Skin evaluation method, evaluation program, evaluation device, cosmetic evaluation method, cosmetic treatment evaluation method, and analysis method

Abstract

To evaluate skin by an index different from a conventional one, or to find an evaluation index different from the conventional one.SOLUTION: A skin evaluation method comprising: a G kurtosis acquisition step of acquiring a G kurtosis which is calculated based on a G image obtained by separating a G component in an RGB color space from a skin image including skin of a subject and indicates a kurtosis of a luminance histogram in the G image; and an evaluation step of evaluating the skin of the subject based on the G kurtosis.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a skin evaluation method, an evaluation program, an evaluation device, a cosmetic evaluation method, and a cosmetic treatment evaluation and analysis method. [Background technology]

[0002] Techniques for evaluating human skin using images have been known for some time. For example, Patent Document 1 discloses a technique for calculating an evaluation value indicating skin age or the like based on color information indicating the brightness and unevenness of the color of a predetermined part of the face. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2021-10652 Summary of the Invention [Problem to be solved by the invention]

[0004] As such, indicators such as brightness, unevenness of color, pores, blemishes, and texture are used to evaluate skin. However, the mechanism by which people actually perceive other people's impressions is still largely unknown, and there may be other factors that influence impressions in addition to the indicators currently used.

[0005] As a result of extensive research, the inventors have discovered that there are images that cannot be explained by existing evaluation indices when it comes to skin impressions using images. This fact suggests the possibility that there may be factors that affect skin evaluation other than existing evaluation indices.

[0006] In view of the above background, an object of the present invention is to evaluate skin using an index different from conventional ones, or to find an evaluation index different from conventional ones. [Means for solving the problem]

[0007] [1] A skin evaluation method comprising: a G kurtosis acquisition step of acquiring G kurtosis, which represents the kurtosis of a brightness histogram in a G image obtained by separating the G component in an RGB color space from a skin image including the skin of a subject, and an evaluation step of evaluating the skin of the subject based on the G kurtosis.

[0008] This configuration allows skin evaluation using a new index called G-kurtosis, which has not been used in skin evaluation using skin images in the past. Furthermore, skin evaluation can be performed using skin images, making it possible to easily evaluate skin with less burden on the subject.

[0009] [2] The skin evaluation method described in [1], wherein in the evaluation step, the lower the G kurtosis, the higher the evaluation of the subject's skin.

[0010] The inventors have obtained experimental data indicating the possibility of a negative correlation between the G kurtosis in a G image acquired from a skin image and the skin evaluation. By adopting this configuration, it is possible to perform an evaluation that appropriately reflects the relationship between the G kurtosis and the skin evaluation.

[0011] [3] The skin evaluation method according to [1] or [2], wherein in the evaluation step, the appearance impression of the subject's skin is estimated based on the G kurtosis.

[0012] [4] A skin evaluation method described in any of [1] to [3], wherein in the evaluation step, an index related to blood in the subject's skin is estimated based on the G kurtosis in order to evaluate the subject's skin for cosmetic purposes.

[0013] The present inventors have found that there is a correlation between the G kurtosis in a G image acquired from a skin image and an index related to blood in the skin. With this configuration, it is possible to estimate an index related to blood in the subject's skin from the skin image based on the correlation between the G kurtosis and the index related to blood in the skin.

[0014] [5] The skin evaluation method described in [4], wherein the blood-related index includes any one of a hemoglobin index, an erythema index, and a blood vessel depth.

[0015] [6] A skin evaluation method as described in claim 1, wherein in the evaluation step, the stratum corneum moisture content in the subject's skin is estimated based on the G kurtosis in order to evaluate the subject's skin for cosmetic purposes.

[0016] [7] A skin evaluation method described in any of [1] to [6], further comprising a suggestion process for making suggestions to reduce the G kurtosis of the subject's skin based on the evaluation results in the evaluation process.

[0017] [8] The skin evaluation method described in [7], wherein the suggestions include suggestions for improving blood-related indicators in the subject's skin for cosmetic purposes.

[0018] The present inventors have found that there is a correlation between the G kurtosis in a G image acquired from a skin image and an index related to blood in the skin. With this configuration, the index related to blood can be improved based on the correlation between the G kurtosis and the index related to blood in the skin, thereby reducing the G kurtosis, which can be used to improve skin evaluation.

[0019] [9] An evaluation program that causes a computer to function as a G kurtosis acquisition unit that acquires G kurtosis, which represents the kurtosis of the brightness histogram in a G image calculated based on a G image obtained by separating the G component in RGB color space from a skin image including the skin of a subject, and an evaluation unit that evaluates the skin of the subject based on the G kurtosis.

[0020]

[10] An evaluation device comprising: a G kurtosis acquisition unit that acquires G kurtosis, which represents the kurtosis of a brightness histogram in a G image calculated based on a G image obtained by separating the G component in an RGB color space from a skin image including the skin of a subject; and an evaluation unit that evaluates the skin of the subject based on the G kurtosis.

[0021]

[11] A method for evaluating cosmetics, comprising the steps of: evaluating the skin of multiple subjects before and after using a cosmetic product using the evaluation method described in [1]; and evaluating the cosmetic product by comparing the evaluation results before and after using the cosmetic product.

[0022] With this configuration, it is possible to evaluate the cosmetic product according to changes in the skin evaluation based on the G kurtosis of the G image acquired from the skin image.

[0023]

[12] A method for evaluating a cosmetic treatment, comprising the steps of: evaluating the skin of a plurality of subjects before and after a cosmetic treatment using the evaluation method described in [1]; and evaluating the cosmetic treatment by comparing the evaluation results before and after the cosmetic treatment.

[0024] With this configuration, it is possible to evaluate the cosmetic treatment according to changes in the skin evaluation based on the G kurtosis of the G image acquired from the skin image.

[0025]

[13] An analytical method for finding indices that affect the appearance impression of skin, comprising: an impression evaluation step of having a plurality of evaluators evaluate the appearance impression of skin for a plurality of skin images; an index evaluation step of having the plurality of evaluators evaluate each skin image using a plurality of existing indices; a feature acquisition step of acquiring a plurality of types of image features from each skin image; a classification step of performing a statistical analysis on the appearance impression evaluation value and the evaluation values ​​of each existing indices for each skin image, and classifying skin images for which a significant difference is found between the appearance impression evaluation value and the evaluation values ​​of all the existing indices as a new index group, and other skin images as an existing index group; and an identification step of comparing the new index group with the existing index group for each image feature, and identifying image features for which a significant difference is found between each group as new indices that affect the appearance impression of skin.

[0026] This configuration makes it possible to discover new indices that differ from existing indices. Specifically, image features that contribute to a good impression and are not reflected in existing indices can be identified as new indices that affect the impression of skin.

[0027]

[14] The analysis method described in

[13] , wherein the skin image is obtained by removing features including at least one of pores, blemishes, acne scars, acne, moles, peeling skin, erythema, and wrinkles from a photographed image of human skin through image processing.

[0028] By adopting such a configuration, it is expected that judgments will be suppressed from being made based on elements that are particularly obvious to the eye and are existing indicators.

[0029]

[15] The analysis method according to

[13] or

[14] , wherein in the impression evaluation step, a time limit is set for the evaluator to check the skin image.

[0030] This configuration is expected to have the effect of preventing the evaluator from unconsciously (or consciously) trying to apply existing indicators to the reasons for the appearance impression evaluation.

[0031]

[16] A skin evaluation method comprising the steps of: acquiring image features, which are new indicators identified by the analysis method described in any one of

[13] to

[15] , based on a skin image including the skin of a subject; and evaluating the skin of the subject based on the image features. [Effects of the Invention]

[0032] According to the present invention, it is possible to evaluate skin using indices different from conventional ones, or to find evaluation indices different from conventional ones. [Brief explanation of the drawings]

[0033] [Figure 1] FIG. 1 is a block diagram showing the configuration of an evaluation system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a hardware configuration diagram of an information processing apparatus and a terminal device according to the embodiment. [Figure 3] 1 is a flowchart showing the steps of a skin evaluation method according to the present embodiment. [Figure 4] 10 is a flowchart showing a procedure for evaluating a skin treatment according to the present embodiment. [Figure 5] FIG. 2 is a block diagram showing the functional configuration of the analysis device of the present embodiment. [Figure 6] 1 is a flowchart showing the procedure of an analysis method according to the present embodiment. [Figure 7] 1 is a flowchart showing a procedure for evaluating a skin image in Example 1. [Figure 8] Analysis results of Example 3. DETAILED DESCRIPTION OF THE INVENTION

[0034] The present invention relates to an evaluation method, an evaluation program, and an evaluation device for evaluating the skin of a subject based on G kurtosis, which represents the kurtosis of a luminance histogram in a G image obtained by separating the G component in an RGB color space from a skin image containing the subject's skin. The present invention also relates to an evaluation method, an evaluation program, and an evaluation device for evaluating a treatment by comparing skin evaluation results based on G kurtosis before and after the treatment. The present invention also relates to an analytical method for finding indicators that affect the impression of skin.

[0035] In this invention, the RGB color space refers to a color space expressed by the three primary colors of light: red, green, and blue, and the G image refers to an image obtained by separating the G component, which is the green component of the three primary colors, from a skin image (extracting only the G component). The G kurtosis is the kurtosis of the luminance histogram in this G image.

[0036] Furthermore, the evaluation in the present invention is performed for cosmetic purposes. In the embodiment described below, blood-related indices are estimated and improvements are proposed in skin evaluation, but such estimation and proposal are not intended for medical procedures including diagnosis and treatment, but are merely estimations for cosmetic evaluation.

[0037] DETAILED DESCRIPTION OF THE INVENTION

[0014] Exemplary embodiments of the present invention will now be described in more detail with reference to the accompanying drawings, in which preferred embodiments are shown, but which may be embodied in many different forms and are not limited to the embodiments set forth herein.

[0038] For example, in this embodiment, the configuration, operation, etc. of the skin evaluation system will be described, but similar effects can be achieved by a device having similar functions, a method executed by the device, a computer program that causes a computer device to execute the method, etc. The program may be provided as a non-transitory computer-readable recording medium, or may be provided so as to be downloadable from an external server.

[0039] In the following embodiments, the term "unit" may include, for example, a combination of hardware resources implemented by a broadly defined circuit and software information processing that can be specifically realized by these hardware resources. In this embodiment, "information" is represented by, for example, the physical value of a signal value representing voltage or current, the high or low value of a signal value as a binary bit set consisting of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculation can be performed on a broadly defined circuit.

[0040] A circuit in the broad sense is a circuit realized by appropriately combining a circuit, a processor, a memory, etc. For example, it is a circuit including any of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), etc.

[0041] <1. System configuration> Fig. 1 is a block diagram showing the configuration of a skin evaluation system according to this embodiment. As shown in Fig. 1, the evaluation system 0 includes an evaluation device 1 and a terminal device 2. The evaluation device 1 and the terminal device 2 are configured to be able to communicate with each other via a network NW. In this embodiment, the network NW is an IP (Internet Protocol) network, but there are no limitations on the type of communication protocol, the type of network, etc.

[0042] The evaluation device 1 has an acquisition unit 11, an evaluation unit 12, a proposal unit 13, and an output unit 14, and is connected to a database DB so as to be able to communicate with each other. These are information processing units that are software-based and are specifically realized by hardware.

[0043] The terminal device 2 has an image capturing unit 21 and is connected to the network NW. The image capturing unit 21 captures an image of the subject's skin and acquires the captured image. The terminal device 2 then transmits the captured image to the evaluation device 1. In this embodiment, an image of the cheek is used as the skin image, so an image including at least the subject's cheek is captured to acquire the captured image. Note that in order to keep the image capturing conditions constant, an image capturing guide may be displayed during image capturing, or facial recognition or angle detection may be performed, and image capturing may be permitted if the conditions are met.

[0044] Next, the processing of each unit included in the evaluation device 1 will be described. The acquisition unit 11 acquires G kurtosis, which represents the kurtosis of the luminance histogram of a G image obtained by separating the G component in the RGB color space from the skin image. Details of the processing procedure will be described later. Note that in this embodiment, it is assumed that the acquisition unit 11 calculates the G kurtosis, but it is also possible to calculate the G kurtosis by analyzing the skin image using an external device, and the acquisition unit 11 acquires the G kurtosis.

[0045] The evaluation unit 12 evaluates the subject's skin based on the G kurtosis acquired by the acquisition unit 11. The evaluation unit 12 of this embodiment particularly evaluates the appearance impression of the subject's skin. The appearance impression of the skin refers to the impression given by the appearance of the skin. For example, the appearance impression includes any impression such as age, whether the skin is what the subject wants, whether the skin looks lively, etc. In particular, in this embodiment, the impression of "whether the skin is what the subject wants" is evaluated as the appearance impression.

[0046] The evaluation unit 12 estimates the evaluation value of the appearance impression of the skin from the G kurtosis based on the correlation between the G kurtosis and the evaluation value of the appearance impression. Here, the inventors have found a relationship in which the lower the G kurtosis, the greater the degree to which the skin is desirable (the evaluation value relating to the impression of whether the skin is desirable). Therefore, the evaluation unit 12 of this embodiment estimates a higher evaluation value of the appearance impression of the subject's skin as the G kurtosis decreases.

[0047] In this embodiment, the evaluation unit 12 further estimates a blood-related index based on the G kurtosis. The blood-related index is any index related to blood or blood vessels, and includes, for example, the hemoglobin index, the erythema index, and the blood vessel depth. The evaluation unit 12 estimates the blood-related index from the G kurtosis based on the correlation between the G kurtosis and the blood-related index.

[0048] Furthermore, in this embodiment, the evaluation unit 12 estimates the moisture content of the stratum corneum in the skin of the subject based on the G kurtosis.

[0049] The estimation of the appearance impression evaluation value, the blood-related index, and the stratum corneum moisture content can be performed, for example, by inputting the G kurtosis into an estimation model generated based on the correlation with the G kurtosis. The estimation model may be generated by any method, but it is expected that the estimation model will be generated by, for example, machine learning or statistical analysis.

[0050] Furthermore, the values ​​output by the estimation model may be divided into multiple levels, and the levels may be used as the evaluation results. For example, it is envisioned that the indices relating to the appearance of the skin, blood, and stratum corneum moisture content may each be evaluated on a five-level scale.

[0051] Furthermore, the evaluation unit 12 of this embodiment evaluates the appearance of the skin before and after the skin treatment and compares the results to further evaluate the skin treatment. Here, skin treatment refers to treatment performed to improve the skin, and primarily includes treatments applied directly to the skin, such as the use of cosmetics applied to the skin, beauty treatments including esthetics and massages. Treatments performed on the whole body, such as taking supplements and exercising, are also included in skin treatments.

[0052] Skin treatments include both one-time treatments and continuous treatments. For example, when evaluating the use of cosmetics, such as foundation, it is preferable to evaluate the appearance before and after a single treatment (application). When evaluating basic cosmetics such as lotions and serums, the appearance may be evaluated before and after continuous treatment (application) for several days to several months, for example. This allows the effects of both a one-time treatment and a continuous treatment to be evaluated.

[0053] The suggestion unit 13 determines and makes suggestions for improving the appearance of the subject's skin based on the evaluation results by the evaluation unit 12. Specifically, the suggestion unit 13 makes suggestions for reducing G-kurtosis, thereby making suggestions for improving the appearance of the subject's skin. For example, it is expected that cosmetics such as foundation that have the effect of reducing G-kurtosis will be suggested. In addition, the suggestion unit 13 may suggest shooting angles, lighting conditions, etc. that reduce G-kurtosis in the image.

[0054] Furthermore, the suggestion unit 13 may make suggestions for reducing G kurtosis by influencing the subject's own body, particularly the condition of the subject's skin, which has a correlation with G kurtosis. In this embodiment, the suggestion unit 13 makes suggestions for improving blood-related indices, which have a correlation with G kurtosis, as suggestions for reducing G kurtosis. For example, it is envisioned that suggestions may be made to improve blood-related indices by improving diet, using supplements, cosmetics, massage, etc., thereby reducing G kurtosis.

[0055] The output unit 14 outputs the evaluation results by the evaluation unit 12 and the proposal contents by the proposal unit 13 to the terminal device 2. Specifically, by displaying the evaluation results and proposals and transmitting the processing results to the terminal device 2, a screen displaying these is displayed on the terminal device 2. There are no limitations on the output format, and output may be performed by audio, video, or any other method.

[0056] One or more information processing devices 10 (computer devices), such as a general-purpose server or a personal computer, can be used as the server device constituting the evaluation device 1 and the database DB. Furthermore, a terminal device 9 (computer device), such as a personal computer, smartphone, or tablet terminal, can be used as the terminal device 2. In this embodiment, the evaluation device 1 is the information processing device 10 in which a computer program (evaluation program) that executes the evaluation method is installed.

[0057] <2. Hardware configuration> Fig. 2(a) is a hardware configuration diagram of the information processing device 10. As shown in Fig. 2, the information processing device 10 has a control unit 101, a storage unit 102, and a communication unit 103, which are used to perform the functions of each unit and each process.

[0058] The control unit 101 has a processor such as a CPU that can execute an instruction set, and executes an OS and programs. The storage unit 102 includes a volatile memory such as a RAM capable of storing an instruction set, and a non-volatile recording medium such as an HDD or SSD capable of recording an OS, an evaluation program, a DBMS, and the like. The communication unit 103 has an interface for physically connecting to a network, and controls communication with the network NW to input and output information.

[0059] Fig. 2(b) is a hardware configuration diagram of the terminal device 9. As shown in Fig. 2, the terminal device 9 has a control unit 901, a storage unit 902, a communication unit 903, an input unit 904, and an output unit 905, which are used to perform the functions of each unit and each process.

[0060] The control unit 901 has a processor such as a CPU that can execute an instruction set, and executes an OS, programs, and the like. The storage unit 902 includes a volatile memory such as a RAM capable of storing an instruction set, and a non-volatile recording medium such as an HDD or SSD capable of recording an OS, a determination program, and the like. The communication unit 903 has an interface for physically connecting to a network, and controls communication with the network NW to input and output information. The input unit 904 includes an operation input device capable of input processing, such as a touch panel or keyboard, and an audio input device capable of audio input, such as a microphone. The output unit 905 includes a display device capable of display processing, such as a display, and an audio output device, such as a speaker.

[0061] <3. Skin evaluation method> Next, a specific description will be given of a method for evaluating skin using the evaluation device 1. Fig. 3 is a flowchart showing the steps of the method for evaluating skin in this embodiment. First, in step S31, the photographing unit 21 photographs the subject's skin and generates a photographed image. Then, the terminal device 2 transmits the photographed image, and the acquisition unit 11 acquires the received photographed image as a skin image. The area of ​​skin to be photographed may be determined arbitrarily, but it is assumed that the skin of the left cheek, for example, is photographed. Alternatively, the acquisition unit may generate a skin image by receiving a photographed image of the entire face from the terminal device 2 and performing processing to cut out only the target area. Furthermore, before evaluation, image processing may be performed to remove factors that cause skin problems, such as noticeable pores, blemishes, acne scars, acne, moles, peeling skin, and wrinkles. Alternatively, a photographed image of an area that does not contain such factors may be used as the skin image.

[0062] In the following step S32, the acquisition unit 11 performs image processing to extract the G component from the skin image to create a G image. Then, in step S33, the acquisition unit 11 creates a luminance histogram of the G image and calculates its kurtosis to obtain the G kurtosis.

[0063] Then, in step S34, the evaluation unit 12 evaluates the subject's skin based on the G-kurtosis. Specifically, the evaluation value of the skin appearance impression is determined based on the correlation registered in the database DB.

[0064] Here, the inventors have found through extensive research that there is a correlation (negative correlation) in which the lower the G kurtosis, the higher the evaluation of the subject's skin appearance impression. Therefore, based on this correlation, the evaluation unit 12 evaluates the subject's skin so that the lower the G kurtosis, the higher the estimated skin appearance impression of the subject.

[0065] Furthermore, the evaluation unit 12 estimates blood-related indices based on the G-kurtosis. Examples of blood-related indices include a hemoglobin index, an erythema index, and an index related to blood vessel depth. The estimation here is performed based on the correlation between blood-related indices and G-kurtosis, which was discovered by the present inventors. Furthermore, the evaluation unit 12 further estimates the stratum corneum moisture content based on the G kurtosis.

[0066] In step S35, the proposing unit 13 determines the content of a proposal for reducing the G kurtosis based on the evaluation result. For example, it is conceivable that the evaluation value and the content of a proposal are associated with each other and registered in advance in the database DB, and the content of a proposal corresponding to the evaluation value is determined. Furthermore, the proposing unit further makes a proposal to reduce the G kurtosis by improving the blood-related index that has a correlation with the G kurtosis. Note that "improvement" here refers to changing the blood-related index in the direction that reduces the G kurtosis.

[0067] The output unit 14 then transmits the evaluation result by the evaluation unit 12 and the proposal content determined by the proposal unit 13 to the terminal device 2, thereby outputting the evaluation result and proposing an improvement. The output unit 14 performs display processing of the evaluation result and the proposal content and transmits the processing result to the terminal device 2, whereby a display screen showing the evaluation result and the proposal content is displayed on the terminal device 2.

[0068] <4. Evaluation method for skin treatment> The evaluation using the above-described evaluation method (steps S31 to S34) is performed before and after the skin treatment, and the effects of the skin treatment can be evaluated by comparing the evaluation results. Figure 4 is a flowchart showing the procedure for evaluating the skin treatment in this embodiment.

[0069] First, in step S41, the subject's skin is evaluated before the skin treatment according to the procedure described above. Next, in step S42, the skin treatment is performed. As described above, skin treatment includes one-time treatment and continuous treatment. Therefore, in step S42, the treatment to be evaluated is performed by any method and for any period. Therefore, when evaluating continuous treatment over a certain period, the period from step S41 to step S43 may be long.

[0070] After the skin treatment in step S42 is completed, the subject's skin after the skin treatment is evaluated again in step S43 using the procedure described above. Then, in step S44, the evaluation result in step S41 is compared with the evaluation result in step S43 to evaluate the treatment.

[0071] Specifically, the evaluation unit 12 determines whether the evaluation has improved based on the difference or ratio of the evaluation results of steps S41 and S43. For example, if the difference obtained by subtracting the evaluation value determined in step S41 from the evaluation value determined in step S43 is less than a predetermined reference value, it is assumed that the treatment is deemed ineffective, and if the difference is equal to or greater than the predetermined reference value, it is assumed that the treatment is effective. Furthermore, multiple reference values ​​may be set and the magnitude of the effect of the treatment may be evaluated in multiple stages.

[0072] As described above, the effectiveness of a treatment can be evaluated by evaluating the appearance before and after the treatment. This makes it possible to evaluate, for example, makeup, the skin improvement effect of continued use of cosmetics, and the effects of treatments such as aesthetic treatments and massages based on G-kurtosis.

[0073] <5. Methods for analyzing indicators that affect skin appearance> The present inventors have invented an analytical method for identifying indicators that affect the appearance of skin, and through this analytical method, have discovered that the G-kurtosis affects the appearance of skin. Such an analytical method will be described in detail below.

[0074] Fig. 5 is a diagram showing the functional configuration of the analysis device 3 in this embodiment. As with the evaluation device 1, the information processing device 10 shown in Fig. 2(a) can be used as the analysis device 3. An analysis program is recorded in the storage unit 102, and the control unit 101 executes various controls in accordance with the analysis program, causing the information processing device 10 to function as each unit described below.

[0075] The analysis device 3 has, as its functional configuration, an index acquisition unit 31, an image acquisition unit 32, a feature acquisition unit 33, a classification unit 34, and an identification unit 35. These are software-based information processing specifically realized by hardware. Details of the processing executed by each unit will be explained below with reference to FIG. 6.

[0076] 6 is a flowchart showing the steps of the analysis method executed using the analysis device 3. Note that the steps shown below are an example, and the order may be changed as desired. For example, step S65 may be performed after the acquisition of a skin image and before step S66. Furthermore, the various pieces of information in step S62 do not need to be acquired simultaneously.

[0077] In the analysis method according to the present embodiment, first, a large number of photographed images of the facial skin of a large number of subjects are prepared in advance. In step S61, the image acquisition unit 32 acquires these photographed images from, for example, the terminal device 2, and processes the images to generate and acquire skin images.

[0078] The content of image processing may be determined arbitrarily. For example, image processing may be performed to remove factors that may cause skin problems, such as noticeable pores, blemishes, acne scars, acne, moles, peeling skin, and wrinkles. Furthermore, appropriate cropping (trimming) may be performed so that the skin areas represented by each skin image are similar. By removing factors that may cause skin problems through such image processing, it is expected that the evaluation in step S62 (described below) will not be strongly influenced by existing skin problems.

[0079] Note that selecting an image that does not have such factors may replace the process of generating a skin image. Image processing is not required, and the captured image may be used as a skin image as is. Furthermore, it is preferable to select an image whose "brightness," which is thought to affect the appearance of the skin, falls within a certain range.

[0080] Next, in step S62, for each skin image acquired by the image acquisition unit 32, a number of evaluators subjectively determine an evaluation value of the skin appearance impression and an evaluation value based on existing indices (an impression evaluation step and an index evaluation step).Then, the index acquisition unit 31 acquires these evaluation values.

[0081] Here, the evaluation of the appearance impression and the evaluation of the existing indexes are preferably carried out by setting a limit on the time for the evaluator to check, presenting a skin image for a short time, and then having the evaluator input evaluation values ​​for the appearance impression and the existing indexes without looking at the skin image. This is because if the skin image is presented for a long time and the evaluator is given time to think, there is a possibility that the evaluator will try to apply the reason for the appearance impression he or she felt to the existing indexes.

[0082] Any impression can be used as the evaluation value of the appearance impression, but in this embodiment, an evaluation value is obtained from the perspective of "whether the skin is what the evaluator wants to have." Furthermore, as existing indices, a plurality of commonly used evaluation indices are used to obtain evaluation values ​​for each. Here, obtaining evaluation values ​​for as many existing indices as possible makes it easier to find more novel indices.

[0083] Here, the existing indexes are existing evaluation indexes known to be usable for facial appearance impressions. In particular, indexes known to be correlated with the evaluation of facial appearance impressions are cited as examples of existing evaluation indexes. More specifically, it is expected that indexes that can be evaluated from an image, such as indexes related to noticeable pores, blemishes, acne scars, acne, moles, peeling skin, wrinkles, etc., will be used as existing evaluation indexes. Note that the order in which the skin image, the evaluation value of the appearance impression, and the evaluation value of the existing index are obtained does not matter.

[0084] Then, in step S63, a statistical analysis of the evaluations by multiple evaluators is performed on the appearance impression evaluation value and the evaluation value of the existing index for each skin image. Specifically, it is confirmed whether there is a significant difference between the appearance impression evaluation value and the evaluation value of the existing index. In this embodiment, a significant difference test is performed between the appearance impression evaluation value and the evaluation value of the existing index, particularly for skin images whose appearance impression evaluation value is equal to or greater than a predetermined reference value. This confirms whether skin images that are highly rated for appearance impression are also similarly rated using the existing index.

[0085] Next, in step S64, the classification unit 34 classifies the skin images whose appearance impression evaluation value is equal to or greater than a predetermined reference value based on the result of step S63 into two groups: a new index group that is thought to be influenced by the new index, and an existing index group that is thought to be evaluated by the existing index (classification step). Specifically, skin images that have a significant difference between the evaluation value of the existing index and the evaluation value of the appearance impression may be classified as the new index group, and images that do not have a significant difference may be classified as the existing index group.

[0086] In this embodiment, evaluation values ​​of multiple types of evaluation indices are obtained as existing indices, and a significant difference test is performed for each of them with respect to the evaluation value of the appearance impression. It is preferable that skin images for which there are significant differences between the evaluation values ​​of the appearance impression and the evaluation values ​​of the appearance impression for multiple evaluation values ​​among these are classified as a new index group. In particular, in this embodiment, only skin images for which there are significant differences between the evaluation values ​​of the appearance impression and the evaluation values ​​of all existing indices are classified as a new index group, and skin images for which there is at least one existing index that is not significantly different from the evaluation value of the appearance impression are classified as a new index group.

[0087] As a result, skin images with high evaluation values ​​for appearance impression and significant differences from the evaluation values ​​of existing indices become the new index group, and skin images with high evaluation values ​​for appearance impression and no significant differences from the evaluation values ​​of existing indices become the existing index group. Therefore, it is thought that the skin images in the new index group may have received high evaluations for appearance impression due to the influence of new indices that are not reflected in the existing indices.

[0088] In step S65, the feature acquisition unit 33 acquires image features for each of the skin images classified in this manner (feature acquisition step). In particular, it is preferable to acquire multiple types of image features. The image features may be acquired using any image analysis method. In particular, it is preferable that the feature acquisition unit 33 acquires multiple types of image features that have not been used in skin evaluation so far.

[0089] In step S66, the identification unit 35 calculates and compares the average values ​​of the new index group and the existing index group for each image feature. The identification unit 35 then identifies the image feature with a significant difference in the average value as a new index that affects the appearance of the skin (identification step).

[0090] As described above, the analysis method of this embodiment makes it possible to identify image features that affect the appearance impression and are not reflected in existing indices. This allows for the definition of new evaluation indices that can be used for skin evaluation and improvement, and even for the evaluation of skin treatments such as cosmetics. [Example]

[0091] Examples of the analytical method according to the present invention will be described below. Note that the examples described below are merely examples, and it goes without saying that the present invention is not limited to the following embodiments.

[0092] (1) Preparation of skin images From a photographed image including the subject's cheek, the cheek area, excluding the outline, eyes, nose, mouth, etc., was cut out, and image processing was performed to remove elements of skin problems, thereby generating a skin image. Specifically, correction was performed to remove visually identifiable elements of skin problems, such as noticeable pores, blemishes, acne scars, acne, moles, peeling skin, and wrinkles. Image processing can be performed using any image editing tool.

[0093] Using the above procedure, skin images were generated for 31 women in their 20s and 30s, and the following evaluations were conducted on a total of 31 skin images. For the skin images, images were selected that had a certain range of "brightness," which is thought to have a significant impact on skin impression. More specifically, the L* value in a color space consisting of the three components "L*: brightness," "a*: redness," and "b*: yellowness" was measured at the intersection of the outer corner of the left eye and the tip of the nose, and images with an L* value (brightness) above a predetermined standard value were used.

[0094] (2) Evaluator The evaluators were 100 ordinary women in each age group: 20s, 30s, 40s, 50s, and 60s, for a total of 500 women.

[0095] (3) Evaluation Procedure Each of the 31 skin images described above was evaluated by each evaluator based on the appearance impression and on existing indices. 7 is a flowchart showing the procedure for an evaluator to evaluate one skin image. As shown here, first, in step S71, one target skin image is presented to the evaluator. The display here is preferably for a short period of time, such as a few seconds, for example, 5 seconds or less. In this embodiment, the skin image presentation time in step S71 is set to 2 seconds.

[0096] Next, in step S72, no skin image was presented, and the evaluator input an evaluation value based on the appearance impression of the skin image presented in step S71 from the perspective of "is this the skin that the evaluator wants to have?" Similarly, in step S73, no skin image was presented, and the evaluator inputs evaluation values ​​of existing indices for the skin image presented in step S71. As for the existing indices, several existing indices used for evaluating skin, such as transparency, brightness, and firmness, were adopted.

[0097] The evaluation scores for the appearance impression and the evaluation scores for the existing indicators were both evaluated on a 6-point scale and selections were accepted. There was no time limit for inputting (selecting) the evaluation scores. In this way, by displaying a skin image for a short time and then prompting the user to input an evaluation value for the appearance impression and an evaluation value for the existing index, the impression received by the user can be acquired directly as an evaluation value.

[0098] (4) Statistical analysis First, the following statistical analysis was performed on skin images whose average evaluation score of the appearance impression indicating "whether the skin is desirable" among 500 evaluators was equal to or greater than a reference value. In this example, the reference value was set to 3.5.

[0099] Next, for each target skin image, a significant difference test was conducted between the evaluation score of the appearance impression by 500 evaluators and the evaluation score of each existing index. This confirmed whether there was a significant difference between the evaluation score of each existing index and the evaluation score of the appearance impression for each skin image.

[0100] Skin images that showed significant differences between the evaluation values ​​of all existing indices and the evaluation values ​​of the appearance impression were classified as a new index group, while skin images that did not show significant differences between the evaluation values ​​of at least one of the existing indices and the evaluation values ​​of the appearance impression were classified as an existing index group.

[0101] (5) Identification of new indicators Image features were obtained for each of the skin images included in the new index group and the existing index group. Specifically, we first created images by decomposing the skin image into color space components. Specifically, we generated images for each component in the color space consisting of the three components "L*: brightness," "a*: redness," and "b*: yellowness," the color space consisting of the three components "Hue: hue," "Saturation: saturation," and "Brightness," and the color space consisting of the three components "R: red," "G: green," and "B: blue." Below, these images will be referred to as "L* image," "R image," etc., respectively.

[0102] Then, for the brightness of the nine images for each color component, basic statistical quantities such as the mean, standard deviation, mode, minimum value, maximum value, skewness, and kurtosis were analyzed and used as image features. Note that skewness and kurtosis refer to the skewness and kurtosis of the brightness histogram.

[0103] By comparing the new index group with the existing index group for these image features, new indexes were identified. Specifically, the average values ​​of the new index group and the existing index group were calculated for each image feature, and significant differences were confirmed. Image features that showed significant differences between the new index group and the existing index group were then identified as new indexes that affect the appearance impression of skin.

[0104] (6) Results Analysis using the above procedure confirmed significant differences between the new index group and the existing index group for eight image feature quantities: L* kurtosis (kurtosis of the brightness histogram of the L* image), Hue maximum value, Saturation mode, G kurtosis (kurtosis of the brightness histogram of the G image), B mean value, B mode value, B maximum value, and fractal dimension. Therefore, these eight image feature quantities were identified as new indices that affect the appearance impression of skin.

[0105] More specifically, a significant positive difference was confirmed in the most frequent saturation value, with the new index group being significantly larger. Additionally, significant negative differences were confirmed in the L* kurtosis (kurtosis of the brightness histogram of the L* image), Hue maximum value, G kurtosis (kurtosis of the brightness histogram of the G image), B mean value, B mode value, B maximum value, and fractal dimension, with the new index group being significantly smaller.

[0106] As described above, this example clarified that among skin images with a predetermined or higher evaluation value for appearance impression, there are images in which there is a significant difference between the evaluation value of the existing index and the evaluation value of appearance impression. This suggests that there are elements that give a good impression that are not reflected in the existing index. By comparing the image features of skin images from the new index group and the existing index group, we found different image features in the skin images of each group. These image features are suggested to be elements that give a good impression of skin, which are not reflected in the existing indexes.

[0107] Based on these results, a skin evaluation method can be provided, which includes an acquisition step of acquiring any of the image features L* kurtosis (kurtosis of the brightness histogram of the L* image), Hue maximum value, Saturation mode, G kurtosis (kurtosis of the brightness histogram of the G image), B average value, B mode value, B maximum value, and fractal dimension, and an evaluation step of evaluating the skin of a subject based on the image features. More specifically, a skin evaluation method is provided in which, based on the significant difference between the new index group and the existing index group in each of the image features described above, if a positive difference is observed (the new index group is larger), the larger the image feature is, the higher the subject's skin is evaluated; and if a negative difference is observed (the new index group is smaller), the smaller the image feature is, the higher the subject's skin is evaluated. [Example]

[0108] Furthermore, the inventors conducted an experiment to examine whether G-kurtosis actually affects the evaluation of the appearance impression that indicates "whether the skin is what one desires." Note that the examples described below are merely examples, and it goes without saying that the present invention is not limited to the following embodiments.

[0109] (1) Preparation of skin images One skin image was created in the same manner as in Example 1. Then, an image was created by processing the skin image so that the G kurtosis was reduced. These two images were evaluated by an evaluator.

[0110] (2) Evaluator The evaluators were 127 ordinary women in their 20s to 60s. However, those who work in a field related to skin evaluation and those with visual impairments were excluded from the evaluators.

[0111] (3) Evaluation Procedure The original image before processing and the image processed to reduce G-kurtosis were shown to each evaluator in turn for the same amount of time. After that, without showing either image, the evaluator was asked to answer which image was rated higher (whether it represented the skin the evaluator wanted to have). Following this procedure, responses were obtained from each evaluator three times.

[0112] (4) Results In a total of 381 evaluations (127 people x 3 times), 95% of the time the images processed to lower the G-kurtosis were rated as having the skin the evaluator wanted. In other words, images with lower G-kurtosis were rated more highly in terms of appearance impression. These results suggest that the lower the G kurtosis, the higher the evaluation of appearance impression. [Example]

[0113] Furthermore, the inventors performed a correlation analysis between G kurtosis and blood-related indices. Specifically, a skin image of a subject was generated using the above-described procedure, and the G kurtosis was measured. Then, various blood-related indices were measured for the same subject. Three blood-related indices were measured: hemoglobin index, erythema index, and vascular depth.

[0114] The analysis results are shown in Figure 8. Figure 8(a) shows the relationship between G kurtosis and hemoglobin index, Figure 8(b) shows the relationship between G kurtosis and erythema index, and Figure 8(c) shows the relationship between G kurtosis and vascular depth. In this way, a tendency for multiple types of blood indices to show a correlation with G kurtosis was confirmed. [Example]

[0115] Furthermore, the present inventors performed a correlation analysis between G-kurtosis and the moisture content of the stratum corneum of skin. Specifically, skin images of subjects were generated using the above-described procedure, and their G-kurtosis was measured. The moisture content of the stratum corneum was then measured for the same subjects. Any known technique can be used to measure the moisture content of the stratum corneum. For example, a device that measures the moisture content of the stratum corneum using a technique such as the capacitance method may be used.

[0116] A correlation analysis revealed a correlation coefficient of -0.337, p=0.0298. These results indicated that there is a correlation between lower G-kurtosis and higher stratum corneum moisture content. Generally, the higher the stratum corneum moisture content, the better the skin condition is evaluated, and the correlation with multiple indicators suggests that the lower the G-kurtosis, the better the skin condition. [Explanation of symbols]

[0117] 0: Rating System 1: Evaluation device 11: Acquisition part 12: Evaluation section 13: Proposal Department 14: Output section 2: Terminal device 21: Photography Department 3: Analyzer 31: Indicator acquisition part 32: Image acquisition unit 33: Feature acquisition unit 34: Classification section 35: Specific part 9: Terminal device 901: Control unit 902: Storage section 903: Communications Department 904: Input section 905: Output section 10: Information processing device 101: Control unit 102: Storage section 103: Communications Department NW: Network

Claims

1. a G kurtosis acquisition step of acquiring G kurtosis representing the kurtosis of a luminance histogram in a G image obtained by separating a G component in an RGB color space from a skin image including the skin of a subject; A skin evaluation method comprising an evaluation step of evaluating the subject's skin based on the G kurtosis.

2. The skin evaluation method according to claim 1 , wherein in the evaluation step, the lower the G kurtosis, the higher the evaluation of the subject's skin.

3. The skin evaluation method according to claim 1 , wherein the evaluation step estimates an appearance impression of the subject's skin based on the G-kurtosis.

4. The skin evaluation method according to claim 1 , wherein in the evaluation step, an index relating to blood in the subject's skin is estimated based on the G kurtosis in order to evaluate the subject's skin for cosmetic purposes.

5. The skin evaluation method according to claim 4 , wherein the blood-related index includes any one of a hemoglobin index, an erythema index, and a blood vessel depth.

6. The skin evaluation method according to claim 1 , wherein in the evaluation step, the stratum corneum moisture content in the subject's skin is estimated based on the G kurtosis in order to evaluate the subject's skin for cosmetic purposes.

7. The skin evaluation method according to claim 1 , further comprising a suggestion step of making a suggestion for reducing G kurtosis in the skin of the subject based on the evaluation result in the evaluation step.

8. The skin evaluation method according to claim 7 , wherein the suggestions include suggestions for improving blood-related indicators in the subject's skin for cosmetic purposes.

9. a G kurtosis acquisition unit that acquires G kurtosis representing the kurtosis of a luminance histogram in a G image obtained by separating a G component in an RGB color space from a skin image including the skin of a subject; and An evaluation program that causes a computer to function as an evaluation unit that evaluates the subject's skin based on the G kurtosis.

10. a G kurtosis acquisition unit that acquires G kurtosis representing the kurtosis of a luminance histogram in a G image obtained by separating a G component in an RGB color space from a skin image including the skin of a subject; and An evaluation device comprising: an evaluation unit that evaluates the subject's skin based on the G kurtosis.

11. a step of evaluating the skin of a plurality of subjects before and after using a cosmetic product by the evaluation method according to claim 1; and evaluating the cosmetic by comparing the evaluation results before and after use of the cosmetic.

12. a step of evaluating the skin of a plurality of subjects before and after a cosmetic treatment using the evaluation method of claim 1; A method for evaluating a cosmetic treatment, comprising a step of evaluating the cosmetic treatment by comparing evaluation results before and after the cosmetic treatment.

13. An analytical method for finding an index that influences skin appearance, comprising: an impression evaluation step of having a plurality of subjects evaluate the appearance impression of skin for the plurality of skin images; an index evaluation step of having a plurality of subjects evaluate each skin image using a plurality of existing indices; a feature acquisition step of acquiring a plurality of types of image feature amounts from each skin image; a classification step of performing a statistical analysis of the evaluation value of the appearance impression and the evaluation values ​​of each of the existing indices for each skin image, and classifying skin images for which a significant difference is found between the evaluation value of the appearance impression and the evaluation values ​​of all of the existing indices as a new index group, and classifying other skin images as an existing index group; and an identifying step of comparing the new index group with an existing index group for each image feature, and identifying image features for which a significant difference is found between the groups as new indexes that affect the appearance impression of skin.

14. The analysis method according to claim 13, wherein the skin image is obtained by removing features including at least one of pores, blemishes, acne scars, acne, moles, peeling skin, erythema, and wrinkles from a photographed image of human skin by image processing.

15. The analysis method according to claim 13 , wherein a time limit is set for the subject to check the skin image in the impression evaluation step.

16. A step of acquiring an image feature amount, which is a new index identified by the analysis method according to any one of claims 13 to 15, based on a skin image including the skin of a subject; and evaluating the subject's skin based on the image features.

Images