Methods for identifying dendritic pores

Abstract

This invention provides a method for identifying dendritic pores. Fine line images and pore images are obtained from digital images of a subject's skin. These fine line and pore images are overlaid to identify pores having at least one fine line intersecting with the pore as dendritic pores.

Description

Technical Field

[0001] The present invention relates generally to a method for identifying dendritic pores on the skin. Background Technology

[0002] Skin features such as enlarged pores and fine lines are common problems in cosmetic dermatology and the non-medical aesthetic industry. For example, the presence of enlarged pores can negatively impact skin elasticity, which in turn leads to sagging skin, particularly around the nose and cheek areas of an individual's face. This causes many individuals, especially younger ones, to seek various treatment options to help address pore-related issues. Fine lines and wrinkles are typical symptoms of aging skin, but identifying them in younger individuals is particularly challenging. Having the opportunity to better identify aging skin features in younger individuals allows for preventative treatments to prevent or delay the onset of age-related symptoms and a better assessment of the effectiveness of any two cosmetic treatments. Non-invasive methods, such as dermoscopy or confocal laser microscopy, are commonly used to assess skin features. However, these methods are limited in that only a very small or narrow area (e.g., 15 mm in diameter) can be examined per measurement. Therefore, it may not be suitable for measuring larger areas, such as the entire cheek area or the entire face, as multiple measurements would have to be performed. Furthermore, these methods are not easy for many users to operate and require a certain level of training. There is a continued opportunity to develop more accurate and / or precise methods to identify user-friendly skin features, thus enabling the recommendation and / or development of optimized non-medical aesthetic skin care treatments for individuals, particularly younger individuals with less pronounced skin features compared to older individuals. Summary of the Invention

[0003] This invention is based at least in part on the discovery of dendritic pores, and further on the role of dendritic pores in estimating skin and therefore individual skin care needs. This method can also be applied to help develop skin care treatments. Dendritic pores are pores in which fine lines or wrinkles intersect with the pore. One aspect of the invention provides a method for identifying dendritic pores, the method comprising the steps of: obtaining a digital image of a subject's skin; generating a fine line image from the obtained digital image; generating a pore image from the obtained digital image; and overlapping the fine line image and the pore image to form a first overlapping image to identify those pores having at least one fine line intersecting with the pore, thereby identifying dendritic pores.

[0004] Compared to other methods, the advantage lies in the increased sensitivity of the method used to estimate skin characteristics, especially in younger subjects. Attached Figure Description

[0005] The invention will now be described by way of illustrative example only with reference to the accompanying drawings, in which:

[0006] Figure 1 A general flowchart is depicted for a method used to identify dendritic skin pores;

[0007] Figure 2 Depicting Figure 1 A more detailed version of the method described therein;

[0008] Figure 3 It's an image of the cheek, and Figure 1 and Figure 2 The corresponding output of the described method, applied to a cheek image to identify multiple dendritic pores comprising clusters of dendritic pores connected to each other; and

[0009] Figure 4 yes Figure 3 A close-up of one of the dendritic pore clusters.

[0010] Figure 5 It is a graph that measures the average dendritic pore area of ​​users aged 20 to under 80. Detailed Implementation

[0011] The features and beneficial effects of various embodiments of the present invention will become apparent from the following description, which includes examples of specific embodiments intended to give a broad representation of the invention. Various modifications will become apparent to those skilled in the art from this description and from practice of the invention. The scope of the invention is not intended to be limited to the specific forms disclosed, and the invention covers all modifications, equivalents, and alternatives falling within the spirit and scope of the invention as defined in the claims.

[0012] As used in this article, the term "cosmetic" refers to non-medical methods that provide a desired visual effect on areas of the body. Visual cosmetic effects can be temporary, semi-permanent, or permanent.

[0013] As used herein, the term "facial region" refers to the entire face of a user or a portion of the user's face, including but not limited to one or more of the following regions: cheeks, nose, forehead, mouth, chin, periorbital region, and neck region.

[0014] As used herein, the term "image capturing device" refers to a means, system, or instrument capable of capturing and / or recording images (e.g., still images or videos), preferably digital images. This device may be part of a clinical imaging system or a skin assessment system based on a skin counter. The device may be part of a mobile or smart device, including mobile phones, smartphones, tablets, laptops, watches, personal digital assistants, or may be part of a personal computer, or may be a standalone camera such as a handheld camera. The device may also include a built-in light source (e.g., a flash) for emitting light.

[0015] As used herein, the term "skin" refers to the outermost protective covering of a mammal, composed of cells such as keratinocytes, fibroblasts, and melanocytes. Skin comprises the outer epidermis and the underlying dermis. Preferably, the skin is the skin of the facial region.

[0016] As used herein, the term “skin features” refers to features on the skin of a subject, including but not limited to one or more of the following: pores, shine, fine lines (including wrinkles), spots, hair, moles, papular acne, blackheads, whiteheads, and any combination thereof.

[0017] As used herein, the term “subject” refers to a person to whom the methods (and systems) described herein are used.

[0018] Figure 1 This provides a general flowchart of a method (1) for identifying dendritic pores. The first step is to obtain a digital image (3) of the subject. From this digital image, a fine line image (5) is generated. Additionally, from this digital image (3), a pore image (7) is generated. Finally, the fine line image (5) and the pore image (7) are superimposed on each other to provide a first superimposed image (9) to identify those pores having at least one fine line intersecting with the pore, thereby identifying dendritic pores.

[0019] Figure 2 yes Figure 1 A more detailed flowchart (1000) is provided. Similarly, method (31) provides a digital image (300) of the subject. The image can be obtained from an image capture device. For example, the image capture device is a Canon with customized light settings and polarization filters. ® The imaging system of the 5D Mark II full-frame digital single-lens reflex (DSLR) camera, such as Visia... ® -CR imaging system (Canfield Scientific, New Jersey, USA), this imaging system includes Canon ®5D Mark II DSLR camera. If the image is a whole human face or other large-area image, the region of interest only involves a portion of such an image. Alternatively, images can be retrieved from computer memory (where the stored images were captured by the image capture device from an earlier point in time).

[0020] The subject's digital image (300) may display multiple skin features. These skin features may include, for example, spots (12) and moles (14). Notably, skin features also include individual pores (16), fine lines (20), and dendritic pores (18). Some of these skin features are not visible to the naked eye. It should be understood that not all of these skin features are readily apparent in the subject's digital image (300) without the digital processing steps described subsequently.

[0021] Dendritic pores (18) are those pores connected to at least one fine line. A dendritic pore is a pore in which a fine line intersects with the pore. For example, intersecting fine lines can be intersecting fine lines that contact the outer boundary of the pore or intersecting fine lines that pass through the pore. A dendritic pore may intersect with 2, 3, 4, or more fine lines. One dendritic pore can connect to other dendritic pores (via a shared fine line (or wrinkle)). Subsequently, these interconnected dendritic pores can form clusters (where clusters can be separated from each other). Without being bound by theory, dendritic pores are associated with actual age. Dendritic pores can appear over time in facial areas due to physically loaded stress caused by frequent facial expressions and / or decreased elasticity due to aging. This correlation is demonstrated in Example 1. That is, the average number of dendritic pores per defined unit area increases with the age of the subject. Studying dendritic pores as a phenotype of skin health and aging is important in cosmetic research. Furthermore, free from theoretical constraints, the distinction between dendritic pores (18) and isolated pores (16) (i.e., pores that do not intersect with any fine lines) may be important in conveying signs of premature skin aging and their perception to younger users (e.g., 18 to 25 years old). The methods described in this paper can contribute to the development of innovative new cosmetic skincare products or programs, as well as cosmetic skin diagnosis.

[0022] Still referencing Figure 2A pore image (700) is generated from a digital image (300) of the subject. The pore image (700) is generated by extracting pores (16, 18A) from the digital image (300) of the subject. These pores may be from individual pores (16) or dendritic pores (18A). Dendritic pores (18A) are present in the generated pore image (700), but there are no intersecting fine lines (18B). Although not shown, in a preferred example, the extracted pore image (700) identifies the boundary of each pore in the pores (16, 18A). The pore image (700) can be extracted by segmenting the digital image (300) of the subject. The segmentation of the digital image (300) of the subject can be performed by one or more methods such as thresholding methods, color-based segmentation methods, transformation methods, texture methods, or combinations thereof. Preferably, the segmentation of the digital image (300) of the subject is performed by a thresholding method, and more preferably by an adaptive thresholding method.

[0023] Similarly, a fine line image (500) is generated from the digital image (300) of the subject. The fine line image (500) is generated by extracting fine lines (18B, 20) from the digital image (300) of the subject. Fine lines may originate from individual fine lines (20) (i.e., those not intersecting with pores) or from fine lines associated with one or more dendritic pores (18A) that intersect with pores. Although not shown, in a preferred example, the extracted fine line image (500) identifies the boundary of each fine line in the fine lines (18B, 20). The fine line image (500) can be extracted by segmenting the digital image (300) of the subject. Segmentation can be performed according to the method described previously.

[0024] Optionally, to improve the accuracy of identifying the boundaries of each of the pores (16, 18A) and / or fine lines (18B, 20), the digital image (300) of the subject may be processed before extracting the pore image (700) and / or the fine line image (500). For example, histogram equalization may be performed on the digital image (300) of the subject to enhance the contrast and / or improve the illumination of the image (300), thereby obtaining a histogram equalized image. The histogram equalized image or the unprocessed image (300) may be filtered to remove one or more skin features (e.g., spots (12) and moles (14)) as desired, thereby obtaining a filtered image. Filters such as frequency filters may be used to filter the histogram equalized image or the unfiltered image (300). Examples of frequency filters include fast Fourier transform filters used in conjunction with bandpass filters and difference-of-Gaussian filters. Preferably, the histogram equalized image of the image (300) is filtered using a difference-of-Gaussian filter. After obtaining the histogram equalized or filtered image, segmentation is performed on the histogram equalized or filtered image to extract pore images (700) and / or fine line images (500).

[0025] Optionally, one or more additional filters may be used to further improve the accuracy of identifying the boundaries of each of skin pores (16, 18A) and / or fine lines (18B, 20). For example, a watershed transformation filter may be used to delineate contiguous skin pores (16, 18A) that would otherwise be identified as boundary fine lines (not shown). Similarly, the same watershed filter may be used to delineate fine lines (18B, 20). Another example of a filter is a shape filter. This shape filter may include defining pores by pore geometry parameters; preferably, the pore geometry parameter is the pore area (preferably 25,000 micrometers). 2 -1×10 6 micrometer 2 ), diameter (preferably 175 micrometers-1100 micrometers), width / length aspect ratio (preferably 0.3-1), and combinations thereof.

[0026] The shape filter includes defining screeds by screed geometry parameters; preferably, the screed geometry parameters are selected from screed thickness (preferably greater than 35 micrometers, more preferably 40 micrometers to 1 cm), screed length (preferably greater than 200 micrometers, more preferably 250 micrometers to less than 5 cm, preferably less than 3 cm, more preferably 250 micrometers to 1 cm), and combinations thereof.

[0027] Although not shown, it is preferable to process the digital image of the subject (according to one or more of the steps described above). More preferably, a fine line image and a pore image are generated from the processed image. Further, a binary fine line image and / or a binary pore image are generated. Optionally but preferably, an intersection step may be applied to the binary pore image and / or the binary fine line image to help identify very small pores detected along with the fine lines in the binary fine line image. A joint step will help to identify the fine lines in the binary fine line image and / or the pores in the binary pore image, respectively. Image processing programs and programming languages ​​such as MATLAB are used. ® Python ™ OpenCV, Java ™ ImageJ can be used to perform the above steps.

[0028] Finally, and still refer to Figure 2 The fine line image (5) and the pore image (7) are overlaid on each other to identify those pores having at least one fine line intersecting with the pore, thereby identifying dendritic pores (18) in the first overlaid image. The second overlaid image (19) provides individual pores (16). The third overlaid image (23) will provide individual fine lines (20). Preferably, morphological reconstruction is applied to the first overlaid image to identify those pores having at least one fine line intersecting with the pore. In a non-limiting example, MATLAB ® Used for morphological reconstruction.

[0029] Figure 3 A method for identifying dendritic pore clusters was described (31). Images of the cheeks of the subject's facial region were provided (3000). Application Figure 1 and Figure 2 The described method identifies a first overlapping image (9000) comprising multiple dendritic pores. An example of identifying a cluster (1800) of dendritic pores in the first overlapping image (9000). Figure 3 The first overlapping image contains 57 distinct dendritic clusters (1800). These identified dendritic pores can be classified into one or more predetermined categories. Such categories may include: dendritic pores with one ridge; dendritic pores with two ridges; dendritic pores with at least three ridges; dendritic pores connected to another dendritic pore via at least one ridge; and combinations thereof.

[0030] Figure 4 yes Figure 3 A close-up of a dendritic pore cluster (1800). The cluster consists of six different dendritic pores. These pores in the cluster are connected by at least one fine line.

[0031] The method described herein may provide an additional display step. That is, the identified dendritic pores or dendritic pore clusters can be displayed to the subject. This display can be made via various well-known methods, including through computer-visualized websites or applications. Preferably, the display is made via a mobile smartphone with a visual screen.

[0032] The method described herein may also provide an additional step of determining a numerical severity that is at least partially associated with the identified dendritic pores in the subject (and displaying the determined numerical severity to the subject). A non-limiting example of such numerical severity is a numerical severity scale based on a 1-5 scale. 1 represents the minimum number of dendritic pores in a defined unit of the facial region, while 5 represents a larger number of dendritic pores. Alternatively, the numerical severity may be correlated with other skin parameters to provide a broader overall numerical severity for the subject. Without limitation, these other skin parameters may include those based at least in part on blemishes, texture, wrinkles, individual pores, skin tone, brightness, and other imaging measurements. Further other skin parameters may include those based on in vivo physical measurements such as dryness, hydration, barrier function, and sebum secretion.

[0033] The method described herein may also provide additional steps to generate comparisons between the numerical severity of the subject and predetermined values ​​associated with the population. The population data used may be specific to the subject's age, geographic location, ethnic origin, or any other factor. See US 6,571,003 B1, column 9, lines 5 through 47, which is incorporated herein by reference.

[0034] Example 1

[0035] Figure 5 This is a table plotting the average dendritic pore area measured across users of different ages. These users ranged from 20 to under 80 years of age. Linear regression determined that the dendritic pore area increased by an arbitrary 2.14 units per year. Digital images of the subjects' facial skin were obtained using an iPhone. ® The data was obtained from the main camera in section 7. Using the method described above, the data in Table 1 was generated.

[0036] Table 1. Identification of dendritic pore area across the first and second user groups, and in the second user group In the body, the area of ​​dendritic pores changes throughout the day. .

[0037]

[0038] A surprising finding was a significant increase in dendritic pore area throughout the day, at least in the first user group. Specifically, when measured between morning and evening of the same day, there was an average change of more than 19.39 arbitrary units. Dividing 19.39 by 2.14, the increase in dendritic pore area on an annual basis (determined by linear regression across all users) represents a change equivalent to 9.07 years. This is a significant increase in “skin age” throughout the day for at least the first user group. Therefore, this provides an opportunity to develop non-medical skin care products and treatments to address this diurnal variation (at least in the first user group). Accordingly, one aspect of the invention provides methods for identifying dendritic pores in subjects at a frequency of more than once a day to track these changes in dendritic pores and / or dendritic pore area. For example, the methods for identifying dendritic pores described herein are applied to subjects 2-10 times a day to understand these changes and how products or treatments affect them. Preferably, the method is performed at intervals of at least 30 minutes, more preferably at least one hour, and more preferably at least once in the morning and at least once in the evening. The method can also measure the area of ​​the identified dendritic pores (therefore any changes in the area of ​​the identified dendritic pores can be estimated).

[0039] The dimensions and values ​​disclosed herein should not be construed as strictly limited to the precise numerical values ​​cited. Rather, unless otherwise specified, each such dimension is intended to represent the stated value and a range around which it is functionally equivalent. For example, a dimension disclosed as “40 mm” is intended to represent “about 40 mm”. All numerical ranges described herein include narrower ranges; the upper and lower limits of the described ranges are interchangeable to further form ranges not explicitly described. Embodiments described herein may comprise the essential components described herein as well as optional components, and embodiments described herein are substantially composed of or comprise of the essential components described herein as well as optional components. As used in the specification and appended claims, unless the context clearly indicates otherwise, the singular forms “an,” “a,” and “the (described)” are intended to also include the plural forms.

[0040] Unless expressly excluded or otherwise limited, every reference cited herein, including any cross-references or related patents or patent applications, and any patent application or patent claiming priority to or benefiting from it, is incorporated herein by reference in its entirety. Reference to any reference is not an endorsement of it as prior art to any disclosed or protected art herein, nor is it an endorsement of any such invention, either on its own or in combination with any one or more references. Furthermore, where any meaning or definition of a term in this invention conflicts with any meaning or definition of the same term in referenced documents, the meaning or definition given to that term in this invention shall prevail.

[0041] While specific embodiments of the invention have been illustrated and described, it will be apparent to those skilled in the art that various other changes and modifications can be made without departing from the spirit and scope of the invention. Therefore, it is intended that all such changes and modifications falling within the scope of the invention be covered in the appended claims.

Claims

1. A method for identifying dendritic pores (18), the method comprising the following steps: (a) Obtain a digital image (300) of the subject’s skin; (b) Generate a striated image (500) with fine lines from the obtained digital image (300); (c) Generate a pore image (700) with pores from the obtained digital image (300); and (d) Overlay the fine line image (500) and the pore image (700) to provide a first overlay image (900) thereby identifying those pores having at least one fine line intersecting with the pore as dendritic pores (18).

2. The method of claim 1, further comprising the step of applying morphological reconstruction to the first overlapping image to identify those pores having at least one fine line intersecting with the pores.

3. The method according to claim 1 or 2, wherein the step of overlapping the fine line image (500) and the pore image (700) to identify those pores having at least one fine line intersecting with the identified skin pore boundary, is to identify the dendritic pores (18).

4. The method of claim 3, wherein the fine line image (500) identifies fine line boundaries, and the step of overlapping the fine line image (500) and the pore image (700) identifies those pores having at least one fine line boundary intersecting with the identified skin pore boundary to identify the dendritic pores (18).

5. The method according to claim 1 further includes the step of generating a binary fine line image (500) or a binary pore image (700).

6. The method according to claim 1 further includes the step of generating both a binary fine line image and a binary pore image.

7. The method of claim 1, further comprising the step of applying a shape filter to the obtained digital image of the subject's skin prior to generating the fine line image and the pore image.

8. The method of claim 7, wherein the shape filter includes defining the pore by pore geometry parameters.

9. The method of claim 8, wherein the pore geometry is selected from pore area, diameter, width / length ratio, and combinations thereof.

10. The method of claim 9, wherein the pore area is 25,000 micrometers. 2 -1×10 6 micrometer 2 .

11. The method of claim 9, wherein the diameter is 175 micrometers to 1100 micrometers.

12. The method of claim 9, wherein the width / length aspect ratio is 0.3-1.

13. The method according to any one of claims 7-12, wherein the shape filter includes defining the striates by striate geometry parameters.

14. The method of claim 13, wherein the stencil geometry parameters are selected from stencil thickness, stencil length, and combinations thereof.

15. The method of claim 14, wherein the fine texture thickness is greater than 35 micrometers.

16. The method of claim 14, wherein the stencil length is greater than 200 micrometers.

17. The method of claim 1, wherein the step of obtaining a digital image of the subject's skin further comprises performing histogram equalization to obtain a histogram equalized image.

18. The method of claim 1, further comprising the step of filtering any one of the above images by removing one or more irrelevant skin features.

19. The method of claim 18, wherein the unrelated skin feature is selected from spots, hair, moles, papules, acne, blackheads, whiteheads, and combinations thereof.

20. The method of claim 18, wherein the filtering step comprises using at least a frequency filter, wherein the frequency filter is selected from the group consisting of: fast Fourier transform filters and bandpass filters, difference Gaussian filters, and any combination thereof.

21. The method of claim 20, wherein the frequency filter is the Gaussian difference filter.

22. The method of claim 1, further comprising classifying the at least one dendritic pore into a predetermined category.

23. The method of claim 22, wherein the predetermined category is selected from: dendritic pores having one fine line; dendritic pores having two fine lines; dendritic pores having at least three fine lines; dendritic pores connected to another dendritic pore via at least one fine line; and combinations thereof.

24. The method of claim 1, further comprising the step of displaying the identified dendritic pores to the subject.

25. The method of claim 1, wherein the display is transmitted via a smartphone.

26. The method of claim 1, further comprising the step of determining a numerical severity associated with the identified dendritic pores.

27. The method of claim 26, further comprising the step of generating a comparison result between the numerical severity of the subject and a predetermined value associated with a population.

28. The method of claim 1, further comprising the step of identifying dendritic pores in the subject at a frequency of more than once per day.

29. The method of claim 28, wherein the frequency is 2-10 times per day.

30. The method of claim 28, wherein the frequency is spaced at least 1 hour apart.

31. The method of claim 28, wherein the step of identifying dendritic pores in a subject includes an additional step of determining the area of ​​the identified dendritic pores.

Images