KR102645445B1 - 2D skin image analysis system and analysis method using thereof - Google Patents

Abstract

The present invention relates to a two-dimensional skin image analysis system and analysis method. The two-dimensional skin image analysis system according to an embodiment of the present invention includes a light source that irradiates light to the skin from the upper direction of the skin, and a two-dimensional skin image. An image measurement unit including an image measurement unit that measures; and an image processing unit including an image extraction unit that extracts a preset color from the measured image to generate an extracted image, and an image binarization unit that binarizes the extracted image.

Description

2D skin image analysis system and analysis method {2D skin image analysis system and analysis method using the same}

The present invention relates to a two-dimensional skin image analysis system and analysis method, and to a two-dimensional skin image analysis system and analysis method that can accurately measure and analyze skin characteristics.

Pores and wrinkles on the skin are the main signs of aging. Accordingly, various methods are being attempted to evaluate skin characteristics such as pores and wrinkles.

Conventionally, a method using 3D images and a method using 2D images were used to evaluate pores.

In the case of a method using 3D images, the depth of pores can be measured, but the shape, distribution, and area of the pores cannot be calculated, and it has the disadvantage of not being able to be measured without expensive special equipment. Additionally, it is difficult to distinguish between pores and skin roughness, and pores vary depending on the focus, so the equipment had to be attached closely to the subject's face in order to focus accurately. However, there was a problem that errors occurred because the subject's pores and wrinkles were pressed and deformed as the equipment adhered closely.

In the case of a method using a 2D image, such as in Patent Document 1, it is used that the pore area is darker than other areas due to blackheads. When measured with this method, skin pigmentation other than the pore, shadow due to facial curvature, and facial finesse are used. The disadvantage is that grain structure, etc. are included in the results, which causes errors, and it is difficult to accurately analyze pores. In addition, even in the case of 2D images, pore deviations appear depending on the focus, and when the equipment is brought into contact with the subject's skin, the shape of the pores or wrinkles is deformed, causing errors.

Meanwhile, Figure 1 is a picture showing a pore analysis method using a 2D facial image according to the prior art. Referring to (a) of FIG. 1, a two-dimensional face image is measured and a region of interest (ROI) (A ₁ ) is set. Referring to (b) of FIG. 1, the image of the region of interest (A ₁ ) is converted to gray scale. Referring to (c) of FIG. 1, color information is removed from the image converted to gray scale and only the shaded area remains. In this case, a problem occurred in which the image was lost due to the curve of the face, as shown in the lower right corner.

In other words, in the case of conventional 3D image analysis, the depth of pores could be measured, but the shape or distribution area could not be calculated, and measurement was difficult with expensive and bulky equipment. In addition, in the case of 2D image analysis, shadows due to facial curvature were included in the results, making accurate analysis difficult. In addition, both 3D and 2D images have errors depending on focus, so they are measured by placing them in close contact with the face. In this case, there is a problem in that the shape of pores or wrinkles is deformed, which causes errors.

J.P. 1995-231883 A

The present invention is intended to solve the above-mentioned problems, and seeks to provide a two-dimensional skin image analysis system and analysis method that can accurately measure and analyze skin features without expensive special equipment.

In addition, we aim to provide a two-dimensional skin image analysis system and analysis method that can analyze pore features by minimizing errors caused by facial curves, fine wrinkles, etc.

In addition, we aim to provide a two-dimensional skin image analysis system and analysis method that can analyze pore features in a variety of ways by analyzing the angle, number, distribution, and elongation length of pores.

In addition, we aim to provide a two-dimensional skin image analysis system and analysis method that can minimize errors due to focus through focus correction and analyze skin characteristics without attaching the measurement device to the face.

The present invention is intended to solve the above-mentioned problems, and includes an image measurement unit including a light source that irradiates light to the skin from the upper direction of the skin and an image measurement unit that measures a two-dimensional skin image; and an image processing unit including an image extraction unit that extracts a preset color from the measured image to generate an extracted image, and an image binarization unit that binarizes the extracted image.

The image measuring unit may be arranged to face the target skin, and the light source may be arranged upward with respect to an axis connecting the skin and the image measuring unit.

The target skin is facial skin, and the light source may be arranged in a range of 0° to 90° to the left or right around the axis in the direction of the crown of the head.

The target skin is facial skin, and the light source may be disposed within a range of 0° to 90° in the frontal direction around the crown direction axis of the head.

The image processing unit may further include a pre-processing unit that pre-processes the measured image by applying a filter.

The filter may be a high-pass filter.

The preset color may be blue.

It may further include a skin analysis unit that analyzes skin features in the binarized image.

The skin analysis unit may analyze one or more of pore elongation length, pore area, pore distribution, pore angle, and pore number.

The image measurement unit is an integrated image measurement device consisting of the light source, the image measurement unit, a fixture that fixes the subject's position with respect to the image measurement unit, and a housing including the light source, the image measurement unit, and the fixture therein. It can be.

A two-dimensional skin image analysis method according to another embodiment of the present invention includes an image measurement step of measuring a two-dimensional skin image by irradiating light to the skin from the upper direction of the skin; An image extraction step of generating an extracted image by extracting a preset color from the measured image; and an image binarization step of binarizing the extracted image.

In the image extraction step, the blue color can be extracted to create a blue extracted image.

In the image extraction step, a binarized image can be generated by increasing the contrast ratio in the blue extracted image.

The method may further include an image analysis step of analyzing skin features in the binarized image.

In the image analysis step, it may include analyzing pores in the binarized image to calculate pore characteristic data consisting of one or more of pore elongation length, pore area, pore distribution, pore angle, and pore number.

The image analysis step includes collecting and storing pore images or pore characteristic data binned by race or age; And it may further include a step of estimating the subject's race or age group by comparing the binarized image or pore characteristic data of the measured skin with the binarized pore image or pore characteristic data by race or age.

The image analysis step includes collecting and storing binarized pore images or pore characteristic data according to the usage time or usage amount of the test cosmetic; And it may further include calculating the degree of influence of the pores of the test cosmetic based on changes in the binarized pore image or pore characteristic data according to the usage time or amount of the test cosmetic.

According to one embodiment of the present invention, a two-dimensional skin image analysis system and analysis method that can accurately measure and analyze skin features without expensive special equipment can be provided.

In addition, it is possible to provide a two-dimensional skin image analysis system and analysis method that can analyze pore features by minimizing errors due to facial curvature, fine wrinkles, etc.

In addition, it is possible to provide a two-dimensional skin image analysis system and analysis method that can analyze pore features from various angles by analyzing the angle, number, distribution, elongation length, etc. of pores.

In addition, it is possible to provide a two-dimensional skin image analysis system and analysis method that can minimize errors due to focus through focus correction and analyze skin characteristics without being placed in close contact with the face.

In addition, pore characteristic data by race or age can be collected and stored to build a database, and can be used to estimate the subject's race or age based on the pore characteristic data measured therefrom.

Additionally, it can be used to analyze the effect of a product on pores through pore images and quantitative pore characteristic data based on them. In other words, it can be used to determine the efficacy of a pore-related product through changes in the pore characteristic data of subjects using the product.

Figure 1 is a photograph showing pore images obtained by a pore analysis method according to the prior art.
Figure 2 is a diagram schematically showing a two-dimensional skin image analysis device according to an embodiment of the present invention.
Figure 3 is a side view schematically showing the arrangement of an image measurement unit according to an embodiment of the present invention.
Figure 4 is a front view schematically showing the arrangement of an image measurement unit according to an embodiment of the present invention.
Figure 5 is a flowchart schematically showing a two-dimensional skin image analysis method according to an embodiment of the present invention.
Figure 6 is a flowchart schematically showing a two-dimensional skin image analysis method according to another embodiment of the present invention.
Figure 7 is a photograph showing pore images obtained by a two-dimensional skin image analysis device according to an embodiment of the present invention.
Figure 8 shows pore images for each age measured according to an embodiment of the present invention and pore images on which image processing was performed on the measured images.
Figure 9 is a graph showing the distribution of pore area by age group measured according to the prior art and an embodiment of the present invention.
Figure 10 is an enlarged view of the processed images of Figure 8, showing them by age.
Figure 11 is a graph showing the average pore elongation rate of subjects by age.
Figure 12 is an image of measured pores of subjects in their 50s and 60s.

Since these embodiments can be subjected to various transformations and have various embodiments, specific embodiments will be described in detail by illustrating them in the drawings. However, this is not intended to limit the scope to specific embodiments, and should be understood to include all transformations, equivalents, and substitutes included in the disclosed spirit and technical scope. In describing the embodiments, if it is determined that detailed description of related known technologies may obscure the point, the detailed description will be omitted.

In an embodiment, a 'module' or 'unit' performs at least one function or operation, and may be implemented as hardware or software, or as a combination of hardware and software. In addition, a plurality of 'modules' or a plurality of 'units' are integrated into at least one module and implemented by at least one processor (not shown), except for 'modules' or 'units' that need to be implemented with specific hardware. It can be.

Hereinafter, the embodiment will be described in detail with reference to the accompanying drawings. In the description with reference to the accompanying drawings, identical or corresponding components will be assigned the same drawing numbers and overlapping descriptions thereof will be omitted.

Figure 2 is a diagram schematically showing a two-dimensional skin image analysis device according to an embodiment of the present invention.

Referring to FIG. 2, the two-dimensional skin image analysis system according to an embodiment of the present invention includes an image measurement unit 1 that measures a two-dimensional skin image and an image processing unit 3 that processes the measured image. do.

The image measurement unit 1 includes a light source 11 that irradiates light to the skin from the upper direction of the skin, and an image measurement unit 13 that measures a two-dimensional skin image.

As used herein, skin features refer to areas where irregularities are formed and there is a height difference with respect to other normal skin surfaces, including pores, wrinkles, scars, etc. According to one embodiment, the skin features can be facial pores.

The light source 11 is configured to induce shadows on features of the skin. It is arranged to induce undistorted shadows on skin features. For example, pores have irregularities, which can cause shadows due to light.

The image measurement unit 13 is a device for measuring skin images, and various image measurement devices such as cameras that can be used in the present technical field can be applied to it.

FIG. 3 is a side view schematically showing the arrangement of an image measurement unit according to an embodiment of the present invention, and FIG. 4 is a front view schematically showing the arrangement of an image measurement unit according to an embodiment of the present invention.

Referring to FIG. 3, the image measuring unit 13 is arranged to face the target skin (i.e., subject (X)), and the light source 11 connects the skin and the image measuring unit 13. It can be arranged in an upward direction with respect to the axis.

When the light source 11 is irradiated from the front, it is difficult for a shadow to appear on the skin's features, so it is difficult to distinguish features such as pores even if a two-dimensional image is measured. Accordingly, the light source 11 may irradiate the subject (X) from one side.

Additionally, when the light source 11 is irradiated from one side, the shadow of a feature may be tilted or distorted. Therefore, according to one embodiment, the image measuring unit 13 is arranged to face the subject (X), and the light source 11 is directed upward with respect to the axis connecting the image measuring unit 13 and the subject (X) ( That is, it may be disposed in a direction opposite to the ground or in a direction toward the top of the subject's head (X) and irradiated with light (X).

More specifically, when the target skin is facial skin, the light source 11 may be disposed within a range of 0° to 90° in the frontal direction around the crown direction axis (X ₁ ) of the head. To measure facial skin, it can be arranged to have a value of 90° or less, because if it exceeds 90°, the shadow will be distorted by other parts of the face.

Referring to FIG. 3, the light source 11 may be arranged at a preset first angle α in the frontal direction, which is the direction toward the nose, with the crown axis (X ₁ ) of the head of the subject (X) as the center. there is. Figure 3 shows an arrangement for inducing an undistorted shadow, especially in the pore area, and the first angle α is approximately 45°.

Additionally, when the target skin is facial skin, the light source 11 may be disposed within a range of 0° to 90° to the left or right around the axis (X ₂ ) in the crown direction of the head.

In particular, when applied to facial skin, the nose is the highest due to the nature of the facial contour, so if it is excessively biased to the left or right of the nose, that is, if it exceeds 90°, the feature may be obscured by the shadow caused by the nose. Alternatively, the shadow of a feature, such as a pore, may be tilted and the length component, such as the stretching length, may be distorted and measured inaccurately. Accordingly, the light source 11 may be arranged within a range of 90° or less to the left or right around the crown direction axis (X ₂ ) of the face.

Specifically, referring to FIG. 4, the light source ₁₁ is set at a second angle ( β) can be arranged. Figure 4 shows an arrangement for inducing an undistorted shadow, especially in the pore area, where the light source 11a shows an embodiment in which the second angle β is approximately 0°, and the light source 11b is angled to the right at a predetermined angle other than 0°. It indicates that it is placed at an angle of . However, it is not necessarily limited to this and of course it can be placed on the left. Additionally, the light source 11 may be arranged to have a preset first angle (α) and a second angle (β) in both the front direction component and the left/right direction. That is, in the case of an embodiment for measuring the pore area, the first angle α may be arranged to be approximately 45° and the second angle β may be approximately 0° to optimize the shadow of the pore area.

According to the above arrangement, it is possible to induce a shadow with a shape and shape similar to the feature while minimizing the shadows that may be caused by the skin, and specifically other structures that may form on the face.

In this specification, the two-dimensional skin image measurement system and method is described with a focus on facial skin, especially pores, but is not necessarily limited thereto, and may be other areas other than the face, and may include wrinkles, scars, etc. other than pores. Of course, it can be used to analyze the characteristics of .

Depending on the structural characteristics of each area, the angle of the light source may be adjusted to radiate light to the target area from the upper direction.

According to one embodiment of the present invention, the image measurement unit 1 includes the light source 11, the image measurement unit 13, and a fixture capable of fixing the position of the subject, and the light source 11 and the image It may be an integrated image measurement device consisting of a housing that includes a measuring unit 13 and a fixing unit (not shown) inside. By including the fixing part, the subject's face can be fixed and errors can be removed by controlling the light source. As an example, the image measurement device may be VISIA-CR ^® manufactured by Canfield (Fairfield, NJ, USA). However, it is not necessarily limited to this, and of course, the light source 11, the image measurement unit 13, and the fixing part may be configured as a separate image measurement device that can be assembled and disassembled.

According to one embodiment, when measuring the pore area using VISIA-CR ^® , only the upper light source has an aperture value of F5.6+1 to F5.6+9, F8.0+0 in standard 1 mode or 2 mode. You can measure in the range from F8.0+9 to F11.0+0 to F11.0+9 with the left, right, and bottom light sources turned on and blocked. Preferably, the clearest pore image can be obtained when a light source in the range of F11.0+0 to F11.0+9 is turned on.

Referring again to FIG. 2, the image processing unit 3 generates an image that eliminates shadows due to skin color elements and skin curvatures and enhances features through image processing.

More specifically, the image processing unit 3 includes an image extraction unit 31 that extracts a preset color from the measured image to generate an extracted image, and an image binarization unit 33 that binarizes the extracted image. do.

The image extraction unit 31 serves to remove shadows caused by skin color elements and skin curves. Specifically, areas caused by pigmentation can be removed from a two-dimensional skin image, and shadows caused by skin curvature can be removed.

In particular, when applied to measuring facial pores, it is possible to remove areas caused by pigmentation such as freckles or spots formed on the face and shadows caused by skin curves such as nasolabial folds.

The image extraction unit 31 generates an extracted image by extracting a preset color to extract only the image of the skin feature from the measured two-dimensional skin image. As an example, when measuring to analyze the pores of the face, blue color can be extracted from the two-dimensional skin image to generate a blue extracted image. In this case, the difference in pore shade can be maximized through the blue extraction image.

It is not necessarily limited to this, and when red color is extracted from a two-dimensional skin image to generate a red color extraction image, the difference in shades of sebum formed on the skin can be maximized. In other words, it can be applied to analyze the distribution and area of sebum on the skin using the red color extracted image.

Additionally, when green is extracted from a two-dimensional skin image to create a green extracted image, the difference in shades of whiteheads formed on the skin can be maximized. In other words, it can be applied to analyze the distribution and area of white heads on the skin using the green extracted image.

By utilizing various color extraction images, the shaded image of desired skin features can be maximized. In other words, by removing skin curves or pigmentation areas from a two-dimensional skin image, only the desired features can be distinguished and analyzed.

Additionally, the image binarization unit 33 can binarize the extracted image and analyze features based on the binarized data. Additionally, the image binary unit 33 can adjust the contrast of the feature within a preset value range to remove facial curves or fine wrinkles and make only pores appear.

According to one embodiment, when analyzing facial pores, the contrast ratio may be set to a value before the pigmentation area appears. Accordingly, the pigmentation area can be removed and a binary image showing only pores can be created.

According to one embodiment of the present invention, the image processing unit 3 may further include a pre-processing unit (not shown) that pre-processes the measured image by applying a filter. The preprocessor may be used for focus correction.

The pre-processing unit may correct focus by applying a filter to the image measured by the image measuring unit 13 before extracting the color in the image extracting unit 31. More specifically, in the case of face images, the protruding nose portion is focused. However, if you want to observe pores on both cheeks, focus correction must be performed. Therefore, a filter can be applied to the measured skin image so that the desired area is focused.

According to one embodiment, the filter may be a high-pass filter, and the clarity of the final image formed can be increased by focusing on both cheeks of the face.

According to one embodiment of the present invention, it may further include a skin analysis unit 5 (also referred to as an “image analysis unit”) that analyzes skin features in the binarized image. The skin analysis unit 5 can calculate the distribution, number, area, etc. of features through the derived binarized skin image.

According to one embodiment, the skin analysis unit 5 may calculate one or more values of pore elongation length, pore area, pore distribution, pore angle, and pore number. For example, if the binarized skin image is a black and white image in which pores are expressed in white, the total pore area can be calculated by adding the areas of the white parts, and the elongation rate of the pore is calculated by dividing the longest diameter in each pore by the shortest length. (elongation) can be calculated, and in addition, pore angles, pore distribution, etc. can be analyzed. In other words, it is possible to solve the shortcoming of not being able to intuitively represent changes in pores and ensure objectivity in the analysis by expressing the characteristics of pores in numerical values.

Furthermore, according to the present invention, pore characteristic data by race or age (e.g., stretching length, pore area, pore distribution, pore angle, number, etc.) is collected and stored, and based on the pore characteristic data measured therefrom. It can be used to estimate the subject's race or age group. Additionally, in the skin analysis unit 5, it can be used to analyze the effect of a product on pores through pore images and quantitative pore characteristic data based thereon. there is. In other words, it can be used to determine the efficacy of a pore-related product through changes in the pore characteristic data of subjects using the product.

The image processing unit 3 and the image analysis unit 5 may be composed of processing software or hardware, or a combination of software and hardware. According to one embodiment, the image processing unit 3 and the image analysis unit It can be combined with (5) or formed to exchange data directly or indirectly.

For example, the image processing unit 3 may be Image-Pro Premier 9.2 from Media Cybernetics (Silver Spring, MD, USA), and the pore area in the skin image measured through the image processing unit 3 is selected as a region of interest. You can perform preprocessing by specifying and applying a high-pass filter. The high-pass filter can have an intensity and pass value specified in the range between 1 and 100. After applying the filter, it extracts only the blue color to create a blue extracted image, and extracts a threshold from 0 to 130 in the extracted image excluding skin color. ) can be specified to rationalize the image by leaving only the pores white and removing the rest.

Figure 5 is a flowchart schematically showing a two-dimensional skin image analysis method according to an embodiment of the present invention.

The two-dimensional skin image analysis method can be performed using the skin image analysis system described above, and of course, the contents of the skin image analysis system described above can also be applied to the analysis method.

A two-dimensional skin image analysis method according to an embodiment of the present invention includes an image measurement step (S1) of measuring a two-dimensional skin image by irradiating light to the skin from the upper direction of the skin; An image extraction step (S2) of generating an extracted image by extracting a preset color from the measured image; and an image binarization step (S3) of binarizing the extracted image.

According to one embodiment, an image analysis step (S4) of analyzing skin features in the binarized image may be further included.

Figure 6 is a flowchart schematically showing a two-dimensional skin image analysis method according to another embodiment of the present invention.

According to one embodiment of the present invention, in particular, measuring pores of facial skin can be performed in the following manner.

A two-dimensional skin image analysis method for analyzing pore features according to an embodiment of the present invention includes an image measurement step (S11) of measuring a face image by irradiating light to the skin from the upper direction of the face; A face image pre-processing step (S13) of performing pre-processing for focus correction by applying a high-pass filter to the measured face image; An extraction image generation step (S15) of generating a blue extracted image by extracting only the blue color from the preprocessed image; A binarized image generation step (S17) of generating a binarized image by increasing the contrast ratio to extract only pore features from the extracted image; and a skin feature analysis step (or image analysis step) (S19) of analyzing the features of the pores based on the binarized image.

In the image analysis step, it may include analyzing pores in the binarized image to calculate pore characteristic data consisting of one or more of pore elongation length, pore area, pore distribution, pore angle, and pore number.

For this purpose, the blue extracted image can be used to highlight only the pore area of the skin and remove the pigmentation area or skin curve area. Accordingly, it is possible to provide a two-dimensional skin image analysis method that can accurately analyze only pore features.

According to one embodiment of the present invention, it is possible to provide a two-dimensional skin image analysis system and analysis method that can measure an accurate two-dimensional skin image by deriving undistorted shadows of skin features, especially pore areas.

Furthermore, in the image analysis step, based on the binarized pore image or pore characteristic data, pore characteristics by race or age can be identified, and the subject's pore age or race can be estimated.

Specifically, the image analysis step includes collecting and storing pore images or pore feature data binarized by race or age, and binarized images or pore feature data of the measured skin and the binarized image or pore feature data by race or age. The method may further include estimating the subject's race or age group by comparing pore images or pore characteristic data.

According to an embodiment of the present invention, since pore characteristics can be accurately and quantitatively determined, it is possible to accurately determine pore characteristics by race or age, and based on this, determine the subject's pore age, etc. For example, by informing the subject of the age of his or her pores, the subject can more easily determine the state of his or her pores.

In addition, based on the binarized pore image or pore characteristic data, the pore impact of the product (eg, pore shrinkage, pore age change, etc.) can be calculated.

Specifically, the image analysis step includes collecting and storing binarized pore images or pore characteristic data according to the usage time or usage amount of the test cosmetic, and binarized pore images or pores according to the usage time or usage amount of the test cosmetic. It may further include calculating the degree of influence of the test cosmetic on pores based on changes in characteristic data.

Accordingly, it can be used to analyze the impact of pore-related products on pores through binarized pore images and quantitative pore characteristic data based on them. For example, through changes in the pore characteristic data of subjects using the product, information such as how much the size of the pores has been reduced, how much the distribution of pores has been reduced, and how much the pore elongation length has been reduced is provided to the subject in quantitative numbers or You can check it with an image. Various embodiments of the present invention can be used to evaluate the effect of cosmetics on pores.

Ultimately, through the various embodiments of the present invention described above, by extracting a preset color from a skin image and using the color extraction image, only the desired features can be extracted and analyzed, regardless of skin color, pigmentation area, or skin curvature. A two-dimensional skin image analysis system and analysis method can be provided.

In addition, it is possible to provide a two-dimensional skin image analysis system and analysis method that can provide clear images through focus correction and provide accurate numerical values for characteristics of features.

And, according to one embodiment of the present invention, since a two-dimensional skin image is used, it is possible to freely adjust the area of the region of interest for which features are to be analyzed. In other words, it is possible to provide a two-dimensional skin image analysis system and analysis method that can quickly and easily measure a wide range of pores.

Equipment using 3D images is expensive and complex equipment, but the 2D skin image analysis system can specifically measure and analyze the characteristics of pores in a cheap and simple way, and allows for easy-to-understand and intuitive evaluation through images. It is possible to provide a two-dimensional skin image analysis system and analysis method that can provide objective evidence that can be used.

In order to evaluate pore features, they were analyzed using a two-dimensional skin image analysis system according to an embodiment of the present invention. The photos in FIG. 7 are photos derived by this two-dimensional skin image analysis system.

VISIA-CR ^® was used as the image measurement unit (1), and only the upper light source F11.0+8 was turned on in Standard 1 mode and all other light sources were blocked. The ISO value was set to 100, the aperture value was set to F16, the resolution was 5616 × 3744, the white balance was set to daylight, and the pores on one cheek were viewed from the front of the subject. A region of interest (ROI) (A ₂ ) was set and photographed (see (a) of FIG. 7 ).

Preprocessing was performed by applying a highpass filter with an intensity of 5 and a pass value of 10 to the region of interest (A ₂ ) of the captured image as shown in (b) of Figure 7.

Then, as shown in Figure 7 (c), only the blue color was extracted to create a blue extracted image. At this time, while the pore area appears clearly emphasized, the pigmentation area also appears minutely.

Afterwards, as shown in (d) of Figure 7, the contrast ratio was increased to the value before the pigmentation area appeared to generate a binarized image. Referring to (d) of FIG. 7, it can be seen that the pigmentation areas (P ₁₁ and P ₁₂ ) that appeared in (b) of FIG. 7 have been removed. It can be seen that the areas that appeared as skin texture (P ₉ , P ₁₀ ) were also removed from the final binarized image, and were converted to a black and white image with only the pore area displayed in white.

subject

By age, participants were in their 20s (27.36 ± 1.36 years, 11 people), 30s (34.64 ± 2.45 years, 33 people), 40s (44.87 ± 3.22 years, 45 people), 50s (52.60 ± 2.54 years, 20 people), or 60s (52.60 ± 2.54 years, 20 people). Pores characteristics were measured in 116 Korean women (age 61.00 ± 1.42 years, 7 people). To prevent environmental influences, subjects were allowed to rest in a laboratory maintained at a temperature of 20 ± 2 °C and relative humidity of 40 ± 2% for 20 min. Then, optical images were measured.

image measurements

First, light was irradiated from all directions on the subject's face according to the prior art as described in FIG. 1, converted to gray scale using an image analysis program, and then the pore area of the pore area was measured in pixels.

And, according to one embodiment of the present invention, facial images were measured for the front in Standard 1 mode using VISIA-CR ^® . Since some subjects may experience loss of light due to the curve of the subject's face, the light source was adjusted so that only the upper part of the subject was illuminated and all light from other directions was eliminated.

image processing

The measured images were processed using Image-Pro Plus 9.2. The region of interest (ROI) was designated as 700 Finally, a threshold was applied to remove images of skin texture and skin pigment, excluding facial pores. In the image after image processing, the pores appeared white in color. The total area of pores was measured by calculating the sum of the areas of the white areas. Pore elongation was calculated as the ratio of the short diameter to the long diameter in oval-shaped pores.

Through image processing, pore images according to an embodiment of the present invention can calculate the pore area by calculating the number of pixels, and it is known that the distribution, direction, and shape of pores can be analyzed using the visual image itself. I was able to.

statistical analysis

The collected results were analyzed using SPSS from SPSS Inc. (Chicago, IL, USA). For normally distributed data, one-way analysis of variance was used, and for non-normally distributed data, the Mann-Whitney U test was used. Following the Kruskal-Wallis test, multiple comparisons were performed with Bonferroni's correction to compare differences by age. Statistical significance was assumed to be P<0.05.

Changes in pore shape and area by age

8 shows age-specific pore images (A, B, C, D, E) measured according to an embodiment of the present invention as described above, and pore images (F, G, H) on which image processing was performed on the measured images. , I, J).

Referring to Figure 8, it was found that the pores of subjects in their 20s were mainly located around the center of the face around the nose (A, F). However, the pores of subjects in their 30s were found to be located around the nose and cheeks (B, G). It was found that the pores of subjects in their 40s appeared more prominently than those of subjects in their 30s and had a U-shaped pore direction (C, H). It was found that the pores previously formed in a U-shape changed to an angle and several pores merged. The pores of subjects in their 50s showed a clearer direction, but the pore area indicated by the number of pixels was smaller than that of the pores of subjects in their 40s (D, I). While most of the pores of subjects in their 60s disappeared, the pores around the nose remained wrinkled (E, J).

Figure 9 is a graph showing the pore area distribution by age group measured according to the prior art (a) and the above-described embodiment (b) of the present invention.

Referring to Figures 9 (a) and (b), it can be seen that according to the prior art and an embodiment of the present invention, the pore area increases from the 20s to the 40s and then tends to decrease from the 50s. .

However, referring to (a) of Figure 9, according to the prior art, the pore area is approximately between 3000 pixel and 6000 pixel across all age groups, and when compared based on the average of people in their 20s and 30s. The averages were similar for those in their 40s, 50s, and 60s. It was difficult to analyze changes in pore distribution, direction, and shape, and only pore area and number of pores could be confirmed.

In contrast, referring to (b) of FIG. 9, according to one embodiment of the present invention, the pore area was distributed in the range of approximately 5000 pixels to 15000 pixels, and when compared based on the average of people in their 20s, the average difference was more detailed. It appeared. In other words, differences between ages could be clearly distinguished. In addition, compared to the graph in (a) of Figure 9, the deviation in the graph in (b) is greatly reduced, and it can be seen that the deviation value does not change significantly across age groups, thereby increasing the measurement accuracy and reliability of the data. It can be confirmed that. In Figure 9(b), the pore area of subjects in their 20s is 5604.45 pixels, and the pore area of subjects in their 30s increases significantly by approximately 80.97% compared to the pore area of subjects in their 20s. could know that Meanwhile, the pore area of subjects in their 40s was found to be the largest among all age groups, and increased by approximately 176.49% compared to subjects in their 20s. It was confirmed that the pore area of subjects in their 40s or older gradually decreased until their 60s.

Changes in pore elongation by age

FIG. 10 is an enlarged view of the processed images of FIG. 8 and shows them by age.

Referring to Figure 10, in the case of subjects in their 20s, the pores show a circular shape (A), and this is maintained until the 30s (B). However, in people in their 40s, pores have an oval shape and direction (C), while in people in their 50s, several pores are combined and have a distinct direction (D). These characteristics were more clearly revealed in subjects in their 60s (E), and their pores could be seen looking like wrinkles.

Figure 11 is a bar graph (a) showing the average pore elongation rate (ratio of long diameter/short diameter of oval pores) of subjects by age and a line graph (b) showing their average and distribution.

Referring to Figure 11 (a), the pore elongation rate of subjects in their 20s was approximately 2.7, and did not show statistically significant changes until their 30s. However, in the case of subjects in their 40s, the pores had a long oval shape and the elongation rate increased by approximately 8.25%, and there was no statistically significant change compared to the pore elongation rate of those in their 40s until the 50s. In the case of subjects in their 60s, rapid pore elongation was observed, and the pore elongation rate was found to increase by approximately 16.3% compared to subjects in their 20s.

Referring to Figure 11 (b), it can be seen that the length of pores gradually increases throughout life, and the pore elongation rate has a positive correlation with age (r2=0.424, p<0.001).

Figure 12 is an image of the measured pores of subjects in their 50s and 60s, and the elongation direction of the pores could be confirmed. As subjects aged in their 50s or 60s, several pores coalesced, and the skin around the pores formed wrinkle-like skin condensations along the nasolabial line.

In other words, it was found that after the age of 50, the pores around the wrinkles look like wrinkles, and the range (range between arrows a1-a2, range between arrows a3-a4) becomes wider and clearer than in the age of 60.

According to the prior art, it was difficult to accurately measure and determine the characteristics of pores due to the influence of skin pigment and skin curvature. However, according to one embodiment of the present invention, by using light irradiated from the upper direction, the influence on remaining parts other than the pores can be minimized and the shade ratio of the pores in the image can be strengthened. Accordingly, the size and size of pores can be clearly measured, and pore characteristics according to age can also be accurately measured.

Specifically, in appearance, the size and number of pores are different and the pore pattern varies depending on age, but this could not be clearly analyzed without accurate measurement of pore characteristics. However, according to the present invention, pore characteristics can be accurately measured to provide quantitative analysis. Since it can be provided as data, pore characteristics by age or race can be clearly analyzed.

Accordingly, according to the present invention, pore characteristics by age group, such as pore elongation rate, pore area, pore direction, and pore distribution, can be converted into data, and the subject's pore age can be estimated by measuring the subject's pore characteristics. In addition, it can be applied to evaluate the efficacy of a product by identifying changes in the subject's pore characteristics depending on the product.

1: Image measurement unit
3: Image processing unit
5: Image analysis unit
11: light source
13: Image measurement unit
31: Image extraction unit
33: Image editing department

Claims (17)

an image measurement unit comprising a light source for irradiating light to the skin from an upper direction of the skin to induce an undistorted shadow on the skin feature, and an image measurement unit for measuring a two-dimensional skin image;
An image processing unit including an image extraction unit for generating an extracted image by extracting a preset color from the measured image to remove skin color elements and shadows due to skin curvature, and an image binarization unit for binarizing the extracted image; and
comprising a skin analysis unit that analyzes skin features in the binarized image,
The skin feature is a pore feature, which is a skin area for pore analysis,
The image extractor removes shadows caused by skin curvatures, including blemishes, pigmentation including spots, and nasolabial folds formed on the face so that the skin analysis unit can analyze from various angles by minimizing errors in the pore features,
The skin analysis unit analyzes one or more of pore elongation length, pore area, pore distribution, pore angle, and pore number,
Collect and store pore images or pore characteristic data rationalized by race or age,
A two-dimensional skin image analysis system characterized in that the race or age group of the subject is estimated by comparing the binarized image or pore characteristic data of the measured skin with the binarized pore image or pore characteristic data by race or age.
According to claim 1,
The image measuring unit is arranged to face the target skin,
A two-dimensional skin image analysis system, wherein the light source is disposed upward with respect to an axis connecting the skin and the image measuring unit.
According to claim 1,
The target skin is facial skin,
A two-dimensional skin image analysis system, wherein the light source is arranged in a range of 0° to 90° to the left or right around the axis in the direction of the crown of the head.
According to claim 1,
The target skin is facial skin,
A two-dimensional skin image analysis system, wherein the light source is arranged in a range of 0° to 90° in the frontal direction around the crown axis of the head.
delete According to paragraph 1,
A two-dimensional skin image analysis system wherein the image extraction unit extracts blue from the two-dimensional skin image and maximizes the difference in shade of pores.
According to paragraph 1,
If the skin feature is a skin area for analyzing sebum distribution and area,
A two-dimensional skin image analysis system wherein the image extraction unit extracts red color from the two-dimensional skin image and maximizes the difference in shade of sebum formed on the skin.
According to paragraph 1,
If the skin feature is a skin area for analyzing whitehead distribution and area,
A two-dimensional skin image analysis system wherein the image extraction unit extracts green color from the two-dimensional skin image and maximizes the difference in shade of white heads formed on the skin.
According to claim 1,
The image processing unit,
A two-dimensional skin image analysis system further comprising a pre-processing unit that applies a filter to the measured image and pre-processes it.
According to clause 9,
A two-dimensional skin image analysis system, characterized in that the filter is a high-pass filter.
According to claim 1,
A two-dimensional skin image analysis system, characterized in that the preset color is blue.
According to paragraph 1,
A two-dimensional skin image analysis system wherein the image encoding unit adjusts the contrast of skin features within a preset value range to remove facial curves or fine wrinkles and make only pores appear.
According to clause 12,
The image binarization unit is set to a value before the pigmentation area appears, and generates a binarized image showing only pores with the pigmentation area removed.
According to clause 1,
The skin analysis unit is
If the binarized skin image is a black-and-white image with pores appearing white, the total pore area is calculated by adding the areas of the white areas, or
Based on the pore elongation length, a two-dimensional skin image analysis system characterized in that the elongation of the pore is calculated by dividing the length with the longest diameter in each pore by the length with the shortest diameter.
According to clause 1,
The image analysis unit is,
Collect and store binarized pore images or pore characteristic data according to the usage time or amount of test cosmetics,
A two-dimensional skin image analysis system that calculates the influence of the test cosmetic on pores based on changes in the binarized pore image or pore characteristic data according to the usage time or amount of the test cosmetic.
According to claim 1,
The image measurement unit is,
2, characterized in that it is an integrated image measurement device consisting of the light source, the image measurement unit, a fixture for fixing the subject's position with respect to the image measurement unit, and a housing containing the light source, the image measurement unit, and the fixture therein. Dimensional skin image analysis system.
A two-dimensional skin image analysis method performed by a processor, comprising:
An image measurement step of measuring a two-dimensional skin image by an image measurement unit by irradiating light to the skin from an upper direction of the skin to induce an undistorted shadow on the skin feature;
An image extraction step of generating an extracted image by extracting a preset color from the measured image by an image extraction unit to remove skin color elements and shadows caused by skin curvature;
an image binarization step of binarizing the extracted image by an image binarization unit; and
A skin analysis step of analyzing skin features in the binarized image by a skin analysis unit,
The skin feature is a pore feature, which is a skin area for pore analysis,
The image extraction step removes shadows caused by skin curves including blemishes, pigmentation including spots, and nasolabial folds formed on the face so that the skin analysis step can analyze the pore features in a multifaceted manner by minimizing errors,
The skin analysis step analyzes one or more of pore elongation length, pore area, pore distribution, pore angle, and pore number,
Collect and store pore images or pore characteristic data rationalized by race or age,
A two-dimensional skin image analysis method comprising estimating the subject's race or age group by comparing the binarized image or pore characteristic data of the measured skin with the binarized pore image or pore characteristic data by race or age. .