or may be a mode of graphing the calculated value of the color difference obtained in the calculation step as it is.

[0041] When displaying the composition ratio of the groups as the appearance frequency of the calculated value of the color difference, the composition ratio of each group based on the total number of calculated values of the color difference as a reference (100%) can be displayed. At this time, as the calculated value of the color difference, all the calculated values of the color difference obtained for each pixel may be used, or the representative value of the calculated value of the color difference for each pixel may be used.

[0042] Also, when the analysis range of the calculated value of the color difference is set in the classification step, it is preferable to display the composition ratio of each group when the total number of calculated values of the color difference within the analysis range is used as a reference (100%).

[0043] Examples of the types of graphs, etc. include, as an example, a stacked bar graph, a bar graph, a pie chart, a graph of cumulative relative frequencies, etc. In the present invention, it is particularly preferable to use a stacked bar graph.

[0044] FIG. 2 shows an example of a display showing the composition ratio of each group as a stacked bar graph after classifying the calculated values of the color difference into three groups based on the calculated value of ΔE * ab calculated in the above classification step.

[0045] As shown in FIG. 2, each group is preferably colored in a different color in order to improve visibility. For example, the group with a smaller calculated color difference value can be colored with a higher brightness and a color closer to white, and the group with a larger calculated color difference value can be colored with a lower brightness and a color closer to black. By using such coloring, it becomes possible to intuitively recognize the degree of color unevenness shown by each group. Also, it may be a form of coloring so that the group to be noted stands out. For example, it can be a form of coloring only the group to be noted with a color having a different hue from other groups, or a form of filling only the group to be noted with a pattern such as diagonal lines.

[0046] Also, in the display step, it is preferable to display the above-described graph or the like and the skin image related to the evaluation region side by side. FIG. 2 shows a mode of displaying the skin image (left in FIG. 2) and the bar graph (right in FIG. 2) side by side. By displaying the skin image in the same region as the calculation region of the color difference side by side with the graph or the like, it is possible to compare the uneven skin color sensuously recognized from the skin image and the objective uneven skin color based on the color difference.

[0047] As an embodiment of the present invention, a form of displaying the uneven skin color obtained from a plurality of skin images in one figure is also preferably mentioned. The form will be described in detail below. Note that the above description of each step is also applicable to the present embodiment, and thus the description will be omitted as appropriate.

[0048] The method for displaying color unevenness according to the present embodiment calculates the color difference from each of a plurality of skin images, displays the composition ratio of the groups based on the calculated color difference values in a graph or the like, and displays the obtained plurality of graphs or the like in one figure.

[0049] When comparing and displaying color unevenness from a plurality of skin images as described above, in the above (3) classification step, it is preferable to set a predetermined analysis range and classify the calculated color difference values included in the analysis range into groups. Hereinafter, a method for determining the analysis range of the calculated color difference value using two skin images will be described.

[0050] First, for two skin images, calculate the frequency of the calculated color difference values for each predetermined class. At this time, it is preferable that the width of the class is 0.3.

[0051] Subsequently, for each skin image, calculate the cumulative frequency for each class using the calculated frequency. Further, for the cumulative frequency, calculate the difference in the cumulative frequency derived from the two skin images for each class. Note that the difference in the cumulative frequency is taken as an absolute value.

[0052] Using the difference in the cumulative frequency, calculate the cumulative frequency difference occupancy ratio. The cumulative frequency difference occupancy ratio represents the occupancy ratio of the difference in the cumulative frequency for each class, and can be calculated by dividing the difference in the cumulative frequency for each class by the total value of the differences in the cumulative frequency for all classes. Also, the cumulative frequency difference occupancy ratio is expressed as a ratio when the total value of the differences in the cumulative frequency for all classes is set to 1.

[0053] Finally, using the cumulative frequency difference occupancy ratio, calculate the cumulative value of the cumulative frequency difference occupancy ratio for each class. Pay attention to the class where the cumulative value of the cumulative frequency difference occupancy ratio is 0.1 to 0.99, and the range of that class can be used as the analysis range of the calculated color difference values used in the classification step. When the analysis range of the calculated color difference values is set in this way, the calculated color difference values used in the above (3) classification step can be performed using only the values included in the analysis range.

[0054] Also, in the color unevenness display method of the present invention, the plurality of skin images may be obtained from different skin regions of the same subject, or may be obtained from the same skin region of the same subject.

[0055] When using skin images obtained from the same skin region, obtain a skin image at a certain point in time (the first point in time) from a predetermined skin region, and then (that is, the second point in time) obtain a skin image again from the same skin region, and each graph or the like generated based on the calculated color difference values calculated from each skin image can be displayed in one figure.

[0056] As an example of the above embodiment, a form of comparing uneven skin color before using cosmetics and uneven skin color after using cosmetics can be cited. For example, FIG. 3 shows an embodiment in which a bar graph showing uneven skin color before using cosmetics and a bar graph showing uneven skin color three months after using cosmetics (beauty essence) are displayed in one figure.

[0057] By arranging and displaying graphs etc. before and after using cosmetics in this way, the change in uneven skin color due to the use of cosmetics can be expressed.

[0058] As the above-mentioned cosmetics, any of basic cosmetics such as lotion, emulsion, and beauty essence used for skin care, and makeup cosmetics such as foundation and base can be used, but the present invention is preferably used to display uneven skin color before and after using basic cosmetics. That is, the present invention is suitable for displaying uneven skin color of bare skin.

[0059] Here, the uneven color display method of the present invention is not limited to a form in which only one set of graphs etc. at the first time point and the second time point is displayed, and may be a form in which two or more sets of graphs etc. at the first time point and the second time point are displayed. That is, skin images of the same area of the subject may be acquired over time two or more times, and a plurality of graphs etc. created for each time point may be displayed in one figure.

[0060] Also, color differences are calculated from the skin at a plurality of time points such as before using cosmetics, one week after using, and two weeks after using, and graphs etc. based on the calculated values of each color difference are displayed in one figure, whereby the change in uneven skin color due to the use of cosmetics can be expressed.

[0061] In this way, by displaying graphs etc. at a plurality of time points related to the same skin area in one figure, it becomes possible to objectively grasp the change over time of uneven skin color. The said display method can be utilized when evaluating whether there is a change in uneven skin color when skin care products, pharmaceuticals, etc. are continuously used.

[0062] In addition, the color unevenness display method of the present invention can also be in a form of displaying graphs or the like related to different skin areas in one figure. For example, by displaying in one figure a graph or the like based on an image obtained from a predetermined skin area (first skin area) and a graph or the like based on an image obtained from a skin area different from the first skin area (second skin area), it is possible to easily compare the color unevenness for each skin area.

[0063] For example, by creating graphs or the like for each of the right cheek and the left cheek and displaying these graphs or the like in one figure, the difference in color unevenness between both cheeks can be recognized at a glance, and it is possible to easily compare the color unevenness.

[0064] <2>Method for evaluating color unevenness The method for evaluating skin color unevenness of the present invention (hereinafter, also referred to as the "color unevenness evaluation method") includes an evaluation step of evaluating skin color unevenness based on the calculated value of the color difference calculated in the calculation step in the above-described color unevenness display method. Note that the image acquisition step, the calculation step, and the classification step shall be applied with the descriptions set forth in <1> above. Hereinafter, the evaluation step in the color unevenness evaluation method of the present invention will be described in detail.

[0065] In the evaluation step, the skin color unevenness of the subject is evaluated based on the frequency of occurrence of the calculated value of the color difference obtained by the above method. Specifically, the calculated values of the color difference are classified into two or more preset groups, and the skin color unevenness in the evaluation area is evaluated based on the composition ratio of each group with the total number of the calculated values of the color difference as a reference (100%).

[0066] The grouping of the calculated values of the color difference is preferably performed according to the magnitude of the calculated values of the color difference. For example, when classifying the calculated values of the color difference into two groups, those with a calculated value of the color difference less than a predetermined value can be classified into a group with small color unevenness, and those with a calculated value of the color difference greater than or equal to the predetermined value can be classified into a group with large color unevenness.

[0067] In addition, the grouping of the calculated values of the color difference is preferably performed using the method described in the above <1>(3) classification step. For example, for the color difference ΔE * abIt is preferable to set the boundary line of the group within the range of 0.3 to 5.0.

[0068] The skin color unevenness in the evaluation area can be evaluated by comparing the composition ratios of each group. For example, when classifying the calculated color difference values into two groups, when the composition ratio of the group with a smaller calculated color difference value is larger than the composition ratio of the group with a larger calculated color difference value, there is an aspect of evaluating that the color unevenness in the evaluation area is small. When classifying the calculated color difference values into three or more groups, when the composition ratio of the group with the largest calculated color difference value is smaller than a predetermined ratio, and the composition ratio of the group with the smallest calculated color difference value is larger than a predetermined ratio, there are aspects of evaluating that the color unevenness in the evaluation area is small, and evaluating the color unevenness in the evaluation area based on the composition ratio of a group determined in advance as the evaluation target.

[0069] In addition, the color unevenness evaluation method of the present invention may be a form of evaluating skin color unevenness by using a plurality of skin images. The plurality of skin images may be obtained from different skin regions or may be obtained from the same skin region.

[0070] Hereinafter, embodiments of evaluating color unevenness using a plurality of skin images will be described in detail.

[0071] When obtaining a plurality of skin images from the same skin region, for example, calculate the color difference (the first color difference) from the skin image at a certain time point (the first time point), and then (the second time point), calculate the color difference (the second color difference) from the skin image of the same skin region. Using each color difference, classify the calculated color difference values into a plurality of groups for each time point, and the skin color unevenness in the evaluation area can be evaluated based on the composition ratio of the group.

[0072] Specifically, when classifying the calculated color difference values into two groups, when the composition ratio related to the group with a larger calculated color difference value is smaller at the second time point compared to the first time point, it can be evaluated that the skin color unevenness in the evaluation area has been improved.

[0073] Here, the color unevenness evaluation method of the present invention is not limited to a form in which only the composition ratio of a group based on the first color difference and the composition ratio of a group based on the second color difference are compared, and may be a form in which the composition ratios of two or more groups are compared. That is, the skin image of the subject may be acquired two or more times over time, the composition ratio of each group may be obtained from the calculated values of a plurality of color differences obtained from the skin image, and the change in color unevenness of the skin over time may be evaluated.

[0074] In one embodiment of the color unevenness evaluation method of the present invention, the first color difference is calculated using a skin image obtained from the skin before the use of the cosmetic, and the second color difference is calculated using a skin image obtained from the skin after the use of the cosmetic.

[0075] Further, the color unevenness evaluation method of the present invention can also be a form of evaluation by comparing the color unevenness related to different skin regions. For example, the first color difference calculated based on the image obtained from a predetermined skin region (the first skin region) and the second color difference calculated based on the image obtained from a skin region different from the first skin region (the second skin region) can be used to compare the color unevenness of both regions.

[0076] Specifically, for the skin images obtained from different skin regions, the color differences are calculated respectively, and the calculated values of the obtained color differences are classified into a plurality of predetermined groups. When the color differences are classified into two groups, by paying attention to the group with a smaller calculated value of the color difference, it can be evaluated that the skin region with a larger composition ratio of the group has less skin color unevenness.

[0077] <3> Color unevenness display device Hereinafter, the color unevenness display device will be described with reference to FIG. 4. The color unevenness display device of the present invention is a device for implementing the color unevenness display method described in the above item <1>. Therefore, the description related to the above <1> is also applicable to the following device.

[0078] As shown in FIG. 4, the color unevenness display device of the present invention includes an image acquisition unit 11 that acquires a skin image of a subject, a calculation unit 12 that calculates the color difference from adjacent pixels for each pixel in the skin image, a storage unit 13 that stores various formulas and the like necessary for calculating the color difference, and a display unit 14 that displays the appearance frequency of the calculated color difference values in a graph or chart.

[0079] The image acquisition unit 11 captures an image of the skin surface, inputs the skin surface image, and converts it into a signal in which color elements are digitized. The signal is not particularly limited as long as it can represent the skin condition. For example, RGB signals, Munsell signals, L * a * b * signals (JIS Z8729), XYZ signals, signals representing lightness, hue, and chroma, and the like can be mentioned. As the image acquisition unit 11, a general video camera or digital camera can be used.

[0080] The calculation unit 12 calculates the color difference from adjacent pixels using the color component values for each pixel in the skin image obtained by the image acquisition unit 11. The method for calculating the color difference refers to the description in <1>(2) above.

[0081] The storage unit 13 stores programs necessary for the color unevenness display device of the present invention to function, as well as various formulas necessary for calculating the color difference and the like.

[0082] The display unit 14 converts the appearance frequency of the calculated color difference values obtained by the calculation unit 12 into a graph or chart. At this time, it is preferable to display graphs and the like based on a plurality of skin images in one figure.

[0083] Further, the display unit 14 may include a classification step of classifying the calculated color difference values obtained by the calculation unit 12 into a plurality of groups before generating a graph or the like. In the color unevenness display device of the present invention, it is preferable to form a graph or chart of the composition ratio of the groups obtained in the classification step.

[0084] In addition, the display means 14 also issues an instruction to display the graph or chart on an arbitrary display device such as a liquid crystal display.

[0085] In the present invention, it is not necessarily required to include the above configuration. For example, it may be in the form of a color unevenness display device including a calculation means 12 that calculates the color difference between adjacent pixels for each pixel from the skin image of the subject, and a display means 14 that displays the appearance frequency of the calculated color difference value in a graph or chart.

[0086] <4> Color unevenness display program The color unevenness display program of the present invention causes a computer, other devices, machines, etc. to function as an image acquisition means 11 for acquiring a skin image of a subject, a calculation means 12 for calculating the color difference between adjacent pixels for each pixel from the skin image, a storage unit 13 for storing various formulas etc. necessary for the calculation of the color difference, and a display means 14 for displaying the appearance frequency of the calculated color difference value in a graph or chart (see FIG. 4).

[0087] In the present invention, it is not necessary to be a program that functions all of the above configurations. For example, it may be in the form of a color unevenness display program that causes a computer, other devices, machines, etc. to function as a calculation means 12 for calculating the color difference between adjacent pixels for each pixel from the skin image of the subject, and a display means 14 for displaying the appearance frequency of the calculated color difference value in a graph or chart.

[0088] The color unevenness display program of the present invention is a program related to the color unevenness display method and the color unevenness display device described in the above items <1> and <3>. Therefore, the explanations related to the above <1> and <3> are also valid for the color unevenness display program of the present invention.

Industrial Applicability

[0089] The present invention can be applied to a method for diagnosing skin conditions in the fields of cosmetics and pharmaceuticals, a method for evaluating cosmetic effects, and devices used for these methods.

Explanation of Signs

[0090] 1 Skin color unevenness display device 11 Image acquisition means 12 Calculation means 13 Storage unit 14 Display means

Claims

1. A computer, for a skin image of a subject, a calculating step of calculating a color difference from an adjacent pixel for each pixel; a classifying step of classifying the color differences calculated in the calculating step into two or more groups based on the calculated values; including executing a display step of displaying the appearance frequency of the calculated values of the color differences in a graph or chart, wherein the color difference is a value of ΔE * ab shown below, the calculating step includes a step of calculating a first color difference from a skin image at a first time point before the use of a base cosmetic, a skin care product, and / or a pharmaceutical, and for the same skin area as the skin area for which the first color difference was calculated, a step of calculating a second color difference from a skin image at a second time point after the use of a base cosmetic, a skin care product, and / or a pharmaceutical after the first time point; in the classifying step, a group boundary line is provided within a range of 0.3 to 5.0 for the color difference, and the calculated values of the color difference are classified into two or more groups; in the display step, as the appearance frequency of the calculated values of the color difference, the composition ratio of the groups classified in the classifying step is displayed; the skin image at the first time point and the skin image at the second time point are displayed side by side with a first graph or chart obtained from the calculated value of the first color difference and a second graph or chart obtained from the calculated value of the second color difference, and the first graph or chart and the second graph or chart are displayed in one figure; the classifying step includes a step of setting an analysis range of the calculated values of the color differences calculated in the calculating step; the step of setting the analysis range is for the calculated value of the first color difference and the calculated value of the second color difference, calculating the frequency of the calculated value of the color difference for each predetermined class; for the frequency of the calculated value of the first color difference and the frequency of the calculated value of the second color difference, calculating the cumulative frequency for each class using the calculated frequencies; for each class, calculating the absolute value of the difference between the cumulative frequency based on the calculated value of the first color difference and the cumulative frequency based on the calculated value of the second color difference; dividing the absolute value of the difference in cumulative frequencies for each class by the total value of the differences in cumulative frequencies for all classes to calculate the cumulative frequency difference occupancy ratio for each class; using the cumulative frequency difference occupancy ratio to calculate the cumulative value of the cumulative frequency difference occupancy ratio for each class, and setting the range of the class where the cumulative value of the cumulative frequency difference occupancy ratio is 0.1 to 0.99 as the analysis range of the calculated values of the color difference; including In the classification step, a method for displaying uneven skin color that performs classification using the calculated values of the color differences included in the analysis range. 【Number 1】

2. The graph or chart is a stacked bar graph, In the display step, each group is colored with a different color, The smaller the calculated value of the color difference, the higher the brightness and the closer to white it is colored, The method for displaying uneven skin color according to claim 1, wherein the larger the calculated value of the color difference, the lower the brightness and the closer to black it is colored.

3. In the classification step, the calculated values of the color differences are classified into three groups, The method for displaying uneven skin color according to claim 1 or 2, wherein the boundaries of the color differences of the groups are set at a total of two locations in the ranges of 0.3 to 0.9 and 1.0 to 2.

5.

4. The evaluation area of the skin image for calculating the color difference is square and 1×1 cm or more, The method for displaying uneven skin color according to any one of claims 1 to 3, wherein the resolution of the skin image is 5 million pixels or more.

5. In the calculation step, among the pixels of the skin image, the pixel for calculating the color difference is the target pixel, the pixels surrounding the target pixel are the adjacent pixels, and for the target pixel, the color difference is calculated using the four adjacent pixels, and the color difference of the target pixel is calculated by calculating at least one of the average value, median value, and mode value of the obtained calculated values of the plurality of color differences. The method for displaying uneven skin color according to claim 4, wherein the calculation of the color difference is performed for all the pixels in the evaluation area as the target pixels.

6. Calculation means for calculating the color difference from adjacent pixels for each pixel of the skin image of the subject, Classification means for classifying the color differences calculated by the calculation means into two or more groups based on the calculated values, Display means for displaying the appearance frequency of the calculated values of the color differences in a graph or chart, Comprising The color difference is the value of ΔE * ab shown below, The calculation means includes means for calculating a first color difference from a skin image at a first time point before the use of a base cosmetic, skin care product, and / or pharmaceutical, and for calculating a second color difference from a skin image at a second time point after the first time point and after the use of a base cosmetic, skin care product, and / or pharmaceutical for the same skin area as the skin area for which the first color difference was calculated. The classification means provides group boundary lines within the range of 0.3 to 5.0 for the color differences and classifies the calculated values of the color differences into two or more groups, The display means is As the appearance frequency of the calculated value of the color difference, display the composition ratio of the group classified by the classification means. Along with the first graph or chart obtained from the calculated value of the first color difference and the second graph or chart obtained from the calculated value of the second color difference, display the skin image at the first time point and the skin image at the second time point, and Display the first graph or chart and the second graph or chart in one figure. The classification means includes means for setting an analysis range of the calculated value of the color difference calculated by the calculation means. The means for setting the analysis range is as follows: For the calculated value of the first color difference and the calculated value of the second color difference, means for calculating the frequency of the calculated value of the color difference for each predetermined class. For the frequency of the calculated value of the first color difference and the frequency of the calculated value of the second color difference, means for calculating the cumulative frequency for each class using the calculated frequencies. For each class, means for calculating the absolute value of the difference between the cumulative frequency based on the calculated value of the first color difference and the cumulative frequency based on the calculated value of the second color difference. By dividing the absolute value of the difference in cumulative frequencies for each class by the total value of the differences in cumulative frequencies for all classes, means for calculating the cumulative frequency difference occupancy ratio for each class. Using the cumulative frequency difference occupancy ratio, calculate the cumulative value of the cumulative frequency difference occupancy ratio for each class, and set the range of classes where the cumulative value of the cumulative frequency difference occupancy ratio is 0.1 to 0.99 as the analysis range of the calculated value of the color difference. Including The classification means is a skin color unevenness display device that performs classification using the calculated value of the color difference included in the analysis range. 【Number 2】

7. A computer, For a skin image of a subject, calculation means for calculating the color difference from an adjacent pixel for each pixel, Classification means for classifying the color difference calculated by the calculation means into two or more groups based on the calculated value, Display means for displaying the appearance frequency of the calculated value of the color difference in a graph or chart, A skin color unevenness display program for functioning as, The calculation means includes means for calculating a first color difference from a skin image at a first time point before the use of a base cosmetic, a skin care product, and / or a pharmaceutical, and for calculating a second color difference from a skin image at a second time point after the first time point and after the use of a base cosmetic, a skin care product, and / or a pharmaceutical for the same skin area as the skin area where the first color difference was calculated. The color difference is the value of ΔE * ab shown below. The classification means sets a boundary line of the group within the range of 0.3 to 5.0 of the color difference, and classifies the calculated values of the color difference into two or more groups. The display means displays the composition ratio of the groups classified by the classification means as the appearance frequency of the calculated value of the color difference. displays the skin image at the first time point and the skin image at the second time point side by side with the first graph or chart obtained from the calculated value of the first color difference and the second graph or chart obtained from the calculated value of the second color difference, and displays the first graph or chart and the second graph or chart in one figure. The classification means includes means for setting an analysis range of the calculated value of the color difference calculated by the calculation means. The means for setting the analysis range includes means for calculating the frequency of the calculated value of the color difference for each predetermined class for the calculated value of the first color difference and the calculated value of the second color difference, means for calculating the cumulative frequency for each class using the calculated frequencies for the frequency of the calculated value of the first color difference and the frequency of the calculated value of the second color difference, means for calculating the absolute value of the difference between the cumulative frequency based on the calculated value of the first color difference and the cumulative frequency based on the calculated value of the second color difference for each class, means for calculating the cumulative frequency difference occupancy ratio for each class by dividing the absolute value of the difference in cumulative frequencies of each class by the total value of the differences in cumulative frequencies of all classes, means for calculating the cumulative value of the cumulative frequency difference occupancy ratio for each class using the cumulative frequency difference occupancy ratio, and setting the range of the class where the cumulative value of the cumulative frequency difference occupancy ratio is 0.1 to 0.99 as the analysis range of the calculated value of the color difference, and includes The classification means is a skin color unevenness display program that performs classification using the calculated values of the color difference included in the analysis range. 【Number 3】

Images