JP2018196426A - Pore detection method and pore detection device - Google Patents

Abstract

To provide a technique capable of highly accurately detecting pores.SOLUTION: A pore detection method includes: a candidate specification step (S11) of specifying as a candidate position, a position in a light-dark skin image corresponding to a light emission position in an ultraviolet ray skin image imaged under ultraviolet ray irradiation; and a determination step (S12) of determining whether the candidate position is a pore position or not on the basis of a pixel value of the candidate position in the light-dark skin image.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a skin image analysis technique, and more particularly to a technique for detecting pores from a skin image.

In Patent Document 1, a difference image is obtained by subtracting an image before smoothing from a smoothed image of an image obtained by photographing the skin surface, and pores are obtained using a pore image obtained by binarizing the difference image. A skin surface state analysis system for measuring the size of the skin is disclosed.
Patent Document 2 discloses that only the surface reflected light is selected for the purpose of analyzing the skin condition, so that the planes of polarization of incident light and reflected light are matched.
In Patent Document 3, at least one of skin pores, horn plug porphyrins, wrinkles, stains, and skin color is analyzed from an image of a subject's face taken under a predetermined illumination condition, and the analysis results and measurement are performed. It is disclosed that date and time and subject information are stored in association with each other to analyze the skin condition.
In Patent Document 4, a difference image between an R component and a B component is acquired from a skin image, a pore image is extracted by binarizing and smoothing the difference image, and the size of the pore is determined from the pore image. It is disclosed to seek.
In Patent Document 5, porphyrin is detected by binarizing an image obtained by extracting an R (red) component from an image photographed by irradiating the skin with ultraviolet rays, and the region, number, and area of the porphyrin detected are determined. It has been proposed to display. In this proposal, the information of porphyrin detected in this way is used for the evaluation of acne and hide-and-seek.

JP-A-7-55447 Japanese Patent Laying-Open No. 2015-62569 JP 2007-152084 A JP 2006-305184 A JP 2009-494 A

In the above-described method, the dark position is detected as a pore from the skin image using the fact that the pore is darker than the surrounding area. Moreover, the detection accuracy of pores can be improved by using a polarization image in which the influence of the internal scattered light is reduced and the surface reflection light component is strengthened.
However, in the case of a pore such as a clogged pore due to a square plug, the position of the pore in the light and dark skin image may not be dark. In addition, in a region with a large undulation such as the periphery of the nose, the light does not uniformly strike the entire region, so that the detection accuracy of pores decreases only from the brightness and darkness of the skin image.
The present invention provides a technique that improves such a situation and can detect pores with high accuracy.

Each aspect of the present invention employs the following configurations in order to solve the above-described problems.
The first aspect relates to a pore detection method. The pore detection method according to the first aspect includes a candidate specifying step of specifying a position in a light and dark skin image corresponding to a light emitting position in an ultraviolet skin image captured under ultraviolet irradiation as a candidate position, and the light and dark skin image And a determination step of determining whether or not the candidate position is a pore position based on the pixel value of the candidate position.

  A 2nd aspect is related with a pore detection apparatus. The pore detection device according to the second aspect is acquired by the acquisition unit that acquires an ultraviolet skin image captured under ultraviolet irradiation and a light and dark skin image captured by a method different from the ultraviolet skin image. A candidate specifying means for specifying a position in the light and dark skin image corresponding to a light emission position in the ultraviolet skin image as a candidate position, and the candidate position is determined based on a pixel value of the candidate position in the light and dark skin image. Determining means for determining whether or not to.

  As another aspect of the present invention, there may be a program that causes a computer to execute the pore detection method according to the first aspect, or a computer-readable storage medium that records such a program. Good. This recording medium includes a non-transitory tangible medium.

  According to each said aspect, the technique which can detect a pore with high precision can be provided.

It is a flowchart which shows the pore detection method which concerns on this embodiment. It is a figure which shows notionally the hardware structural example of the pore detection apparatus which concerns on this embodiment. It is a figure which shows notionally the process structural example of the pore detection apparatus which concerns on this embodiment. It is a conceptual diagram which shows the hardware constitutions of the skin diagnostic system which concerns on this embodiment. It is a flowchart which shows the pore detection method which concerns on this embodiment. It is a detailed flowchart of a pore addition step. It is a figure which shows the pseudo code which implement | achieves a pore determination step (S62) and a pore position addition step (S63). It is a figure which shows notionally the flow of the pore detection in this embodiment.

  Embodiments of the present invention will be described below. In addition, embodiment mentioned below is an illustration and this invention is not limited to the structure of the following embodiment.

First, the outline | summary of the pore detection method which concerns on embodiment of this invention is demonstrated using FIG.
FIG. 1 is a flowchart showing a pore detection method according to the present embodiment.
There are the following knowledge about pores. The pore is a visible part of the follicle, and sebum is secreted from the sebaceous glands present in the upper part of the follicle at the back of the pore. Acne bacteria, one of the resident bacteria, prefers sebum and produces porphyrin by metabolism, so porphyrin is present in the pores.
Porphyrin is known as a substance that emits light when excited by ultraviolet rays. Of course, the substance that exists in pores and emits light by ultraviolet rays can be other than porphyrin. Based on the knowledge about such pores and porphyrins, the present inventors have made clues to the existence of substances that emit light by ultraviolet rays such as porphyrins. The idea that the detection accuracy of the pores can be improved is obtained, and the present invention has been conceived.
Furthermore, in addition to the open pores that are not clogged with normal square plugs, the pores are clogged with clogging due to sebum and keratin, and there are also open pores. It is known that when pores are clogged, sebum accumulates and acne bacteria grow to cause skin diseases such as acne. There is also a substance that emits light by ultraviolet rays, such as porphyrin, in the back of such clogged pores.

  As shown in FIG. 1, the pore detection method according to the present embodiment (hereinafter sometimes referred to as the present method) includes a candidate specifying step (S11) and a determination step (S12).

In the step (S11), a position in the light and dark skin image corresponding to the light emission position in the ultraviolet skin image captured by irradiating ultraviolet rays is specified as a candidate position.
Here, the “ultraviolet skin image” means an image obtained by imaging human skin in an environment irradiated with ultraviolet rays. For example, an “ultraviolet skin image” can be obtained by capturing an image of the skin with a digital camera in an environment in which human skin is irradiated with ultraviolet light. As the digital camera at this time, a general digital camera that captures visible light is used.
The “light emission position in the ultraviolet skin image” is the position of the light emission location in the ultraviolet skin image. For example, a position with high luminance in the image is the light emission position. That is, a location where a substance that emits light by ultraviolet rays, such as porphyrin, is the emission position.
The “light / dark skin image” refers to an image obtained by imaging human skin by a method different from the “ultraviolet skin image” and includes at least light / dark information in the visible light region. Examples of the “light / dark skin image” include a general color image, a polarization image captured using a polarization filter, a gray scale image, a black and white image, and the like.
The ultraviolet skin image and the light and dark skin image should include at least the same area of the skin of the same person, and it is only necessary that the same position of the actual skin can be associated between the images. For example, the ultraviolet skin image and the light and dark skin image may be images obtained by shifting the same region of the same person's skin with the same angle of view and shifting the time. In addition, the two images may be images obtained by capturing the skin of the same person from different shooting directions. In this case, the correspondence between the same positions of the actual skin between the images according to the two shooting directions. It is only necessary that the attachment is made.

In step (S12), based on the pixel value of the candidate position in the light and dark skin image, it is determined whether or not the candidate position is a pore position.
Here, the “pixel value” means an index value indicating the brightness of a certain pixel, such as a luminance value and a brightness value. The pixel value of the candidate position that is the basis for the determination in the step (S12) may be a pixel value of one pixel, or may be a pixel value of the darkest or brightest pixel among a plurality of pixels. Alternatively, it may be an average value of pixel values of a plurality of pixels.
If the pore is not clogged with a square plug, the position of the pore in the light and dark skin image should be dark because there is a depression. On the other hand, in the case of a pore such as a clogged pore due to a square plug, the position of the pore in the light and dark skin image may not be dark.
Therefore, for example, when the pixel value of the candidate position specified in the step (S11) in the light and dark skin image indicates a predetermined darkness, the candidate position can be determined as an open pore position.
As described above, in this embodiment, since the position where the luminescent substance such as porphyrin exists and the position indicating the predetermined darkness is determined as the pore position, the pore can be detected with high accuracy. it can.

  By the way, depending on the shooting environment and the shooting site of the light and dark skin image, there is a possibility that portions other than the pores appear dark as well as the pores due to undulations. Therefore, whether or not the candidate position is an open pore position is determined by referring to not only the pixel value of the candidate position but also the surrounding pixel values, based on whether or not the candidate position is darker than the surrounding area. . Specifically, in the determination step (S12), when the difference between the pixel value at a predetermined distance from the candidate position in the light and dark skin image and the pixel value at the candidate position exceeds a predetermined threshold, The candidate position may be determined as the pore position. In this way, the position of the luminescent substance due to ultraviolet rays such as porphyrin is determined as a pore position where a place darker than the surrounding, that is, a place with a depression, is opened, so the pore position with high accuracy. Can be specified.

The pore detection method according to the present embodiment described above can be executed by, for example, the pore detection device illustrated in FIGS. 2 and 3.
FIG. 2 is a diagram conceptually illustrating a hardware configuration example of the pore detection apparatus 10 according to the present embodiment.
The pore detection device 10 is a so-called computer, and includes, for example, a CPU (Central Processing Unit) 11, a memory 12, an input / output interface (I / F) 13, a communication unit 14, and the like that are connected to each other via a bus. The number of hardware elements that form the pore detection device 10 is not limited, and these hardware elements may be collectively referred to as an information processing circuit. Moreover, the pore detection apparatus 10 may include hardware elements not shown in FIG. 2, and the hardware configuration is not limited.

The CPU 11 may be configured by an application specific integrated circuit (ASIC), a DSP (Digital Signal Processor), a GPU (Graphics Processing Unit), or the like in addition to a general CPU.
The memory 12 is a RAM (Random Access Memory), a ROM (Read Only Memory), or an auxiliary storage device (such as a hard disk).
The input / output I / F 13 can be connected to user interface devices such as the output device 15 and the input device 16. The output device 15 is at least one of a device such as an LCD (Liquid Crystal Display) or a CRT (Cathode Ray Tube) display, a device that displays a screen corresponding to drawing data processed by the CPU 11, a printing device, and the like. The input device 16 is a device that receives an input of a user operation such as a keyboard and a mouse. The output device 15 and the input device 16 may be integrated and realized as a touch panel.
The communication unit 14 performs communication via a communication network with other computers, exchange of signals with other devices, and the like. A portable recording medium or the like can be connected to the communication unit 14. The communication unit 14 may be connected to a camera (not shown) that captures one or both of an ultraviolet skin image and a light and dark skin image.
Further, the pore detection device 10 may be a device incorporating a camera.

FIG. 3 is a diagram conceptually illustrating a processing configuration example of the pore detection apparatus 10 according to the present embodiment.
The pore detection apparatus 10 includes an acquisition unit 21, a candidate identification unit 22, and a determination unit 23. These processing modules are software elements, and are realized, for example, by executing a program stored in the memory 12 by the CPU 11. This program is installed from, for example, a portable recording medium such as a CD (Compact Disc) or a memory card or another computer on the network via the input / output I / F 13 or the communication unit 14, and stored in the memory 12. May be.

The acquisition unit 21 acquires an ultraviolet skin image captured under ultraviolet irradiation and a light and dark skin image captured by a method different from the ultraviolet skin image. The meanings of “ultraviolet skin image” and “light / dark skin image” are as described above.
The acquisition unit 21 can acquire an ultraviolet skin image and a light and dark skin image from an external computer, device, portable recording medium, or the like via the input / output I / F 13 or the communication unit 14. Moreover, when the acquisition part 21 is implement | achieved as a camera, the acquisition part 21 may image and produce | generate both the images.
The candidate specifying unit 22 specifies a position in the light and dark skin image corresponding to the light emission position in the ultraviolet skin image acquired by the acquiring unit 21 as a candidate position. The processing content of the candidate specifying unit 22 may be the same as in the above-described candidate specifying step (S11).
The determination unit 23 determines whether or not the candidate position is set as a pore position based on the pixel value of the candidate position in the light and dark skin image. The processing content of the determination part 23 may be the same as that of the above-mentioned determination process (S12).

Hereinafter, a preferred embodiment of the present invention (hereinafter referred to as the present embodiment) will be described in more detail.
The pore detection method according to the present embodiment is implemented as part of skin diagnosis using an image obtained by photographing a skin by executing a software program on the skin diagnosis system 30 shown in FIG. When the pore detection method according to the present embodiment is executed, the whole or part of the skin diagnostic system 30 can also be called a pore detection device.

FIG. 4 is a conceptual diagram showing a hardware configuration of the skin diagnostic system 30 according to the present embodiment.
The skin diagnosis system 30 includes an imaging device 31, a projector 32, an information processing device 33, a touch panel 34, a porphyrin image imaging device 35, and the like.
The imaging device 31 images a skin surface such as the face of the subject S. The imaging device 31 is provided with an imaging polarization filter 36 for selecting light to be imaged on the plane of polarization.
The projector 32 projects photographing light onto the skin surface such as the face of the subject S. The projector 32 is provided with a projection polarizing filter 37 for selecting a polarization plane of light to be projected onto the subject S.
The information processing device 33 performs various calculations for skin diagnosis using an image of the skin surface photographed by the imaging device 31. The information processing device 33 is a general-purpose computer device such as a personal computer. The information processing device 33 includes an external storage device (memory 12) for storing various programs and captured skin surface images, as illustrated in FIG. 2, and a printer (not shown) for printing diagnosis results. An output device 15) is provided.
The touch panel 34 functions as a display unit that displays a photographed skin surface image, a result of skin diagnosis, and the like, and also serves as an input unit for a user of the skin diagnosis system 30 to input various commands to the information processing device 33. Function.
The porphyrin image capturing device 35 captures a porphyrin image. For example, the porphyrin image capturing device 35 is a camera that irradiates ultraviolet rays and acquires a visible light region.
The “porphyrin image” is the same as the above-described ultraviolet skin image.

  Various known skin diagnoses are executed by starting and executing the program in the information processing apparatus 33 of the skin diagnosis system 30 as described above, and the pore detection method according to the present embodiment is executed. The information processing device 33 corresponds to the pore detection device 10 described above.

Hereinafter, the pore detection method according to the present embodiment will be described with reference to FIGS. 5 and 6.
FIG. 5 is a flowchart showing the pore detection method according to the present embodiment.
In the present embodiment, the pore display program P0 is executed in the information processing apparatus 33, whereby the following pore detection method is executed.

  In the present embodiment, the pore detection step (S51) is first executed. Subsequently, a porphyrin position detection step (S52) is executed. Then, it is determined whether or not the position of the detected porphyrin is a pore, and when it is determined that it is a pore, the position is added to the pore position detected in the pore detection step (S51), and a new one is added. A pore adding step (S53) for setting the pore position is executed. Finally, a pore position writing process step (S54) for displaying the determined pore position so as to overlap the skin image is executed.

Hereinafter, each step described above will be described in more detail.
In the pore detection step (S51), a so-called S-polarized image (also referred to as an S-S polarized image) obtained by matching the plane of polarization of the projection polarizing filter 37 with the plane of polarization of the imaging polarizing filter 36 is used as an original image (also referred to as an SS polarized image). A surface reflected light image) is processed, and a pore image is processed by binarizing a difference between an image subjected to low-pass filtering such as a Gaussian filter and an original image. By this process, the dark part of the skin pore image of the subject S can be detected as a pore.
The surface reflected light image as the original image may be acquired as follows. Surface reflection that further suppresses noise other than the surface reflected light component by taking the difference between the S-polarized image and the polarized image obtained by projecting S-polarized light and imaging P-polarized light (SP-polarized image) An optical image may be acquired. Further, a surface reflected light image is obtained by further using a PS polarized image obtained by projecting P polarized light and imaging S polarized light, and a PP polarized image obtained by projecting P polarized light and imaging P polarized light. May be obtained. Of course, the method for acquiring the surface reflected light image is not limited.
The pore detection step (S51) identifies the pore position in the skin surface reflected light image by detecting the dark part of the pore image as a pore. That is, the pore detection step (S51) can also be called a position specifying step. In the pore detection step (S51), a conventional pore detection method can be used.

In the porphyrin position detection step (S52), the position of the porphyrin is detected by using the UV (Ultra Violet) image (porphyrin image) picked up by the porphyrin image pickup device 35 and using the fact that the porphyrin emits light by ultraviolet irradiation. . Specifically, porphyrin is detected in a UV image (porphyrin image), and for each group of porphyrins, the center position, that is, the midpoint of two orthogonal directions, or the position of the center of gravity is obtained and set as the position of the porphyrin. Porphyrin in the UV image is detected as a pixel having a luminance equal to or higher than a predetermined threshold, and the position of the porphyrin is specified as the position of the target pixel. By executing this porphyrin position detection step (S52), a porphyrin position table indicating a plurality of porphyrin positions is formed inside the information processing apparatus 4. The porphyrin positions stored in this porphyrin position table correspond to the above-mentioned candidate positions.
Here, the UV image used in step (S52) and the original image (surface reflected light image) used in step (S51) are images obtained by imaging the skin of subject S in different shooting directions. The alignment is performed between the porphyrin position in the UV image and the pore position in the original image. In this alignment, the same parts of the actual skin are set to the same position.
Hereinafter, the porphyrin position detected in the step (S52) will be described as being aligned in the same coordinate system as the pore position detected in the step (S51).

Using the porphyrin position detected in this porphyrin position detection step (S52), the pore addition step (S53) is executed.
FIG. 6 is a detailed flowchart of the pore addition step. Here, an example is shown in which the pore addition step (S53) is realized as a part of predefined processing of the pore display program P0, a so-called subroutine, but as a series of steps with other steps shown in FIG. It goes without saying that the following steps shown may be realized.

In the pore addition step (S53), a porphyrin position pointer initialization processing step (pointer initialization step) (S61) is first executed. In the pointer initialization step (S61), an address indicating the head of the porphyrin position table is set in the pointer.
Subsequently, a pore determination step (S62) is executed with respect to the target porphyrin position pointed to by the pointer. In the pore determination step (S62), the luminance between the target porphyrin position in the original image (surface reflected light image) used in the step (S51) and the position away from the target porphyrin position by a predetermined distance δ is determined in the vertical and horizontal directions of the image. Compare in direction. When the brightness at a position separated by δ is bright (large) with a difference greater than or equal to a predetermined threshold TH, that is, when the brightness at the target porphyrin position is dark (small) with a difference greater than or equal to the threshold TH from the periphery, the target porphyrin position is It determines with it being a pore (S62; Y), and when that is not right, it determines with the object porphyrin position not being a pore (S62; N). Here, the image used for the luminance comparison may be the original image (surface reflected light image) as described above, or may be a pore image (binarized image) obtained from the original image. Good.
Here, the predetermined distance δ is appropriately determined as a distance at which the difference in luminance becomes clear, but in the present embodiment, it is desirable that the predetermined distance δ is set in a range of 0.05 to 1 mm on the actual skin in consideration of resolution and the like.

When it is determined that the target porphyrin position is a pore (S62; Y), a pore position addition step (S63) is executed, and then a pointer update step (S64) is executed. On the other hand, when it is determined that the target porphyrin position is not a pore (S62; N), the pointer update step (S64) is executed without executing step (S63).
FIG. 7 is a diagram showing pseudo code for realizing the pore determining step (S62) and the pore position adding step (S63).
In FIG. 7, I (x, y) is a function that returns the pixel value of the porphyrin position (x, y) in the original image (surface reflected light image). The pixel value I (x, y) at the porphyrin position (x, y) and the pixel values I (x + δ, y), I (x) in the surrounding four directions are determined by the determination code (IF sentence) that realizes the pore determination step (S62). , Y + δ), I (x−δ, y), and I (x, y−δ), respectively. When all the pixel values in the surrounding four directions are larger (brighter) than the pixel value at the porphyrin position than the predetermined threshold value TH, the code for realizing the pore position adding step (S63) is executed.
“Add (x, y) to Keana” is a code that realizes the pore position addition step (S63), and adds a new pore position (x, y) to the pore position table. In addition, in this pore position addition code, if the same pore position (x, y) as the pore position already detected in the pore detection step (S51) is specified, the pore position is not added. Also good.
In the pointer update step (S64), an address indicating the next porphyrin position in the porphyrin position table is set in the pointer.

The pore determining step (S62), the pore position adding step (S63), and the pointer updating step (S64) are executed until there is no address indicating the porphyrin position to be set in the pointer in the porphyrin position table.
If there is no address indicating the porphyrin position to be set in the pointer in the porphyrin position table (S65; Y), that is, if it is determined whether or not all candidate positions are pores, the subroutine is exited, and FIG. The indicated pore position writing step (S54) is executed.

  Here, in the pore position adding step (S63), even if the pore position is identified in the pore detection step (S51), the pore position that does not match the porphyrin position may be excluded from the final pore position. . In this way, erroneous detection of pores based on the surface reflected light image can be compensated by the porphyrin position.

In the pore position writing step (S54), based on the pore position specified in the pore detection step (S51) and the pore position added in the pore addition step (S53), the subject S used in step (S51) Information indicating that pores are present on the original image (surface reflected light image) of the skin is written. Specifically, a red circle surrounding the position corresponding to the pore position is written in the surface reflected light image. For example, the red circle is drawn in a size corresponding to a diameter of about 0.1 mm to 1 mm in actual skin.
Hereinafter, the image generated in the pore position writing step (S54) may be referred to as a pore detection image.

FIG. 8 is a diagram conceptually showing the flow of pore detection in the present embodiment. As shown in FIG. 8, in this embodiment, pores are detected based on the surface reflected light image of the skin of the subject S, and the pores are detected based on the porphyrin position displayed in the UV image of the skin of the subject S. By detecting and combining them, the total pore position is obtained.
Thereby, according to this embodiment, since the pore which cannot be detected from the original image (surface reflected light image) can be supplemented by the position of the porphyrin, the detection of the pore can be performed more reliably.

The information processing apparatus 33 that executes the above-described pore detection method includes a position specifying unit (not shown) and a determining unit (not shown) in addition to the acquisition unit 21, the candidate specifying unit 22, and the determination unit 23 shown in FIG. It can also be said that it is further provided.
In the present embodiment, the acquisition unit 21 acquires a UV image (porphyrin image) and an original image.
The candidate specifying unit 22 executes a porphyrin position detection step (S52).
The determination unit 23 executes step (S62). That is, when the difference between the pixel value at a position that is a predetermined distance δ away from the candidate position in the original image (surface reflected light image) and the pixel value at the candidate position exceeds the predetermined threshold TH, The position is determined as the pore position.
The position specifying unit executes a pore detection step (S51).
The determining unit executes a pore adding step (S53). That is, the determination unit determines both the pore position determined by the determination unit 23 and the pore position specified by the position specifying unit as the pore position.

However, this embodiment is not limited to the above-mentioned content.
For example, a step of removing the pore position detected in the pore detection step (S51) from the porphyrin position table may be added before the pore addition step (S53) in order to avoid duplicate determination. That is, a position other than the pore position specified in the pore detection step (S51) (position specifying step) may be specified as a porphyrin position (candidate position) to be added in the pore addition step (S53). Similarly, the candidate specifying unit 22 of the information processing device 33 may specify a candidate position other than the pore position specified by the position specifying unit.

Further, the pore detection step (S51) may be omitted. In this case, a corresponding position in the original image is obtained from the position of the porphyrin on the porphyrin image (UV image) and set as a candidate position, and the brightness with respect to the periphery of the candidate position is determined in the original image. The position may be a pore position.
Further, in any step, a plurality of pixels can be handled as a group of areas instead of a single pixel as a pore position and a candidate position. For example, in the porphyrin position detection step (S52), porphyrin on the UV image (porphyrin image) is detected, and one or more positions are specified for each group of porphyrins. Then, a pore determining step (S62) is executed for each specified position, and if it is determined that a predetermined number (for example, one) or more positions included in a group of porphyrins are pores, The position in the group may be added as a pore position (pore position addition step (S63)).

In addition, the image showing the pore position may not be the original image (surface reflected light image) used in the pore detection step (S51), and other images (color image, gray) in which the skin of the target site of the subject S is captured. Scale image, monochrome image).
Further, the wavelength and polarization plane used for imaging the porphyrin image (UV image) for detecting the position of the porphyrin can also depend on various methods different from the above-described embodiment including the detection algorithm.
In addition, the pore position information may be indicated in a tabular form, for example, in which the pore position information is directly indicated as a coordinate value in addition to the pore position being superimposed on the image. In addition, in the present embodiment, new processing is added, the order of each step and the specific processing content are changed, and the method of displaying the obtained results is changed or obtained within a range that does not impair the essence of the invention. You can also change how you use the results.

For example, the pore detection method according to the present embodiment is based on the determined pore position in addition to the pore detection method flow shown in FIG. 5 or instead of the pore position writing step (S54). The method may further include generating pore information regarding the number, area, or shape.
In this case, the area and shape of the pores can be determined as follows, for example.
In the porphyrin position detection step (S52), porphyrin on the UV image (porphyrin image) is detected, and the area (number of pixels) or shape information is obtained for each group of porphyrins. The shape information only needs to be stored. What is necessary is just to calculate circularity etc. as shape information. The group of porphyrins can be detected by a well-known method for detecting a series of pixels indicating porphyrins. Then, in the pore position adding step (S63), it is only necessary to maintain the area or shape together with the position determined as the pore. As another method, the pore area (number of pixels) or shape information may be obtained by comparing the pixel values of the porphyrin position in the original image or the pore image and the periphery thereof.
Moreover, although the direction to determine was made into 4 directions, it is preferable to carry out in the range of 2-8 directions.
In this case, the predetermined distance δ or the threshold value TH (predetermined threshold value) used in the pore determination step (S62) may be determined according to the generated pore information. For example, when the number of pores is generated as pore information, the predetermined distance δ may be set smaller and the predetermined threshold TH may be set smaller than when the area or shape is generated as pore information. That is, the plurality of types of pore information (for example, the number and area) may be generated using a plurality of different predetermined distances δ or predetermined thresholds TH.
In this case, the information processing device 33 may further include a generation unit (not shown) that generates pore information related to the number, area, or shape of the pores based on the pore position determined by the determination unit 23. And the production | generation part can produce | generate multiple types of pore information about each of the several predetermined distance or predetermined threshold value which mutually differ, for example.

Moreover, in the pore determination step (S62) in the pore addition step (S53), it may be determined whether or not it is a pore as follows. That is, based on one or more of the amount, area and shape of the porphyrin (light emission position) on the UV image (porphyrin image) detected in the porphyrin position detection step (S52), the process (S51) It may be determined whether or not the target porphyrin position in the original image (surface reflected light image) used in is a pore. Similarly, in the determination step (S12), the candidate position specified in the candidate specification step (S11) is determined based on one or more of the light emission amount, area, and shape of the light emission position in the ultraviolet skin image. You may determine whether it is set as a pore position.
In this case, the determination unit 23 of the pore detection device 10 or the determination unit 23 of the information processing device 33 may perform such a pore determination step (S62) or a determination step (S12).
Here, the light emission amount of porphyrin (light emission position) on the porphyrin image can be calculated from the pixel value at the corresponding position on the porphyrin image, and may be, for example, luminance. In addition, the area and shape of the porphyrin can be determined by the number of pixels of the porphyrin pixel (light emission position) on the porphyrin image and how it is arranged.
Such pore determination based on at least one of the amount, area, and shape of the porphyrin (emission position) on the porphyrin image may be executed in addition to the determination of the above-described embodiment, It may be executed instead of the determination. In the latter case, the pore detection step (S51) may not be executed. In addition, the light emitting area is calculated | required for every pore, and various evaluation can also be performed by this.

When the pore detection method according to the above-described embodiment was applied to the nose and cheeks of a subject, the following detection results were obtained.
As for the nose, 84 pores are detected based on the S-polarized image (surface reflected light image), and finally 585 pieces of pores are detected by considering the porphyrin position detected from the porphyrin image (UV image). A pore was detected. According to this result, 501 pores are supplemented by porphyrin positions. Since the nose is undulating, it was verified that pores that could not be detected only by the surface reflected light image could be detected by the porphyrin position.
For the cheek, 176 pores were detected based on the S-polarized image (surface reflected light image), and finally 585 pieces of pores were detected by considering the porphyrin position detected from the porphyrin image (UV image). A pore was detected. According to this result, 409 pores are supplemented by porphyrin positions. The cheek part has less undulations than the nose part, but there are pores that can not be detected only by the surface reflected light image even if there are few such undulations, and such pores can be detected by the porphyrin position Verified.
In this embodiment, in addition to the number of pores, the circularity (0.881) and the average area (95.76) are generated as pore information.
The resolution of the S-polarized image (surface reflected light image) and porphyrin image (UV image) was 3648 × 5472 pixels (1 pixel = 35 μm).

  Part or all of the above-described embodiments and modifications may be specified as follows. However, the above-described embodiment and modification examples are not limited to the following description.

<1> a candidate specifying step of specifying a position in a light and dark skin image corresponding to a light emission position in an ultraviolet skin image captured under ultraviolet irradiation as a candidate position;
A determination step of determining whether or not the candidate position is a pore position based on a pixel value of the candidate position in the light and dark skin image;
A method for detecting pores.

<2> In the determination step, if the difference between the pixel value at a predetermined distance from the candidate position in the light and dark skin image and the pixel value at the candidate position exceeds a predetermined threshold, the candidate position is determined as a pore. To determine the position,
<1> The pore detection method according to <1>.
<3> A step of generating pore information regarding the number, area, or shape of pores based on the pore position determined in the determination step,
Further including
The plurality of types of pore information is generated using a plurality of the predetermined distances or the predetermined threshold values different from each other.
The pore detection method according to <2>.
<4> a position specifying step for specifying a pore position based on the skin surface reflected light image;
A determination step of determining both the pore position determined in the determination step and the pore position specified in the position specification step as a pore position;
The pore detection method according to any one of <1> to <3>, further including:
<5> The position specifying step is performed prior to the candidate specifying step, and in the candidate specifying step, a position other than the pore position specified in the position specifying step is specified as the candidate position.
<4> The pore detection method according to <4>.
<6> In the determination step, based on at least one of a light emission amount, an area, and a shape of the light emission position in the ultraviolet skin image, it is determined whether or not the candidate position is a pore position.
The pore detection method according to any one of <1> to <5>.
<7> An acquisition unit that acquires an ultraviolet skin image captured under ultraviolet irradiation and a light and dark skin image captured by a method different from the ultraviolet skin image;
Candidate specifying means for specifying a position in the light and dark skin image corresponding to a light emission position in the ultraviolet skin image acquired by the acquiring means as a candidate position;
Determination means for determining whether or not the candidate position is a pore position based on a pixel value of the candidate position in the light and dark skin image;
A pore detection apparatus comprising:
<8> When the difference between a pixel value at a position away from the candidate position in the light and dark skin image by a predetermined distance and a pixel value at the candidate position exceeds a predetermined threshold, the determination unit determines the candidate position as a pore. To determine the position,
<7> The pore detection apparatus according to <7>.
<9> Generating means for generating pore information related to the number, area or shape of pores based on the pore position determined by the determining means,
Further comprising
The generating means generates a plurality of types of pore information using each of a plurality of the predetermined distances or the predetermined threshold values that are different from each other.
<8> The pore detection apparatus according to <8>.
<10> Position specifying means for specifying the pore position based on the surface reflected light image of the skin;
Determining means for determining both the pore position determined by the determining means and the pore position specified by the position specifying means as pore positions;
The pore detection device according to any one of <7> to <9>, further comprising:
<11> The candidate specifying unit specifies a position other than the pore position specified by the position specifying unit as the candidate position.
<10> The pore detection apparatus according to <10>.
<12> The determination unit determines whether or not the candidate position is a pore position based on at least one of a light emission amount, an area, and a shape of the light emission position in the ultraviolet skin image.
The pore detection device according to any one of <7> to <9>.

10 pore detection device 11 CPU
12 Memory 13 Input / output I / F
14 Communication Unit 15 Output Device 16 Input Device 21 Acquisition Unit 22 Candidate Identification Unit 23 Determination Unit 30 Skin Diagnosis System 31 Imaging Device 32 Projector 33 Information Processing Device 34 Touch Panel 35 Porphyrin Imaging Device 36 Imaging Polarization Filter 37 Projection Polarization Filter

Claims (7)

A candidate specifying step for specifying a position in a light and dark skin image corresponding to a light emitting position in an ultraviolet skin image captured under ultraviolet irradiation as a candidate position;
A determination step of determining whether or not the candidate position is a pore position based on a pixel value of the candidate position in the light and dark skin image;
A method for detecting pores. In the determination step, when a difference between a pixel value at a predetermined distance from the candidate position in the light and dark skin image and a pixel value at the candidate position exceeds a predetermined threshold, the candidate position is determined as a pore position. To
The pore detection method according to claim 1. Generating pore information on the number, area, or shape of pores based on the pore positions determined in the determination step;
Further including
The plurality of types of pore information is generated using a plurality of the predetermined distances or the predetermined threshold values different from each other.
The pore detection method according to claim 2. A position specifying step for specifying a pore position based on the surface reflection light image of the skin;
A determination step of determining both the pore position determined in the determination step and the pore position specified in the position specification step as a pore position;
The pore detection method according to any one of claims 1 to 3, further comprising: Performing the position specifying step prior to the candidate specifying step;
In the candidate specifying step, the position other than the pore position specified in the position specifying step is specified as the candidate position.
The pore detection method according to claim 4. In the determination step, based on at least one of the light emission amount, area and shape of the light emission position in the ultraviolet skin image, it is determined whether or not the candidate position is a pore position.
The pore detection method according to any one of claims 1 to 5. An acquisition means for acquiring an ultraviolet skin image captured under ultraviolet irradiation and a light and dark skin image captured by a method different from the ultraviolet skin image;
Candidate specifying means for specifying a position in the light and dark skin image corresponding to a light emission position in the ultraviolet skin image acquired by the acquiring means as a candidate position;
Determination means for determining whether or not the candidate position is a pore position based on a pixel value of the candidate position in the light and dark skin image;
A pore detection apparatus comprising:

Images