KR20200068440A - Skin diagnostic system using pore and pigmentation condition of skin and diagnostic method therefor - Google Patents

Abstract

The present invention relates to a skin diagnosis system which includes a skin diagnosis device for diagnosing a skin condition using a skin image of a subject. The skin diagnosis device comprises: a pore area extractor for extracting a pore area of the skin of a subject; and a pore feature value calculator for calculating feature values of at least one pore for each of at least one designated area in an image from which the pore area is extracted. The feature values of the at least one pore calculated by the pore feature value calculator includes a ratio of an area occupied by the total pores of an area to the total area of the area calculated for each of the at least one designated area.

Description

SKIN DIAGNOSTIC SYSTEM USING PORE AND PIGMENTATION CONDITION OF SKIN AND DIAGNOSTIC METHOD THEREFOR} Skin diagnostic system using skin pores and pigmentation

The present invention relates to a skin diagnostic system using skin pores and pigmentation, and a diagnostic method thereof.

In general, skin pores and pigmentation are analyzed by processing skin images. However, the analyzed results are derived from the status information of the pores and/or the state of pigmentation, and it is expected that it is difficult for the user to directly utilize the analyzed results.

That is, it will be necessary to increase the utilization of the user by allowing the user to easily know the state information of the skin from the state of the pores and the state of pigmentation.

Domestic Publication No. 10-2017-0014919: Electronic device and method for detecting skin condition in electronic device (published on February 08, 2017).

The present invention is an invention aimed at solving the technical problems as described above, and a skin diagnosis system and a method for diagnosing the skin and enabling the user to easily grasp the state information of the skin by indexing the state of the pores and the state of pigmentation The purpose is to provide.

In addition, it is an object of the present invention to provide a skin diagnosis system and a method for diagnosis so that the state of pores and the state of pigmentation can be properly fused and expressed as one recommended makeup concentration information.

The skin diagnosis method of the present invention includes: a pore region extraction step of extracting a pore region of a subject's skin; And a pore feature value calculating step of calculating a feature value of at least one pore for each of at least one designated region of the image from which the pore region was extracted in the pore region extraction step.

Specifically, the feature values of at least one pore calculated in the step of calculating the feature values of the pores are the ratio of the area occupied by all the pores of the region to the total area of the region and the total pores of the region of the pore shape value It is characterized by including the average value for.

In addition, the shape value of the pores is preferably a value obtained by dividing the length of the long axis, which is the longest side across the pore region, by the length of the shortest variable, the short axis.

In addition, the skin diagnostic method of the present invention, using the feature value of at least one pore calculated in the pore feature value calculation step, the first to calculate the first cover index indicating the concentration of makeup for the cover of the pore area It characterized in that it further comprises a; cover index calculation step.

Specifically, the first cover index is the ratio of the area occupied by the total pores of the corresponding area to the total area of the corresponding area calculated for each of the at least one designated area, relative to the total pores of the corresponding area of the shape value of the pores It is preferable to use the 1-1 value, which is the product of the average value.

In addition, the pore region extraction step, the polarizing plate is inserted into the rear end of the light source to irradiate the polarized light to the subject, and the polarizing plate is not inserted in the front end of the camera to capture an unpolarized image; Alternatively, the pore region may be extracted using an image captured in an environment in which no polarizing plate is inserted between the light source and the camera.

In addition, the skin diagnosis method of the present invention, a pigmentation region extraction step of extracting a pigmentation region of the subject's skin; And a pigmentation feature value calculating step of calculating at least one characteristic value of at least one pigmentation for each of the at least one designated region of the image from which the pigmentation region was extracted in the pigmentation region extraction step. do.

Preferably, the characteristic value of at least one pigmentation calculated in the step of calculating the pigmentation characteristic value is characterized in that an average value of blue pixel values in the RGB domain of the corresponding region is used.

In addition, the skin diagnosis method of the present invention calculates a second cover index indicating the concentration of makeup for covering the pigmentation site by using at least one characteristic value of pigmentation calculated in the pigmentation feature value calculation step. It is preferable to further include; a second cover index calculation step.

Specifically, the second cover index is characterized by using a second value obtained by subtracting the average value of the blue pixel value of the corresponding region from the maximum blue pixel value that can be represented in the RGB domain.

In addition, in the step of extracting the pigmentation region, an image captured in an environment in which a first polarizing plate is inserted at a rear end of a light source to irradiate polarized light to a subject, and a second polarizing plate is inserted at a front end of the camera to capture a polarized image. It is preferable to extract the pigmentation region by using, but the axis of the first polarizing plate and the axis of the second polarizing plate are preferably at right angles.

The skin diagnosis system of the present invention includes a skin diagnosis device for diagnosing a skin condition using a skin image of a subject, wherein the skin diagnosis device comprises: a pore area extractor for extracting a pore area of the subject's skin; And a pore feature value calculator that calculates a feature value of at least one pore for each of the at least one designated region of the image from which the pore region has been extracted by the pore region extractor.

In addition, the feature value of at least one pore calculated by the pore feature value calculator is the ratio of the area occupied by all the pores in the region to the total area of the region and the pores' shape values for the entire pores in the region. It is preferred to include the average value.

Specifically, the shape value of the pores is characterized by dividing the length of the longest axis, which is the longest side that crosses the pore region of the pore, by the length of the shortest, variable short axis.

Preferably, the skin diagnosis apparatus of the present invention, using the feature value of at least one pore calculated by the pore feature value calculator, calculates a first cover index indicating the concentration of makeup for covering the pore area It characterized in that it further comprises a; first cover index calculator.

Specifically, the first cover index is a value of 1-1, which is a ratio of a ratio of an area occupied by all the pores of the area to the total area of the area by multiplying the average value of all pores of the area by the shape value of the pores. It is preferred to use.

In addition, the pore region extractor, the polarizing plate is inserted into the rear end of the light source to irradiate the polarized light to the subject, and the polarizing plate is not inserted in the front end of the camera to capture an unpolarized image; Alternatively, the pore region may be extracted using an image captured in an environment in which no polarizing plate is inserted between the light source and the camera.

In addition, the skin diagnosis apparatus, a pigmentation region extractor for extracting a pigmentation region of the subject's skin; And a pigmentation feature value calculator for calculating at least one characteristic value of at least one pigmentation for each of at least one designated region of the image from which the pigmentation region was extracted by the pigmentation region extractor; ,

Specifically, it is preferable to use the average value of the blue pixel values in the RGB domain of the corresponding region as the feature value of at least one pigmentation calculated by the pigmentation feature value calculator.

In addition, the skin diagnostic apparatus calculates a second cover index indicating the concentration of makeup for covering the pigmentation site by using at least one characteristic value of pigmentation calculated by the pigmentation feature value calculator. 2 cover index calculator; characterized in that it further comprises.

The second cover index is characterized by using a second value obtained by subtracting the average value of the blue pixel values of the corresponding area from the maximum blue pixel value that can be represented in the RGB domain.

In addition, the dye deposition region extractor, the first polarizing plate is inserted into the rear end of the light source to irradiate the polarized light to the subject, the second polarizing plate is inserted in the front end of the camera to capture the image captured in the environment to capture the polarized image Using the extraction of the pigmentation region, the axis of the first polarizing plate and the axis of the second polarizing plate are characterized in that they form a right angle.

According to the skin diagnosis system and the diagnostic method of the present invention, the state of the pores and the state of pigmentation can be indexed, so that the user can easily grasp the state information of the skin and fuse the state of the pores and the state of pigmentation appropriately It can be expressed as the concentration information of one recommended makeup.

1 is a block diagram of a skin diagnosis system using skin pores and pigmentation according to a preferred embodiment of the present invention.
Figure 2 is a block diagram of a skin image acquisition device according to an embodiment of the present invention.
Figure 3 is a block diagram of a skin diagnostic apparatus according to an embodiment of the present invention.
4 is an exemplary view of a first acquired image output from an image acquisition device processed by a pore region extractor.
5 is an exemplary view of a designated area for analyzing a pore area from a user by the first designated area input device.
6 is an exemplary view of an image obtained by combining a first acquired image and an image obtained by secondarily removing noise by an image preprocessor.
7 is an explanatory diagram for calculating the shape value and the slope value of the pores.
8 is an exemplary view of a second acquired image output from an image acquisition apparatus processed by a pigmentation region extractor.
9 is an exemplary view of a designated area to analyze a pigmentation area from a user by a second designated area input device.

Hereinafter, a skin diagnosis system using the skin pores and pigmentation state according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

It goes without saying that the following examples of the present invention are only intended to embody the present invention and do not limit or limit the scope of the present invention. From the detailed description and examples of the present invention, what can be easily inferred by experts in the technical field to which the present invention pertains is interpreted as belonging to the scope of the present invention.

First, FIG. 1 shows a configuration diagram of a skin diagnosis system 1000 using pores and pigmentation of skin according to a preferred embodiment of the present invention.

As can be seen from FIG. 1, the skin diagnosis system 1000 of the present invention includes a skin image acquisition device 100 and a skin diagnosis device 200.

The skin image acquisition device 100 serves to acquire skin images of a subject under various light sources 110 and a polarization environment. In addition, the skin diagnosis apparatus 200 serves to diagnose the skin condition using the skin image acquired by the skin image acquisition apparatus 100.

2 is a block diagram of an apparatus 100 for acquiring skin images according to a preferred embodiment of the present invention.

As can be seen from FIG. 2, the skin image acquisition device 100 includes a light source 110, a polarization unit 120, and an image acquisition unit 130.

The light source 110 of the present invention is preferably irradiated to the skin by indirect light.

The polarization unit 120 includes a first polarizer 121 and a second polarizer 122. That is, the polarization unit 120 directly irradiates light from the light source 110 to the skin of the subject or irradiates light polarized by the first polarizer 121. In addition, the polarization unit 120 polarizes the image of the subject's skin by the second polarizer 122 so that the image acquisition unit 130 can capture the image.

A polarizing plate may be inserted into the first polarizer 121 and the second polarizer 122. That is, it goes without saying that any polarizing plate may not be inserted into the first polarizer 121 and the second polarizer 122 as necessary.

When the polarizers of the same axis are inserted into the first polarizer 121 and the second polarizer 122, that is, the case where the polarizers with parallel axes are inserted is referred to as parallel polarization. In addition, a case in which a polarizing plate having an orthogonal axis is inserted into the first polarizer 121 and the second polarizer 122 is referred to as cross polarization.

The image acquisition unit 130 may acquire a skin image of the subject. The image acquisition unit 130 may be, for example, a camera. The skin image obtained by the image acquisition unit 130 is preferably stored in a storage unit (not shown) and processed by the skin diagnosis apparatus 200. The storage unit may be implemented using a memory or a hard disk, or may be configured as a part of the skin diagnostic device 200, or a separate storage device may be used.

The skin diagnosis apparatus 200 may be implemented by a computing device including a processor, a display, and the like on which a computer program is mounted.

The skin diagnosis apparatus 200 of the present invention processes an image using a first acquired image and a second acquired image.

The first acquired image is an image captured by the image acquisition unit 130 by the first polarization environment, and the second acquired image is an image captured by the image acquisition unit 130 by the second polarization environment. The first polarization environment and the second polarization environment may be determined by inserting a polarizing plate into the first polarizer 121 and the second polarizer 122. That is, the first polarization environment and the second polarization environment may be determined by polarization between the image acquisition unit 130 from the light source 110 and the type of polarization.

Specifically, in the first polarization environment, a polarizing plate is inserted into the first polarizer 121 which is a rear end of the light source 110 to irradiate polarized light to the subject, and the second polarizer 122 that is a front end of the image acquisition unit 130 An environment in which an unpolarized image is captured because a polarizing plate is not inserted into the environment; Or, an environment in which no polarizing plate is inserted between the light source 110 and the image acquisition unit 130. That is, the first polarization environment, when irradiating the light of the light source 110 to the subject irradiated with polarized light or directly irradiated with the light of the light source 110, the front of the image acquisition unit 130 is imaged without polarization Environment.

In addition, in the second polarization environment, the first polarizer is inserted into the first polarizer 121 at the rear end of the light source 110 to irradiate the polarized light to the subject, and the second polarizer at the front end of the image acquisition unit 130 ( 122) is an environment in which a second polarizing plate is inserted to capture a polarized image. Moreover, it is preferable that the axis|shaft of a 1st polarizing plate and the axis|shaft of a 2nd polarizing plate are a cross polarization environment which forms a vertical.

3 is a block diagram of a skin diagnosis apparatus 200 according to a preferred embodiment of the present invention.

As can be seen from Figure 3, the skin diagnostic apparatus 200 according to an embodiment of the present invention, the first cover index calculation unit 210, the second cover index calculation unit 220 and the final makeup concentration recommendation It may be configured to include a portion 230.

The first cover index calculating unit 210 serves to calculate a first cover index indicating the concentration of makeup for covering the pore area using the first acquired image. In addition, the second cover index calculation unit 220 serves to calculate a second cover index indicating the concentration of makeup for covering the pigmentation site using the second acquired image.

In addition, the final makeup concentration recommendation unit 230 uses the first cover index and the second cover index calculated from the first cover index calculation unit 210 and the second cover index calculation unit 220, respectively, to determine the final makeup concentration. It serves to recommend.

Specifically, the final makeup concentration recommendation unit 230 uses a table in which the concentration of makeup is set in multiple stages according to the value of the first cover index in relation to the pores. In addition, the final makeup concentration recommendation unit 230 uses a table or the like in which the concentration of makeup is set in multiple stages according to the value of the second cover index in relation to pigmentation. Finally, the final makeup concentration recommendation unit 230 selects a step indicating the concentration of the darker makeup among the concentration stage of makeup by the first cover index and the makeup stage by the second cover index, to the user It is recommended to use the appropriate makeup level. This is because light makeup is required for the pores, and if the pigmentation is severe, it is necessary to wear dark makeup. If pigmentation is required, the thin makeup is required.

Specifically, the first cover index calculator 210 includes a pore area extractor 211, a first designated area input device 212, an image preprocessor 213, a pore feature value calculator 214, and a first cover index It is preferably configured to include a calculator (215).

The pore region extractor 211 primarily serves to remove noise from the first acquired image and to generate the first image. The high-pass filter may be used to remove noise by the pore region extractor 211. Alternatively, the removal of noise by the pore region extractor 211 may be implemented by processing a first acquired image by a Gaussian blur using a Gaussian blur method and subtracting it from the original first acquired image. The noise removed by the pore region extractor 211 may also include a pigmentation region corresponding to a lower frequency than the pore.

In addition, the pore region extractor 211 primarily serves to extract the pore region of the subject's skin by binarizing or monochromatizing the first image from which noise has been removed by a brightness level.

That is, the pore region extractor 211 extracts a region in which the brightness level is equal to or less than a predetermined value from the first image as a pore region.

4 is an exemplary view of a first acquired image output from an image acquisition apparatus processed by the pore region extractor 211. However, FIG. 4 shows only a partial image, not an entire skin image.

5 shows an exemplary view of a designated area for analyzing a pore area from a user by the first designated area input device 212.

The first designated area input device 212 serves to receive a designated area for analyzing a pore area from a user. It is preferable that the first designated area input device 212 designates at least one designated area. For example, the first designated area input device 212 can designate areas of the forehead, nose, and cheeks. In addition, the size of the area for each area is preferably set in advance by the size of the horizontal pixel and the size of the vertical pixel. In addition, it is needless to say that the size of the area for each area may be set equally or differently.

The image preprocessor 213 secondarily removes noise for each region designated by the first designated region inputter 212, and displays an image of the original first acquired image and the second noise-removed image. The role of convergence.

6 shows an exemplary view of an image obtained by combining the first acquired image and the second noise-removed image by the image preprocessor 213.

Secondary noise removal by the image preprocessor 213 may be performed by removing pores having an area of a size less than or equal to a first pixel among the pore regions extracted by the pore region extractor 211. However, in some cases, the image preprocessor 213 may be excluded from the configuration of the skin diagnosis apparatus 200 of the present invention.

The pore feature value calculator 214 uses an image output from the pore region extractor 211 or an image designated by the first designated region inputter 212 or an image processed by the image preprocessor 213. , It serves to calculate the feature value of at least one pore.

That is, the pore feature value calculator 214 serves to calculate feature values of at least one pore for each designated region from an image from which the pore region has been extracted.

The feature values of the at least one pore calculated by the pore feature value calculator 214 are the number of pores calculated for each of the at least one designated region, the average size of the pores, and the total of the corresponding region with respect to the total area of the region. It is preferable to include at least one of the ratio of the area occupied by the pores, the average value of all the pores in the corresponding area of the shape value of the pores, and the average value of all the pores in the corresponding area of the slope value of the pores. The area occupied by all the pores in the region can be calculated by multiplying the number of pores by the average size of the pores. After the pore feature value calculator 214 calculates the shape values of each pore included in the corresponding region, the average value of all the pores in the corresponding region of the shape values for the entire pores in the corresponding region is calculated. Similarly, after calculating the slope value of each pore included in the corresponding area, an average value of the total pores of the corresponding area of the slope value for the total pores of the corresponding area is calculated.

7 is an explanatory diagram for calculating the shape value and the slope value of the pores.

As can be seen from FIG. 7, the longest variable, the longest axis, and the shortest variable, can be searched from the pore area.

The shape value of the pores can be calculated by the ratio of the length of the major axis of the pore and the length of the minor axis. Specifically, the shape value of the pore can be calculated by dividing the length of the major axis of the pore by the length of the minor axis. That is, the shape value of the pores means a value for knowing the degree of flatness of the pores.

In addition, the inclination value of the pores is characterized in that the horizontal axis connected at the point where the long axis meets the edge of the pore region is an angle formed by the long axis.

The first cover index calculator 215 uses the feature values of at least one pore calculated by the pore feature value calculator 214 to calculate a first cover index indicating the concentration of makeup for covering the pore area. Plays a role. Specifically, the first cover index calculator 215 uses a feature value of at least one pore for a pore region having a predetermined size or more to set a first cover index indicating the concentration of makeup for covering the pore area. It is desirable to calculate.

The first cover index calculator 215 includes the number of pores, the average size of the pores, the ratio of the area occupied by all the pores in the area to the total area of the area, and the shape values of the pores for all the pores in the area The first cover index may be calculated using both the average value and the average value of all pores in the corresponding region of the inclination value of the pores. However, the number of pores, the average size of the pores, and the average value of all pores in the corresponding area of the slope value of the pores may be provided to the user only as reference information.

For example, the first cover index is the first value that is the ratio of the area occupied by the entire pores of the corresponding area to the total area of the at least one designated corresponding area multiplied by the average value of the total pores of the corresponding area by the shape value of the pores. The value -1 can be used. However, if necessary, the first cover index may be calculated using the first or second value, which is a value obtained by multiplying the 1-1 value by an appropriate weight. In addition, the first cover index calculator 215 may finally calculate the first cover index by calculating the average value of the 1-1 value or the 1-2 value calculated for each region.

Hereinafter, the second cover index calculation unit 220 will be described.

Specifically, the second cover index calculator 220 includes a dye deposition region extractor 221, a second designated region inputter 222, a dye deposition feature value calculator 223, and a second cover index calculator 224. It is preferably configured to include.

The pigmentation region extractor 221 primarily removes noise from the second acquired image to generate a second image from which the pigmentation region is extracted. The high-pass filter may be used to remove noise by the pigmentation region extractor 221. Alternatively, the removal of noise by the pigmentation region extractor 221 may be implemented by processing a second acquired image with a Gaussian blur using a Gaussian blur method and subtracting it from the original second acquired image. The noise removed by the dye deposition region extractor 221 becomes a component biased toward a higher frequency region than the noise removed by the pore region extractor 211. That is, it is preferable that the high-pass characteristic frequency of the pore region extractor 211 is lower than the high-pass characteristic frequency of the pigmentation region extractor 221. Since the second acquired image is an image by cross-polarization, a large portion of the pore region having a higher frequency than the pigmentation region is removed, so the pigmentation region extractor 221 efficiently noises by image processing for high pass Will be able to handle it.

8 is an exemplary view in which the second acquired image output from the image acquisition apparatus is processed by the pigmentation region extractor 221.

9 shows an exemplary view of a designated area to analyze a pigmentation area from a user by the second designated area input device 222.

The second designated area input unit 222 serves to receive a designated area for analyzing a pigmentation area from a user. It is preferable that the second designated area input device 222 designates at least one designated area. For example, the second designated area input device 222 may designate areas of both cheekbones. In addition, the size of the area for each area is preferably set in advance by the size of the horizontal pixel and the size of the vertical pixel. In addition, it is needless to say that the size of the area for each area may be set equally or differently.

The pigmentation feature value calculator 223, for each of at least one designated region of the image from which the pigmentation region is extracted by the pigmentation region extractor 221 specified by the second designated region inputter 222, is at least one It serves to calculate the characteristic values of pigmentation.

At least one characteristic value of the pigmentation calculated by the pigmentation feature value calculator 223 is, from each image for at least one designated region, an average value of a red pixel value and an average value of a green pixel value in an RGB domain. Or, it is preferable that it is at least one of the average values of the blue pixel values.

However, in general, the pigmented region tends to exhibit a low blue pixel value, and it is more preferable to use a blue pixel value in the RGB domain as a pigmented feature value from each image for at least one designated region.

In addition, the second cover index calculator 224 uses a feature value of at least one pigmentation calculated by the pigmentation feature value calculator 223 to display a concentration of makeup for the cover of the pigmentation site. 2 It serves to calculate the cover index. Specifically, as the second cover index, it is preferable to use a second value obtained by subtracting the average value of the blue pixel values of the image of the corresponding region calculated from the maximum blue pixel values that can be represented in the RGB domain. For example, for each 256-gradation image for each RGB, the second value may be calculated by (an average value of 255-calculated blue pixel values). That is, the pigmentation region tends to exhibit a low blue pixel value, so that the second value is inversely proportional to the average value of the blue pixel values, by subtracting the average value of the blue pixel values calculated from the maximum blue pixel values. Was calculated. In addition, the second cover index calculator 224 may finally calculate the second cover index by calculating an average value of the second values calculated for each region.

Hereinafter, a method for diagnosing skin using pores and pigmentation of skin according to a preferred embodiment of the present invention will be described. However, since the skin diagnosis method of the present invention uses the skin diagnosis system 1000 described above, it is needless to say that it includes all the features of the skin diagnosis system 1000 even if there is no separate description.

Specifically, the skin diagnosis method of the present invention is characterized in that it is performed by a processor in the skin diagnosis device 200 which is a computing device.

The skin diagnosis method of the present invention includes a first cover index calculation step (S210) by the first cover index calculation unit 210 and a second cover index calculation step (S220) by the second cover index calculation unit 220 and And a final makeup concentration recommendation step S230 by the final makeup concentration recommendation unit 230.

The first cover index calculation step (S210) serves to calculate a first cover index indicating the concentration of makeup for covering the pore area using the first acquired image. In addition, the second cover index calculation step (S220) serves to calculate a second cover index indicating the concentration of makeup for the cover of the pigmentation site using the second acquired image.

In addition, the final makeup concentration recommendation step (S230) is the final makeup concentration using the first cover index and the second cover index calculated from the first cover index calculation step (S210) and the second cover index calculation step (S220), respectively. It serves to recommend.

Specifically, in the final makeup concentration recommendation step (S230), the concentration of the makeup is set in multiple stages according to the value of the first cover index in relation to the pores. In addition, in the final makeup concentration recommendation step (S230), the concentration of makeup is set in multiple stages according to the value of the second cover index in relation to pigmentation. Finally, in the final makeup concentration recommendation step (S230), among the concentration stage of makeup by the first cover index and the concentration stage of makeup by the second cover index, a step representing a concentration of darker makeup is selected and presented to the user. It is recommended to use the appropriate makeup level. This is because light makeup is required for the pores, and if the pigmentation is severe, it is necessary to wear dark makeup. If pigmentation is required, the thin makeup is required.

Specifically, the first cover index calculation step (S210), the pore area extraction step (S211) by the pore area extractor 211, the first designated area input step (S212) by the first designated area inputter 212, The image pre-processing step S213 by the image pre-processor 213, the pore feature value calculating step S214 by the pore feature value calculator 214, and the first cover index by the first cover index calculator 215 It is preferably configured to include the calculation step (S215).

In the pore region extraction step S211, noise is primarily removed from the first acquired image to generate a first image. The high-pass filter may be used to remove noise by the pore region extraction step (S211). Alternatively, the removal of noise by the pore region extraction step (S211) may be implemented by processing the first acquired image with a Gaussian blur using the Gaussian blur method and subtracting it from the original first acquired image. The noise removed by the pore region extraction step S211 may also include a pigmentation region corresponding to a lower frequency than the pore.

In addition, the pore region extraction step (S211) serves to extract the pore region of the subject's skin by binarizing or monochromatizing the first image from which noise is primarily removed by a brightness level.

That is, in the step of extracting the pore area (S211 ), an area having a brightness level equal to or less than a predetermined value is extracted from the first image as the pore area.

The first designated area input step (S212) serves to receive a designated area to analyze the pore area from the user. In the first designated area input step (S212), it is preferable to designate at least one designated area. For example, in the first designated area input step (S212 ), areas of the forehead, nose, and cheeks can be designated. In addition, the size of the area for each area is preferably set in advance by the size of the horizontal pixel and the size of the vertical pixel.

In the image pre-processing step (S213 ), for each region designated by the first designated region input step (S212 ), noise is secondarily removed, and the original first acquired image and the secondly removed image are combined. Plays a role.

The secondary noise removal by the image pre-processing step S213 may be performed by removing pores having an area less than or equal to a first pixel size among the pore areas extracted by the pore area extraction step S211. However, in some cases, the image pre-processing step S213 may be excluded from the configuration of the skin diagnosis apparatus 200 of the present invention.

In the step of calculating the pore feature value (S214 ), the designated image by the first designated region input step (S212) or the image after being processed by the image preprocessing step (S213) among the images output from the pore region extraction step (S211 ). By using, it serves to calculate the feature value of at least one pore.

That is, in the step of calculating the pore feature value (S214 ), the feature value of at least one pore is calculated for at least one designated region from the image from which the pore region is extracted.

The feature values of the at least one pore calculated in the pore feature value calculation step (S214) are the number of pores calculated for each of the at least one designated region, the average size of the pores, and the entirety of the region with respect to the total area of the region. It is preferable to include at least one of the ratio of the area occupied by the pores, the average value of all the pores in the corresponding area of the shape value of the pores, and the average value of all the pores in the corresponding area of the slope value of the pores. The area occupied by all the pores in the region can be calculated by multiplying the number of pores by the average size of the pores. After calculating the shape values of the pores included in the corresponding region in the step of calculating the pore feature value (S214 ), the average value of the total pores of the corresponding region of the shape value for the total pores of the corresponding region is calculated. Similarly, after calculating the slope value of each pore included in the corresponding area, an average value of the total pores of the corresponding area of the slope value for the total pores of the corresponding area is calculated.

In the step of calculating the pore feature value (S214 ), the longest side, the longest side, and the shortest side, the shortest side, across the pore region of the corresponding pore may be searched.

The shape value of the pores can be calculated by the ratio of the length of the major axis of the pore to the length of the minor axis. Specifically, the shape value of the pore can be calculated by dividing the length of the major axis of the pore by the length of the minor axis. In addition, the inclination value of the pores is characterized in that the horizontal axis connected at the point where the long axis meets the edge of the pore region is an angle formed by the long axis.

The first cover index calculation step (S215) uses the feature values of at least one pore calculated in the pore feature value calculation step (S214) to calculate a first cover index indicating the concentration of makeup for covering the pore area. Plays a role. In addition, the first cover index calculation step (S215), using a feature value of at least one pore for a pore area of a predetermined size or more, calculates a first cover index indicating the concentration of makeup for covering the pore area It is desirable to do.

In the first cover index calculation step (S215), the number of pores, the average size of the pores, the ratio of the area occupied by all the pores in the region to the total area of the region, the shape value of the pores for the entire pores in the region The first cover index may be calculated using both the average value and the average value of all pores in the corresponding region of the inclination value of the pores. However, the number of pores, the average size of the pores, and the average value of all pores in the corresponding area of the slope value of the pores may be provided to the user only as reference information.

For example, the first cover index is the ratio of the area occupied by the total pores of the corresponding area to the total area of the corresponding area calculated for each of the at least one designated area, relative to the total pores of the corresponding area of the shape value of the pores The 1-1 value which is the product of the average value can be used. However, if necessary, the first cover index may be calculated using the first or second value, which is a value obtained by multiplying the 1-1 value by an appropriate weight. In addition, the first cover index calculation step (S215) may finally calculate the first cover index by calculating the average value of the 1-1 value or the 1-2 value calculated for each region.

The second cover index calculation step (S220) will be described below.

Specifically, in the second cover index calculation step (S220), the dye deposition area extraction step (S221) by the dye deposition area extractor 221 and the second designated area input step (S222) by the second designated area input device 222 (S222). ), it is preferably configured to include a dye deposition feature value calculating step 223 by the pigmentation feature value calculator 223 and a second cover index calculating step S224 by the second cover index calculator 224. Do.

In the pigmentation region extraction step (S221 ), noise is primarily removed from the second acquired image to generate a second image from which the pigmentation region is extracted. The high-pass filter may be used to remove noise by the pigmentation region extraction step (S221 ). Alternatively, the removal of noise by the pigmentation region extraction step (S221) may be implemented by processing a second acquired image with a Gaussian blur using a Gaussian blur method and subtracting it from the original second acquired image. The noise removed by the dye deposition region extraction step (S221) becomes a component biased toward a higher frequency region than the noise removed by the pore region extraction step (S211 ). That is, it is preferable that the characteristic frequency of the high pass in the pore region extraction step (S211) is lower than the characteristic frequency of the high pass in the pigmentation region extraction step (S221). Since the second acquired image is an image by cross-polarization, since a large number of pore regions having a higher frequency than the pigmentation region are removed, the pigmentation region extraction step (S221) is efficiently performed by image processing for high pass. It can handle noise.

The second designated area input step (S222) serves to receive a designated area to analyze the pigmentation area from the user. In the second designated area input step (S222), it is preferable to designate at least one designated area. For example, in the second designated area input step (S222 ), areas of both cheekbones may be designated. In addition, the size of the area for each area is preferably set in advance by the size of the horizontal pixel and the size of the vertical pixel.

The dye deposition feature value calculation step (S223) is performed for each of at least one designated region of the image in which the dye deposition region is extracted by the dye deposition region extraction step (S221) designated by the second designated region input step (S222). , It serves to calculate the characteristic value of at least one pigmentation.

The feature value of the at least one pigmentation calculated in the step of calculating the pigmentation feature value (S223) is, from each image for at least one designated region, an average value of a red pixel value and an average value of a green pixel value in an RGB domain. Or, it is preferable that it is at least one of the average values of the blue pixel values.

However, in general, the pigmented region tends to exhibit a low blue pixel value, and it is more preferable to use a blue pixel value in the RGB domain as a pigmented feature value from each image for at least one designated region.

In addition, the second cover index calculation step (S224) is, by using the feature value of at least one pigmentation calculated in the pigmentation feature value calculation step (S223), the agent showing the concentration of makeup for the cover of the pigmentation site 2 It serves to calculate the cover index. Specifically, as the second cover index, it is preferable to use a second value obtained by subtracting the average value of the blue pixel values calculated from the maximum blue pixel values that can be represented in the RGB domain. For example, for each 256-gradation image for each RGB, the second value may be calculated by (an average value of 255-calculated blue pixel values). That is, the pigmentation region tends to exhibit a low blue pixel value, so that the second value is inversely proportional to the average value of the blue pixel values, by subtracting the average value of the blue pixel values calculated from the maximum blue pixel values. Was calculated. In addition, the second cover index calculation step S224 may finally calculate the second cover index by calculating an average value of the second values calculated for each region.

As described above, according to the skin diagnosis system 1000 of the present invention and its diagnostic method, the state of the pores and the state of pigmentation are indexed into the first cover index and the second cover index, respectively, so that the user can easily obtain the skin condition information. It can be seen that not only can be grasped, but also the state of the pores and the state of pigmentation can be properly fused and expressed as recommended makeup concentration information using one final cover index.

1000: skin diagnosis system
100: skin image acquisition device 200: skin diagnostic device
110: light source 120: polarization unit
130: image acquisition unit 210: first cover index calculation unit
220: second cover index calculation unit 230: final makeup concentration recommendation unit
121: first polarizer 122: second polarizer
211: pore area extractor 212: first designated area input device
213: image preprocessor 214: pore feature value calculator
215: first cover index calculator 221: pigmentation area extractor
222: second designated area input device 223: pigmentation feature value calculator
224: second cover index calculator

Claims (24)

In the skin diagnosis method,
A pore region extraction step of extracting a pore region of the subject's skin; And
And a pore feature value calculating step of calculating a feature value of at least one pore for each of at least one designated region of the image from which the pore region is extracted in the pore region extraction step. According to claim 1,
The feature value of the at least one pore calculated in the step of calculating the feature value of the pore,
And a ratio of an area occupied by the total pores of the corresponding area to the total area of the corresponding area. According to claim 2,
The feature value of the at least one pore calculated in the step of calculating the feature value of the pore,
The method of diagnosing skin, further comprising an average value of all pores in a corresponding area of the shape value of the pores. According to claim 3,
The shape value of the pores,
The method of diagnosing skin, characterized in that the length of the longest axis, which is the longest side that crosses the pore area of the corresponding pore, is divided by the length of the shortest, shortest side. According to claim 1,
The skin diagnostic method,
Further comprising; a first cover index calculation step of calculating a first cover index indicating the concentration of makeup for the cover of the pore area, by using the feature values of at least one pore calculated in the pore feature value calculation step Skin diagnostic method characterized by. The method of claim 5,
The first cover index,
The 1-1 value, which is the ratio of the area occupied by all the pores in the area to the total area of the area calculated for each of the at least one designated area, is the product of the pore shape value multiplied by the average value of all the pores in the area. Skin diagnostic method characterized by using. According to claim 1,
The pore region extraction step,
An environment in which a polarizing plate is inserted at a rear end of the light source to irradiate polarized light to a subject, and an image that is not polarized is captured because a polarizing plate is not inserted at the front end of the camera; or,
An environment in which no polarizing plate is inserted between the light source and the camera; a method for diagnosing skin, characterized by extracting a pore area using an image captured in the image. According to claim 1,
The skin diagnostic method,
A pigmentation region extraction step of extracting a pigmentation region of a subject's skin; And
It characterized in that it further comprises; for each of the at least one designated region of the image in which the pigmentation region is extracted in the pigmentation region extraction step, calculating a feature value of at least one pigmentation; Skin diagnostic methods. The method of claim 8,
At least one of the feature values of the pigmentation calculated in the step of calculating the feature of pigmentation,
Skin diagnosis method characterized by using the average value of the blue pixel value in the RGB domain of the region. The method of claim 9,
The skin diagnostic method,
A second cover index calculation step of calculating a second cover index indicating the concentration of makeup for the cover of the pigmentation site by using at least one feature value of the pigmentation calculated in the pigmentation feature value calculation step; further Skin diagnostic method comprising a. The method of claim 10,
The second cover index,
A skin diagnosis method characterized by using a second value obtained by subtracting the average value of the blue pixel values of the corresponding region from the maximum blue pixel value that can be represented in the RGB domain. The method of claim 8,
The pigmentation region extraction step,
The first polarizing plate is inserted at the rear end of the light source to irradiate the polarized light to the subject, and the dye deposition region is extracted by using the captured image in an environment in which the second polarizing plate is inserted at the front end of the camera to capture the polarized image.
The axis of the first polarizing plate and the axis of the second polarizing plate, the skin diagnostic method, characterized in that forming a right angle. In the skin diagnostic system,
Includes a skin diagnostic device for diagnosing a skin condition using the skin image of the subject;
The skin diagnostic device,
A pore region extractor for extracting the pore region of the subject's skin; And
And a pore feature value calculator that calculates a feature value of at least one pore for each of the at least one designated region of the image from which the pore region has been extracted from the pore region extractor. The method of claim 13,
The feature value of at least one pore calculated by the pore feature value calculator is
Skin diagnostic system comprising a ratio of the area occupied by the total pores of the area to the total area of the area. The method of claim 14,
The feature value of at least one pore calculated by the pore feature value calculator is
Skin diagnostic system, characterized in that it further comprises the average value of the total pores of the area of the shape of the pores. The method of claim 15,
The shape value of the pores,
A skin diagnosis system characterized by dividing the length of the longest axis, the longest side that crosses the pore area of the pore, by the length of the shortest, shorter side. The method of claim 13,
The skin diagnostic device,
Further comprising; a first cover index calculator for calculating a first cover index indicating the concentration of makeup for the cover of the pore area by using the feature values of at least one pore calculated by the pore feature value calculator Skin diagnostic system characterized by. The method of claim 17,
The first cover index,
A skin diagnosis system characterized by using a 1-1 value, which is a value obtained by multiplying the ratio of the area occupied by all the pores in the area to the total area of the area by the average value of all the pores in the area by the shape value of the pores. The method of claim 13,
The pore region extractor,
An environment in which a polarizing plate is inserted at a rear end of the light source to irradiate polarized light to a subject, and an image that is not polarized is captured because a polarizing plate is not inserted at the front end of the camera; or,
An environment in which no polarizing plate is inserted between the light source and the camera; a skin diagnosis system characterized by extracting a pore area using an image captured in the image. The method of claim 13,
The skin diagnostic device,
A pigmentation region extractor for extracting a pigmentation region of a subject's skin; And
Skin characterized in that it further comprises; for each of the at least one designated region of the image where the pigmentation region is extracted from the pigmentation region extractor, a pigmentation feature value calculator that calculates a feature value of at least one pigmentation; Diagnostic system. The method of claim 20,
The feature value of at least one pigmentation calculated by the pigmentation feature value calculator is:
A skin diagnosis system characterized by using an average value of blue pixel values in the RGB domain of the corresponding area. The method of claim 21,
The skin diagnostic device,
A second cover index calculator that calculates a second cover index indicating the concentration of makeup for the cover of the pigmentation site by using at least one feature value of the pigmentation calculated by the pigmentation feature value calculator; Skin diagnostic system comprising a. The method of claim 22,
The second cover index,
A skin diagnosis system characterized by using a second value obtained by subtracting the average value of the blue pixel values of the corresponding area from the maximum blue pixel value that can be represented in the RGB domain. The method of claim 20,
The pigmentation region extractor,
The first polarizing plate is inserted at the rear end of the light source to irradiate the polarized light to the subject, and the dye deposition region is extracted by using the captured image in an environment in which the second polarizing plate is inserted at the front end of the camera to capture the polarized image.
The axis of the first polarizing plate and the axis of the second polarizing plate, the skin diagnostic system, characterized in that forming a right angle.

Images