KR20220023507A - Device and method for evaluating roughness of skin texture based on analysis of image - Google Patents

Abstract

Examples of the present invention relate to a device and method for evaluating roughness of skin texture based on image analysis. The device comprises: a step of acquiring an analysis image of target skin; a step of obtaining a low-frequency image from the analysis image using a low-pass filter; a step of obtaining a high-frequency image by subtracting the low-frequency image from the analysis image; and a step of calculating the roughness level of the skin texture using the high-frequency image.

Description

Apparatus and method for evaluating skin texture roughness based on image analysis

Embodiments of the present invention relate to an apparatus and method for evaluating skin texture based on image analysis.

A representative method for analyzing skin texture relates to a method for three-dimensional imaging of a skin surface using fringe projection. The interference fringe irradiation method projects a regular sine wave (e.g. sine) waveform intensity pattern by projecting a certain stripe on the skin surface at a fixed projection angle, and the height value of the generated surface is a distorted intensity pattern through phase change. , and they are recorded by appropriate imaging techniques. While this method has high accuracy, it has the disadvantage of using a large device and a complex program, so it is difficult to respond to the skin texture analysis needs based on digital image analysis such as smartphones/portable diagnostic devices.

In addition, the fringe projection has noise by imaging including the curvature of the skin surface. Therefore, as shown in Fig. 1, it is necessary to extract the fine grains by a polynomial method or the like in order to remove the influence of the large curvature of the skin surface.

In Figure 1, the following index values were calculated for the texture part from which the curvature was removed, and the skin texture was evaluated using the values.

[Equation 1]

[Equation 2]

Here, Ra is the average of the peak values of the entire measurement length, n is the total number of pixels, yi is the peak value in the corresponding pixel, and Sa is the average of the peak values of the analysis region, respectively.

Conventionally, it was evaluated that the smaller the index values of Equations 1 and 2, the better the texture. In analyzing the height value of any analysis area on the skin texture, a reference height is required, and an average is usually used as the reference height.

Sa is calculated in various ways based on Equation 2, and when calculated directly from Equation 2, when the average is used as the reference height, Sa becomes a negative number, and thus Sa is inevitably 0. In order to solve this problem, the absolute value process can be applied to Equation 2, but there is a limitation that one more processing is applied. Ra can also be calculated through an absolute value conversion process that turns a negative value into a positive value, but there is a limitation that one more processing is applied.

In addition, in the existing fringe projection, the skin texture can be measured in 3D only when there are interference fringes, and when there are shaded areas such as dots and shadows on the skin, it is impossible to measure because the interference fringes are not generated due to the influence of the background. Therefore, there was a problem in that it was difficult to measure the skin texture because only the effect of the curvature of the given skin surface was removed in the past, and the background influence such as shadows and blemishes on the skin surface could not be removed.

The present invention is to solve the above problems, and is an image analysis-based skin that can provide a reliable and accurate skin texture roughness value in order to respond to the recent digital image analysis-based skin texture analysis needs such as smartphones/portable diagnostic devices. This is to provide a texture evaluation method.

The apparatus for evaluating skin texture roughness according to an aspect of the present invention can accurately detect or quantitatively calculate a skin roughness value. As a result, it is possible to recommend cosmetics suitable for each skin condition together with a skin care method customized according to the skin condition of the subject.

In addition, in calculating the skin roughness value, the present invention does not require the use of interference fringes affected by background elements, unlike the conventional fringe projection, so that the skin texture can be measured more conveniently. In addition, since the present invention removes the background effect by using a filter, it is possible to more accurately measure the roughness of the skin texture regardless of the background influence such as shadows and blemishes on the skin surface.

Effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description of the claims.

In order to more clearly explain the technical solutions of the embodiments of the present invention or the prior art, drawings necessary for the description of the embodiments are briefly introduced below. It should be understood that the drawings below are for the purpose of explaining the embodiments of the present specification and not for the purpose of limitation. In addition, some elements to which various modifications such as exaggeration and omission have been applied may be shown in the drawings below for clarity of description.
1 shows a method of extracting fine grains by a method such as polynomials in order to remove the influence of curvature in order to calculate the grain of a conventional object.
2 shows an apparatus 1 for evaluating skin texture roughness according to an embodiment of the present invention.
3 is a flowchart of a method for evaluating skin texture roughness according to another embodiment of the present invention.
4A shows an original image, a low-frequency image, and a high-frequency image, and shows a perspective view for 3D-based skin texture measurement.
Figure 4b shows a cross-sectional view for 3D-based skin texture measurement of each of the original image, the low-frequency image and the high-frequency image.
5 is a graph according to a low-frequency filter condition during an optimization operation for obtaining a low-frequency image according to an experimental example of the present invention.
6 is a graph illustrating a significant correlation between a skin roughness level value calculated through an image analysis-based skin texture level evaluation apparatus and an actual skin roughness value according to an experimental example of the present invention.

The terminology used herein is for the purpose of referring to specific embodiments only, and is not intended to limit the present invention. As used herein, the singular forms also include the plural forms unless the phrases clearly indicate the opposite. The meaning of "comprising," as used herein, specifies a particular characteristic, region, integer, step, operation, item, and/or component, and includes the presence or absence of another characteristic, region, integer, step, operation, item and/or component. It does not exclude additions.

Although not defined otherwise, all terms including technical terms and scientific terms used herein have the same meaning as those commonly understood by those of ordinary skill in the art to which the present invention belongs. Commonly used terms defined in the dictionary are additionally interpreted as having a meaning consistent with the related technical literature and the presently disclosed content, and unless defined, are not interpreted in an ideal or very formal meaning.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

2 shows an apparatus 1 for evaluating skin texture roughness according to an embodiment of the present invention.

The image analysis-based skin texture roughness evaluation apparatus 1 includes an image capturing unit 10 that acquires an image of the target skin, and a first image processing unit 20 that obtains a low-frequency image representing the ground and curves from the acquired image through a filter ), a second image processing unit 30 for obtaining a high frequency image obtained by subtracting a low frequency image from the obtained analysis image, and a calculation unit 40 for calculating a skin texture roughness value of the high frequency image.

The image analysis-based skin texture roughness evaluation apparatus 1 according to embodiments may be entirely hardware, entirely software, or partially hardware and partially software. For example, the system may collectively refer to hardware equipped with data processing capability and operating software for driving the same. As used herein, terms such as “unit,” “system,” and “device” are intended to refer to a combination of hardware and software run by the hardware. For example, the hardware may be a data processing device including a central processing unit (CPU), a graphic processing unit (GPU), or another processor. In addition, software may refer to a running process, an object, an executable file, a thread of execution, a program, and the like.

For example, the concepts of the skin texture roughness evaluation device 1 (eg, the operations of the components 10 , 20 , 30 and 40 ) are (Canfield's (Fairfield, NJ, USA) VISIA- It may also be integrated into various computing devices (eg, smart devices, computers, etc.) as imaging systems for skin analysis (such as the CR® system) or software for computers or as customer apps. The software is designed to process the photographed image of the target skin, and may detect and/or evaluate the roughness characteristics of the skin texture roughness value of the target skin in the image, and further provide the operation result to the user.

The image photographing unit 10 is a component for photographing the skin of a subject, and may be implemented with various image photographing devices that can be used in the art. In an embodiment, the image capturing unit 10 is configured to obtain an optical image by capturing a skin image of the subject. For example, the image capturing unit 10 may be implemented with VISIA-CR® of Canfield Corporation (Fairfield, NJ, USA). Various imaging devices that can be used in the art can be used for this, but in the case of Visa-CR, standard 2 image is selected.

The image capturing unit 10 may provide skin image data of the target to the other components 20 , 30 or 40 . Also, the image capturing unit 10 may be further configured to perform an operation of extracting a partial region of the target skin image, for example, a cropping operation.

The first image processing unit 20 may include a low-pass filter. The low-pass filter extracts the curved and background components of the image by passing only the low-frequency components.

In an embodiment, the filter condition of the low-pass filter may have a size within a pixel range of 1 to 30 as a kernel size.

The second image processing unit 30 obtains a low frequency image from the first image processing unit 20 , and then subtracts the low frequency image from the original image to obtain a high frequency image. By the second image processing unit 30 , the curvature in the skin image of the target is removed, the height of the skin texture is more clearly deformed, and the skin texture roughness value can be calculated.

The calculator 40 may obtain a degree of roughness of the skin texture based on the high-frequency image acquired by the second image processing unit 30 . In order to obtain the roughness of the skin texture, it is necessary to calculate an index value corresponding to Ra (average of peak values among the entire measurement length) or Sa (average of peak values of analysis areas) of the prior art. Since it is analyzed as a height value, if it is lower than the reference height, it becomes a negative number, so if the average is the reference, Sa has no choice but to be 0. Ra can also be calculated as an absolute value conversion that turns a negative value into a positive value, but has the disadvantage of requiring one more processing. In order to quantitatively calculate the degree of roughness, the calculator 40 calculates a roughness value X _roughness based on a height value of at least a partial region of the high-frequency image. The calculator 40 applies the high-frequency image to the following equation.

[Equation 3]

Here, n denotes the total number of pixels in the high-frequency image, and xi denotes a height value of the corresponding pixel. In another embodiment, xi may be the difference (deviation) between the average height of all pixels and the height of the corresponding pixel.

The calculator 40 quantitatively calculates the degree of roughness through Equation 1, thereby overcoming the limitations of the conventional index values Ra and Sa.

Also, the calculator 40 may be further configured to apply a preset threshold value to the high-frequency image. The threshold value means a time point at which a color change in an image is minimally generated.

In an embodiment, the threshold value represents a minimum color change corresponding to a reference height. If a pixel has a height greater than the threshold value, it indicates that a color change greater than the minimum color change has occurred in that pixel. In addition, as the color is red, it indicates that the degree of roughness in the corresponding pixel is large. The calculator 40 may detect the degree of roughness of the skin displayed in the pixel by applying the threshold value. A region having a height value higher than the threshold value is detected as a region having a large roughness.

It will be apparent to a person skilled in the art that the apparatus 1 for evaluating skin texture roughness may include other components not explicitly described herein. For example, the prediction system may include network interfaces, input devices for data entry, and output devices for display, printing or other data presentation, storage devices such as memory, and other hardware necessary for the operations described herein. It may contain elements.

3 is a flowchart of a method for evaluating the degree of roughness of skin texture, according to another embodiment of the present invention. A method of evaluating the degree of roughness of the skin texture (hereinafter, "roughness evaluation method") is performed by a computing device including a processor. The computing device including the processor is, for example, performed by the apparatus 1 or some components (eg, the image processing units 20 and 30 and/or the calculation unit 40) for evaluating the roughness of the skin texture, or It may also be performed by other computing devices. Hereinafter, for clarity of explanation, the present invention will be described in more detail with examples in which the roughness evaluation method is performed by the skin texture roughness evaluation device 1 .

Referring to FIG. 3, the roughness evaluation method includes: (eg, by the image capturing unit 10) photographing the target skin and obtaining an analysis image (S10), and in the obtained analysis image, Applying the high-frequency image to Equation 1 in the steps (S20) of obtaining a low-frequency image representing the background and the curvature through a filter, obtaining a high-frequency image obtained by subtracting the low-frequency image from the obtained analysis image (S30) and (S40) and calculating a skin texture roughness value.

Figure 4a shows a 3D-based perspective view of the skin texture of each of the original image, the low-frequency image and the high-frequency image.

4B is a 3D-based cross-sectional view of the skin texture of each of the original image, the low-frequency image and the high-frequency image. Various parts, such as a face, a trunk, an arm, and a leg, may be measured as the target skin.

Referring to FIGS. 4A and 4B , in the calculation of the skin texture roughness value based on image analysis according to an embodiment of the present invention, it can be seen that the influence of the curvature and the background is removed from the original image. Referring to FIG. 1, in comparison with the prior art, which calculates the index values (Ra, Sa) for the grain portion from which only the curvature is removed and determines that the smaller the value, the better the texture, FIGS. 4A and 4B show the skin through image processing. It removes curves and backgrounds, such as removing shadows from color elements and skin curves, and allowing better measurement of skin texture roughness. In particular, when applied to the measurement of the skin texture of the face, it is possible to remove a part due to pigmentation such as freckles and spots formed on the face, and a shadow due to skin curves such as nasolabial folds.

Referring to FIGS. 4A and 4B , a specific region having a high degree of roughness of the skin has a larger color difference than a portion of a region having a small degree of roughness. The specific region has a relatively high skin texture and has a low flatness (eg, it is not flat and has a relatively high slope). On the other hand, other specific regions do not have a greater difference in color than regions having a large degree of roughness, and have relatively high flatness (completely or almost flat).

A roughness value is calculated by applying the high-frequency image of FIGS. 4A and 4B to Equation 1 (S40). Since the roughness value is calculated based on the height value of the high-frequency image from which the skin curves and the background are substantially removed, the roughness degree of the skin texture can be accurately calculated.

Experimental example

5 to 6 are diagrams for explaining an experiment for verifying the performance of the skin texture roughness evaluation apparatus 1 of FIG. 1 according to an experiment of the present invention.

5 is a graph according to a low-frequency filter condition during an optimization operation for obtaining a low-frequency image according to an experimental example of the present invention. The low-frequency image has different results depending on the conditions of the low-pass filter. When the average value of the kernel size is used, the correlation between the actual 3D skin results and the image analysis results is analyzed according to the size of the kernel, and the appropriate range of the kernel size can be calculated with a size within a pixel range of 1 to 30.

6 is a graph illustrating a significant correlation between a skin roughness degree value calculated through an image analysis-based skin texture roughness evaluation apparatus and an actual skin roughness value, according to an experimental example of the present invention. In the experimental example of Fig. 6, 32 subjects participated. For the same subject, the roughness value was evaluated using the skin texture roughness evaluation device of FIG. 1 and the Ra index value, which is a conventional roughness calculation method, was compared to compare the correlation. As a result, a graph of the form shown in FIG. 6 appeared. For each of 32 persons, the roughness value was evaluated using the skin texture roughness evaluation device, and the Ra index value, which is the conventional roughness calculation method, showed a significant correlation coefficient ( r=0.520, p-value=0.002). That is, the skin texture roughness evaluation apparatus 1 may accurately calculate the skin roughness level by analyzing the image.

The apparatus and method for evaluating the roughness of the skin texture do not need to use an interference fringe affected by a background element to measure the skin texture, unlike the existing fringe projection. In addition, since the present invention removes the background effect by using a filter, it has the effect of more accurately measuring the roughness of the skin texture regardless of the background influence such as shadows and blemishes on the skin surface.

The operation by the apparatus 1 and method for evaluating skin texture roughness according to the above-described embodiments may be at least partially implemented as a computer program and recorded in a computer-readable recording medium. For example, embodied with a program product consisting of a computer-readable medium containing program code, which may be executed by a processor for performing any or all steps, operations, or processes described.

The computer-readable recording medium includes all types of recording devices in which computer-readable data is stored. Examples of the computer-readable recording medium include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like. In addition, the computer-readable recording medium may be distributed in a network-connected computer system, and the computer-readable code may be stored and executed in a distributed manner. In addition, functional programs, codes, and code segments for implementing the present embodiment may be easily understood by those skilled in the art to which the present embodiment belongs.

Although the present invention as described above has been described with reference to the embodiments shown in the drawings, it will be understood that these are merely exemplary and that various modifications and variations of the embodiments are possible therefrom by those of ordinary skill in the art. However, such modifications should be considered to be within the technical protection scope of the present invention. Accordingly, the true technical protection scope of the present invention should be determined by the technical spirit of the appended claims.

Claims (8)

an image capturing unit to obtain an analysis image by photographing the target skin;
a first image processing unit configured to obtain a low-frequency image from the analyzed image by using a low-pass filter;
a second image processing unit for obtaining a high-frequency image by subtracting the low-frequency image from the analyzed image; and
An image analysis-based skin texture evaluation device comprising a; a calculator for obtaining a roughness degree of the skin texture using the high-frequency image.
The method of claim 1,
Image analysis-based skin texture evaluation apparatus, characterized in that the filter condition of the low-pass filter is determined as a kernel size in the range of 1 to 30 pixels to obtain the low-frequency image.
The method of claim 1,
The value (X _roughness ) of the roughness of the skin texture is calculated by applying the high-frequency image to the following equation by the calculator,
[Equation]

Here, wherein n represents the total number of pixels of the cropped high-frequency image, and x _i is a height value of the corresponding pixel.
4. The method of claim 3,
The x _i is an image analysis-based skin texture evaluation device, characterized in that the difference between the average height of all pixels and the height of the corresponding pixel.
As an image analysis-based skin texture evaluation method performed by a processor,
acquiring an analysis image of the target skin;
obtaining a low-frequency image from the analyzed image using a low-pass filter;
obtaining a high-frequency image by subtracting the low-frequency image from the analyzed image; and
Calculating the degree of roughness of the skin texture using the high-frequency image; image analysis-based skin texture evaluation method comprising a.
6. The method of claim 5,
An image analysis-based skin texture evaluation method, characterized in that the filter condition of the low-pass filter is set to a kernel size in the range of 1 to 30 pixels to obtain the low-frequency image.
6. The method of claim 5,
The value (X _roughness ) of the roughness of the skin texture is calculated by applying the high-frequency image to the following equation by the calculating step,
[Equation]

Here, where n represents the total number of pixels of the cropped high-frequency image, and x _i is a height value of the corresponding pixel.
8. The method of claim 7,
The x _i is an image analysis-based skin texture evaluation method, characterized in that the difference between the average height of all pixels and the height of the corresponding pixel.

Images