Skin inflammation detection method, device, electronic device and storage medium

Abstract

The present application belongs to the technical field of skin detection, and discloses a skin inflammation detection method, device, electronic device and storage medium, which obtains an LED white light image of the skin under LED white light irradiation, and obtains polarization images of the skin at four polarization angles under red laser, green laser and blue laser irradiation respectively; according to the polarization image, obtains polarization images corresponding to lasers of various colors; determines the abnormal area of ​​the skin according to the LED white light image and the polarization image; extracts the image of the abnormal area in each polarization image to obtain the corresponding abnormal area image; calculates the similarity between the abnormal area images corresponding to lasers of various colors; and judges the abnormal area as a pigmentation area or an inflammation area according to the similarity; thereby effectively distinguishing pigmentation areas from inflammation areas, and improving the accuracy of skin inflammation area detection.

Description

Technical Field

[0001] The present application relates to the technical field of skin detection, and in particular to a method, device, electronic device and storage medium for detecting skin inflammation. Background Art

[0002] At present, in the field of automatic skin health detection technology, LED white light is usually used to illuminate the skin of the subject to collect skin images, and then the skin health status is identified through image recognition methods. Among them, whether there is an inflammatory area on the skin is a relatively important detection item. However, in the LED white light image, the pigmentation area and the inflammatory area have a high similarity. When using the LED white light image to identify the inflammatory area, the pigmentation area is often misjudged as the inflammatory area, resulting in inaccurate detection results. Summary of the invention

[0003] The purpose of this application is to provide a skin inflammation detection method, device, electronic device and storage medium, which can effectively distinguish between pigmentation areas and inflammation areas and improve the accuracy of skin inflammation area detection.

[0004] In a first aspect, the present application provides a method for detecting skin inflammation, comprising the steps of:

[0005] A1. Acquire an LED white light image of the skin under LED white light, and acquire polarization images of the skin at four polarization angles under red laser, green laser, and blue laser, respectively;

[0006] A2. According to the polarization image, obtain polarization images corresponding to various color lasers;

[0007] A3. Determine the abnormal area of ​​the skin according to the LED white light image and the polarization image;

[0008] A4. Extracting the image of the abnormal region from each of the polarization images to obtain a corresponding abnormal region image;

[0009] A5. Calculate the similarity between the two images of the abnormal area corresponding to the various color lasers;

[0010] A6. According to the similarity, the abnormal area is judged to be a pigmentation area or an inflammation area.

[0011] In fact, compared with ordinary images, polarized images contain richer skin surface information and superficial skin information. The polarized images of pigmented areas under red, green and blue laser irradiation are more similar, while the polarized images of inflammatory areas under red, green and blue laser irradiation are very different. Based on this, the abnormal area is judged as a pigmented area or an inflammatory area according to the similarity between the images of the abnormal area under the irradiation of the three colors of laser, which can effectively distinguish between pigmented areas and inflammatory areas and improve the accuracy of detection of skin inflammatory areas.

[0012] Preferably, the wavelength of the red laser is 620 nm -690nm;

[0013] The wavelength of the green laser is 500 nm -560nm;

[0014] The wavelength of the blue laser is 380 nm-490 nm.

[0015] Preferably, step A1 comprises:

[0016] Polarization images of the skin at four polarization angles were acquired when the skin was illuminated by red laser, green laser, and blue laser at an illumination angle of 30°-50°, respectively.

[0017] The polarization imaging effect at this illumination angle is better, which can ensure that the obtained polarization image is clear.

[0018] Preferably, the four polarization angles include 0°, 45°, 90° and 135°;

[0019] The LED white light image, the polarization image and the degree of polarization image have the same pixel size;

[0020] Step A2 includes:

[0021] Calculating the polarization degree of each pixel point of the polarization degree image according to the light intensity of each pixel point of the polarization image at the four polarization angles corresponding to the same color laser;

[0022] The polarization degree image is generated according to the polarization degree of each pixel point.

[0023] Preferably, step A3 comprises:

[0024] Identifying a preliminary normal area of ​​the skin according to the LED white light image;

[0025] calculating an average polarization degree of the preliminary normal area in the polarization degree image;

[0026] The polarization degree of each pixel in the polarization degree image is compared with the average polarization degree of the preliminary normal area to determine the abnormal area of ​​the skin.

[0027] By comparing the average polarization degree with the preliminary normal area to determine the abnormal area of ​​the skin, it can be ensured that the abnormal area of ​​the skin with different shades can be identified relatively accurately, further ensuring the accuracy of the test results.

[0028] Preferably, the step of comparing the polarization degree of each pixel in the polarization degree image with the average polarization degree of the preliminary normal area to determine the abnormal area of ​​the skin includes:

[0029] Calculating the polarization degree difference between the polarization degree of each pixel in the polarization degree image and the average polarization degree of the preliminary normal area;

[0030] Determine the type of each pixel point according to the polarization degree difference; the type is a normal pixel point or an abnormal pixel point;

[0031] The abnormal area of ​​the skin is determined according to the position of the abnormal pixel point.

[0032] Preferably, step A6 comprises:

[0033] Calculating an average or weighted average of the similarities to obtain an equivalent similarity;

[0034] The equivalent similarity is compared with a preset similarity threshold to determine whether the abnormal area is a pigmentation area or an inflammation area.

[0035] In a second aspect, the present application provides a skin inflammation detection device, comprising:

[0036] A first acquisition module is used to acquire an LED white light image of the skin under the irradiation of LED white light, and acquire polarization images of the skin at four polarization angles under the irradiation of red laser, green laser and blue laser respectively;

[0037] A second acquisition module is used to acquire polarization degree images corresponding to lasers of various colors according to the polarization image;

[0038] A positioning module, used for determining the abnormal area of ​​the skin according to the LED white light image and the polarization image;

[0039] An extraction module, used for extracting the image of the abnormal region from each of the polarization degree images to obtain a corresponding abnormal region image;

[0040] A calculation module, used for calculating the similarity between the images of the abnormal region corresponding to the lasers of various colors;

[0041] A judgment module is used to judge whether the abnormal area is a pigmentation area or an inflammation area according to the similarity.

[0042] In fact, compared with ordinary images, polarized images contain richer skin surface information and superficial skin information. The polarized images of pigmented areas under red, green and blue laser irradiation are more similar, while the polarized images of inflammatory areas under red, green and blue laser irradiation are very different. Based on this, the abnormal area is judged as a pigmented area or an inflammatory area according to the similarity between the images of the abnormal area under the irradiation of the three colors of laser, which can effectively distinguish between pigmented areas and inflammatory areas and improve the accuracy of detection of skin inflammatory areas.

[0043] In a third aspect, the present application provides an electronic device comprising a processor and a memory, wherein the memory stores a computer program executable by the processor, and when the processor executes the computer program, it runs the steps in the skin inflammation detection method described above.

[0044] In a fourth aspect, a storage medium stores a computer program thereon, wherein when the computer program is executed by a processor, the steps in the skin inflammation detection method described above are performed.

[0045] Beneficial effects:

[0046] The skin inflammation detection method, device, electronic device and storage medium provided in the present application obtain an LED white light image of the skin under LED white light irradiation, and obtain polarization images of the skin at four polarization angles under red laser, green laser and blue laser irradiation respectively; according to the polarization image, obtain polarization images corresponding to lasers of various colors; determine the abnormal area of ​​the skin according to the LED white light image and the polarization image; extract the image of the abnormal area in each polarization image to obtain the corresponding abnormal area image; calculate the similarity between the abnormal area images corresponding to lasers of various colors; judge the abnormal area as a pigmentation area or an inflammation area according to the similarity; thereby effectively distinguishing pigmentation areas from inflammation areas, and improving the accuracy of skin inflammation area detection.

[0047] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent from the description, or may be understood through implementation of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A flow chart of a method for detecting skin inflammation provided in an embodiment of the present application.

[0049] Figure 2 This is a schematic diagram of the structure of a skin inflammation detection device provided in an embodiment of the present application.

[0050] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0052] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0053] Please refer to Figure 1 , Figure 1 A method for detecting skin inflammation in some embodiments of the present application comprises the steps of:

[0054] A1. Acquire an LED white light image of the skin under LED white light, and acquire polarization images of the skin at four polarization angles under red laser, green laser, and blue laser, respectively;

[0055] A2. Obtain polarization images corresponding to various color lasers based on the polarization images;

[0056] A3. Determine the abnormal area of ​​the skin based on the LED white light image and the polarization image;

[0057] A4. Extracting an image of an abnormal region from each polarization image to obtain a corresponding abnormal region image;

[0058] A5. Calculate the similarity between the two images of abnormal regions corresponding to lasers of different colors;

[0059] A6. Based on the similarity, the abnormal area is judged to be a pigmentation area or an inflammation area.

[0060] In fact, compared with ordinary images, polarized images contain richer skin surface information and superficial skin information. The polarized images of pigmented areas under red, green and blue laser irradiation are more similar, while the polarized images of inflammatory areas under red, green and blue laser irradiation are very different. Based on this, the abnormal area is judged as a pigmented area or an inflammatory area according to the similarity between the images of the abnormal area under the irradiation of the three colors of laser, which can effectively distinguish between pigmented areas and inflammatory areas and improve the accuracy of detection of skin inflammatory areas.

[0061] In order to ensure the accuracy of the final determination result, the wavelength of the red laser is 620 nm-690 nm; the wavelength of the green laser is 500 nm-560 nm; and the wavelength of the blue laser is 380 nm-490 nm. For example, in some embodiments, the wavelength of the red laser is 660 nm, the wavelength of the green laser is 520 nm, and the wavelength of the blue laser is 488 nm.

[0062] Preferably, step A1 comprises:

[0063] Polarization images of the skin at four polarization angles were acquired when the skin was illuminated by red laser, green laser, and blue laser at an illumination angle of 30°-50°, respectively.

[0064] The illumination angle refers to the angle between the incident light and the illuminated plane. The polarization imaging effect is better at this illumination angle, and the obtained polarization image can be ensured to be clear. For example, in some embodiments, the illumination angle is 40°.

[0065] Among them, for the same color of irradiation laser, polarization images with four different polarization angles are collected in order to obtain a polarization degree image of the skin, and the polarization angle value can be set according to actual needs; according to different polarization angle values, the corresponding polarization degree can be calculated according to the light intensity of the same pixel point at four different polarization angles (the calculation method is the existing technology and will not be described in detail here), thereby obtaining the corresponding polarization degree image.

[0066] For example, in this embodiment, the four polarization angles include 0°, 45°, 90°, and 135°;

[0067] The LED white light image, polarization image, and degree of polarization image have the same pixel size;

[0068] Thus, step A2 comprises:

[0069] Calculate the polarization degree of each pixel in the polarization image according to the light intensity of each pixel in the polarization image at four polarization angles corresponding to the same color laser;

[0070] A polarization degree image is generated based on the polarization degree of each pixel.

[0071] The polarization degree of each pixel in the polarization degree image can be calculated using the following formula:

[0072] ;

[0073] ;

[0074] ;

[0075] in, is the polarization image at Row, No. The polarization degree of the pixel point of the column, To represent the polarization image located at Row, No. The Stokes parameter of the total light intensity of the pixel in the column, To represent the polarization image located at Row, No. The Stokes parameter of the difference between the horizontal polarization intensity and the vertical polarization intensity of the pixel point of the column, To represent the polarization image located at Row, No. The Stokes parameter of the difference between the 45° polarized light intensity and the 135° polarized light intensity of the pixel point of the column, To represent the polarization image located at Row, No. The Stokes parameter of the difference between the right-handed circularly polarized light intensity and the left-handed circularly polarized light intensity of the pixel point of the column, , , , The polarization images of 0° (i.e., the polarization images corresponding to the polarization angle of 0°), 45° (i.e., the polarization images corresponding to the polarization angle of 45°), 90° (i.e., the polarization images corresponding to the polarization angle of 90°), and 135° (i.e., the polarization images corresponding to the polarization angle of 135°) corresponding to the same color laser are located in the first Row, No. The light intensity of the pixel in the column, is the polarization image at Row, No. The right-handed circularly polarized light intensity of the pixel point of the column, is the polarization image at Row, No. The right-handed circularly polarized light intensity of the laser beam after passing through 1 / 4 of the glass slide (based on , , and Calculated, the calculation method is prior art and will not be described in detail here);

[0076] For the three colors of laser light, the above process is used to obtain corresponding polarization images respectively, thereby obtaining three polarization images.

[0077] The degree of polarization of each pixel can be converted into a pixel value according to a preset conversion rule, and the pixel value of each pixel can be assigned a value according to the calculation result, so as to obtain a degree of polarization image. For example, the degree of polarization of each pixel can be directly used as a pixel value, and since the degree of polarization is a value of 0-1, a degree of polarization image with normalized pixel values ​​can be obtained; or for example, the degree of polarization of each pixel can be multiplied by 255 to obtain a gray value, and the gray value can be used as a pixel value to obtain a degree of polarization image (a grayscale image); but it is not limited to this.

[0078] In some embodiments, step A3 comprises:

[0079] Identify the preliminary normal area of ​​the skin based on the LED white light image;

[0080] Calculate the average polarization degree of the preliminary normal area in the polarization degree image;

[0081] The polarization degree of each pixel in the polarization degree image is compared with the average polarization degree of the preliminary normal area to determine the abnormal area of ​​the skin.

[0082] By comparing the average polarization degree with the preliminary normal area to determine the abnormal area of ​​the skin, it can be ensured that the abnormal area of ​​the skin with different shades can be identified relatively accurately, further ensuring the accuracy of the test results.

[0083] When the preliminary normal area of ​​the skin is identified based on the LED white light image, the existing image recognition method can be used for identification, which is not limited here.

[0084] The step of calculating the average polarization degree of the preliminary normal area in the polarization degree image includes: extracting the polarization degree (hereinafter referred to as the first polarization degree) of the pixel points in the polarization degree image located in the preliminary normal area, calculating the average value of the first polarization degree, and obtaining the average polarization degree of the preliminary normal area. The average polarization degree reflects the reflection level of the normal skin of the detected person to polarized light, and the average polarization degree will vary with the actual skin color and skin condition of the detected person (such as non-pathological conditions such as smoothness and dryness). By comparing the polarization degree of each pixel point in the polarization degree image with the average polarization degree of the preliminary normal area, it is determined whether each pixel point is a point in the abnormal area, which is conducive to eliminating the adverse effects of factors such as the actual skin color and skin condition of the detected person on the accuracy of the detection result.

[0085] In some embodiments, the step of comparing the polarization degree of each pixel in the polarization degree image with the average polarization degree of the preliminary normal area to determine the abnormal area of ​​the skin includes:

[0086] Calculate the polarization degree difference between the polarization degree of each pixel in the polarization degree image and the average polarization degree of the preliminary normal area;

[0087] Determine the type of each pixel point according to the polarization degree difference; the type is a normal pixel point or an abnormal pixel point;

[0088] The abnormal area of ​​the skin is determined according to the location of the abnormal pixel points.

[0089] In one experiment, the normal area, pigmentation area and inflammation area of ​​the skin of a subject were irradiated by 488nm laser, 520nm laser and 660nm laser respectively. The following Tables 1 to 3 are the polarization data (average polarization value of each area) obtained by irradiating the normal area, pigmentation area and inflammation area of ​​the skin of the subject by 488nm laser, 520nm laser and 660nm laser respectively:

[0090] Table 1 Polarization degree of each region under 488nm laser irradiation

[0091] area Normal area Pigmented areas Inflammatory areas Degree of Polarization 28.636% 61.786% 48.442%

[0092] Table 2 Polarization degree of each region under 520nm laser irradiation

[0093] area Normal area Pigmented areas Inflammatory areas Degree of Polarization 22.288% 68.438% 61.084%

[0094] Table 3 Polarization degree of each region under 660nm laser irradiation

[0095] area Normal area Pigmented areas Inflammatory areas Degree of Polarization 21.091% 50.081% 24.889%

[0096] It can be seen from Tables 1 to 3 that, for the pigmented area, under the irradiation of the three color lasers, its polarization degree is significantly greater than that of the normal area, and for the inflammation area, under the irradiation of the blue laser and the green laser, its polarization degree is significantly greater than that of the normal area, and under the irradiation of the red laser, its polarization degree is close to that of the normal area. Therefore, when judging the type of each pixel point according to the polarization degree difference, it can be judged whether the polarization degree difference corresponding to the blue laser and the green laser of the pixel point is greater than the preset first polarization degree difference threshold (which can be set according to actual needs, such as 10%, 15%, etc., but not limited to this), if so, the pixel point is judged to be an abnormal pixel point, otherwise the pixel point is judged to be a normal pixel point; or, the average value or weighted average value of the polarization degree difference corresponding to the three color lasers is calculated to obtain a comprehensive polarization degree difference, and it is judged whether the comprehensive polarization degree difference is greater than the preset second polarization degree difference threshold (which can be set according to actual needs, such as 10%, but not limited to this), if so, the pixel point is judged to be an abnormal pixel point, otherwise the pixel point is judged to be a normal pixel point; or

[0097] The step of determining the abnormal area of ​​the skin according to the position of the abnormal pixel point includes:

[0098] If there are isolated abnormal pixels, the isolated abnormal pixels are reclassified as normal pixels; an isolated abnormal pixel means that all the pixels adjacent to the abnormal pixel are normal pixels;

[0099] If there are isolated normal pixels, the isolated normal pixels are reclassified as abnormal pixels; an isolated normal pixel means that all the pixels adjacent to the normal pixel are abnormal pixels;

[0100] Obtaining each connected area of ​​abnormal pixel distribution (the method of obtaining the connected area is a prior art and will not be described in detail here);

[0101] If there is a hole in the connected area whose area is smaller than the preset area threshold (set according to actual needs) (the hole is actually one of the normal pixel distribution areas), then fill the hole as part of the corresponding connected area;

[0102] The set of the above-mentioned connected areas is regarded as the abnormal area of ​​the skin.

[0103] In some other embodiments, the step of comparing the polarization degree of each pixel in the polarization degree image with the average polarization degree of the preliminary normal area to determine the abnormal area of ​​the skin includes:

[0104] Calculate the ratio of the polarization degree of each pixel in the polarization degree image to the average polarization degree of the preliminary normal area;

[0105] The type of each pixel is determined according to the ratio; the type is a normal pixel or an abnormal pixel;

[0106] Determine the abnormal area of ​​the skin based on the location of the abnormal pixel points (refer to the specific steps in the previous article).

[0107] Among them, when judging the type of each pixel point according to the ratio, it can be judged whether the ratios corresponding to the blue laser and the green laser of the pixel point are both greater than a preset first ratio threshold (which can be set according to actual needs). If so, the pixel point is judged to be an abnormal pixel point, otherwise the pixel point is judged to be a normal pixel point; or, the average value or weighted average value of the ratios corresponding to the three color lasers is calculated to obtain a comprehensive ratio, and it is judged whether the comprehensive ratio is greater than a preset second ratio threshold (which can be set according to actual needs). If so, the pixel point is judged to be an abnormal pixel point, otherwise the pixel point is judged to be a normal pixel point.

[0108] In step A5, the structural similarity (SSIM) between each pair of abnormal region images corresponding to various color lasers can be calculated, but it is not limited to structural similarity. The calculation method of structural similarity is prior art and will not be described in detail here.

[0109] In the above-mentioned experiment, for the pigmentation area, the structural similarity between the abnormal area images corresponding to various color lasers is shown in Table 4 below. For the inflammation area, the structural similarity between the abnormal area images corresponding to various color lasers is shown in Table 5 below:

[0110] Table 4 Structural similarity comparison of pigment deposition areas

[0111] combination 488nm vs. 520nm 488nm vs 660nm 520nm vs 660nm SSIM 0.1373 0.1643 0.1452

[0112] Table 5 Comparison of structural similarities in inflammatory regions

[0113] combination 488nm vs. 520nm 488nm vs 660nm 520nm vs 660nm SSIM 0.0590 0.0742 0.0682

[0114] It can be seen from Tables 4 and 5 that for the pigmentation area, the structural similarity between the two images of the abnormal area corresponding to the three colors of laser is relatively large, and for the inflammation area, the structural similarity between the two images of the abnormal area corresponding to the three colors of laser is relatively small. Based on this characteristic, the pigmentation area and the inflammation area can be effectively distinguished according to the structural similarity.

[0115] In some embodiments, step A6 comprises:

[0116] Calculate the average or weighted average of each similarity to obtain equivalent similarity;

[0117] The equivalent similarity is compared with the preset similarity threshold to determine whether the abnormal area is a pigmentation area or an inflammatory area.

[0118] Specifically, if the ratio of equivalent similarity is greater than a preset similarity threshold, the abnormal area is determined to be a pigmentation area, otherwise, the abnormal area is determined to be an inflammation area.

[0119] The preset similarity threshold may be set according to actual needs. For example, if the similarity is structural similarity, it may be set to 0.1, but is not limited thereto.

[0120] As can be seen from the above, the skin inflammation detection method obtains an LED white light image of the skin under LED white light irradiation, and obtains polarization images of the skin at four polarization angles under red laser, green laser, and blue laser irradiation, respectively; according to the polarization image, polarization images corresponding to lasers of various colors are obtained; the abnormal area of ​​the skin is determined according to the LED white light image and the polarization image; the image of the abnormal area is extracted from each polarization image to obtain the corresponding abnormal area image; the similarity between the abnormal area images corresponding to lasers of various colors is calculated; and the abnormal area is judged to be a pigmentation area or an inflammation area according to the similarity; thereby, the pigmentation area and the inflammation area can be effectively distinguished, and the accuracy of skin inflammation area detection is improved.

[0121] refer to Figure 2 , the present application provides a skin inflammation detection device, comprising:

[0122] The first acquisition module 1 is used to acquire an LED white light image of the skin under the irradiation of LED white light, and acquire polarization images of the skin at four polarization angles under the irradiation of red laser, green laser and blue laser respectively;

[0123] The second acquisition module 2 is used to acquire polarization degree images corresponding to lasers of various colors according to the polarization image;

[0124] A positioning module 3 is used to determine the abnormal area of ​​the skin according to the LED white light image and the polarization image;

[0125] An extraction module 4 is used to extract an image of an abnormal region from each polarization image to obtain a corresponding abnormal region image;

[0126] A calculation module 5, used for calculating the similarity between each pair of abnormal area images corresponding to lasers of various colors;

[0127] The judgment module 6 is used to judge whether the abnormal area is a pigmentation area or an inflammation area according to the similarity.

[0128] In fact, compared with ordinary images, polarized images contain richer skin surface information and superficial skin information. The polarized images of pigmented areas under red, green and blue laser irradiation are more similar, while the polarized images of inflammatory areas under red, green and blue laser irradiation are very different. Based on this, the abnormal area is judged as a pigmented area or an inflammatory area according to the similarity between the images of the abnormal area under the irradiation of the three colors of laser, which can effectively distinguish between pigmented areas and inflammatory areas and improve the accuracy of detection of skin inflammatory areas.

[0129] To ensure the accuracy of the final determination result, the wavelength of the red laser is 620nm-690nm; the wavelength of the green laser is 500nm-560nm; the wavelength of the blue laser is 380nm-490nm; for example, in some embodiments, the wavelength of the red laser is 660nm, the wavelength of the green laser is 520nm, and the wavelength of the blue laser is 488nm.

[0130] Preferably, when acquiring polarization images of the skin at four polarization angles irradiated by red laser, green laser and blue laser, respectively, the first acquisition module 1 executes:

[0131] Polarization images of the skin at four polarization angles were acquired when the skin was illuminated by red laser, green laser, and blue laser at an illumination angle of 30°-50°, respectively.

[0132] The illumination angle refers to the angle between the incident light and the illuminated plane. The polarization imaging effect is better at this illumination angle, and the obtained polarization image can be ensured to be clear. For example, in some embodiments, the illumination angle is 40°.

[0133] Among them, for the same color of irradiation laser, polarization images with four different polarization angles are collected in order to obtain a polarization degree image of the skin, and the polarization angle value can be set according to actual needs; according to different polarization angle values, the corresponding polarization degree can be calculated according to the light intensity of the same pixel point at four different polarization angles (the calculation method is the existing technology and will not be described in detail here), thereby obtaining the corresponding polarization degree image.

[0134] For example, in this embodiment, the four polarization angles include 0°, 45°, 90°, and 135°;

[0135] The LED white light image, polarization image, and degree of polarization image have the same pixel size;

[0136] Therefore, the second acquisition module 2 is used to perform the following when acquiring polarization degree images corresponding to lasers of various colors according to the polarization image:

[0137] Calculate the polarization degree of each pixel in the polarization image according to the light intensity of each pixel in the polarization image at four polarization angles corresponding to the same color laser;

[0138] A polarization degree image is generated based on the polarization degree of each pixel.

[0139] The polarization degree of each pixel in the polarization degree image can be calculated using the following formula:

[0140] ;

[0141] ;

[0142] ;

[0143] in, is the polarization image at Row, No. The polarization degree of the pixel point of the column, To represent the polarization image located at Row, No. Stokes parameter of the total light intensity of the pixel points of the column, To represent the polarization image located at Row, No. The Stokes parameter of the difference between the horizontal polarization intensity and the vertical polarization intensity of the pixel point of the column, To represent the polarization image located at Row, No. The Stokes parameter of the difference between the 45° polarized light intensity and the 135° polarized light intensity of the pixel point of the column, To represent the polarization image located at Row, No. The Stokes parameter of the difference between the right-handed circularly polarized light intensity and the left-handed circularly polarized light intensity of the pixel point of the column, , , , The polarization images of 0° (i.e., the polarization images corresponding to the polarization angle of 0°), 45° (i.e., the polarization images corresponding to the polarization angle of 45°), 90° (i.e., the polarization images corresponding to the polarization angle of 90°), and 135° (i.e., the polarization images corresponding to the polarization angle of 135°) corresponding to the same color laser are located in the first Row, No. The light intensity of the pixel in the column, is the polarization image at Row, No. The right-handed circularly polarized light intensity of the pixel point of the column, is the polarization image at Row, No. The right-handed circularly polarized light intensity of the laser beam after passing through 1 / 4 of the glass slide (based on , , and Calculated, the calculation method is prior art and will not be described in detail here);

[0144] For the three colors of laser light, the above process is used to obtain corresponding polarization images respectively, thereby obtaining three polarization images.

[0145] The degree of polarization of each pixel can be converted into a pixel value according to a preset conversion rule, and the pixel value of each pixel can be assigned a value according to the calculation result, so as to obtain a degree of polarization image. For example, the degree of polarization of each pixel can be directly used as a pixel value, and since the degree of polarization is a value of 0-1, a degree of polarization image with normalized pixel values ​​can be obtained; or for example, the degree of polarization of each pixel can be multiplied by 255 to obtain a gray value, and the gray value can be used as a pixel value to obtain a degree of polarization image (a grayscale image); but it is not limited to this.

[0146] In some embodiments, the positioning module 3 is used to perform the following when determining the abnormal area of ​​the skin according to the LED white light image and the polarization image:

[0147] Identify the preliminary normal area of ​​the skin based on the LED white light image;

[0148] Calculate the average polarization degree of the preliminary normal area in the polarization degree image;

[0149] The polarization degree of each pixel in the polarization degree image is compared with the average polarization degree of the preliminary normal area to determine the abnormal area of ​​the skin.

[0150] By comparing the average polarization degree with the preliminary normal area to determine the abnormal area of ​​the skin, it can be ensured that the abnormal area of ​​the skin with different shades can be identified relatively accurately, further ensuring the accuracy of the test results.

[0151] When the preliminary normal area of ​​the skin is identified based on the LED white light image, the existing image recognition method can be used for identification, which is not limited here.

[0152] Wherein, when calculating the average polarization degree of the preliminary normal area in the polarization degree image, the positioning module 3 performs: extracting the polarization degree (hereinafter referred to as the first polarization degree) of the pixel points in the polarization degree image located in the preliminary normal area, calculating the average value of the first polarization degree, and obtaining the average polarization degree of the preliminary normal area. The average polarization degree reflects the reflection level of the normal skin of the detected person to polarized light, and the average polarization degree will vary with the actual skin color and skin condition of the detected person (such as non-pathological conditions such as smoothness and dryness). By comparing the polarization degree of each pixel point in the polarization degree image with the average polarization degree of the preliminary normal area, it is determined whether each pixel point is a point in the abnormal area, which is conducive to eliminating the adverse effects of factors such as the actual skin color and skin condition of the detected person on the accuracy of the detection result.

[0153] In some embodiments, when comparing the polarization degree of each pixel in the polarization degree image with the average polarization degree of the preliminary normal area to determine the abnormal area of ​​the skin, the positioning module 3 performs:

[0154] Calculate the polarization degree difference between the polarization degree of each pixel in the polarization degree image and the average polarization degree of the preliminary normal area;

[0155] Determine the type of each pixel point according to the polarization degree difference; the type is a normal pixel point or an abnormal pixel point;

[0156] The abnormal area of ​​the skin is determined according to the location of the abnormal pixel points.

[0157] It can be seen from Tables 1 to 3 that, for the pigmented area, under the irradiation of the three color lasers, its polarization degree is significantly greater than that of the normal area, and for the inflammation area, under the irradiation of the blue laser and the green laser, its polarization degree is significantly greater than that of the normal area, and under the irradiation of the red laser, its polarization degree is close to that of the normal area. Therefore, when judging the type of each pixel point according to the polarization degree difference, it can be judged whether the polarization degree difference corresponding to the blue laser and the green laser of the pixel point is greater than the preset first polarization degree difference threshold (which can be set according to actual needs, such as 10%, 15%, etc., but not limited to this), if so, the pixel point is judged to be an abnormal pixel point, otherwise the pixel point is judged to be a normal pixel point; or, the average value or weighted average value of the polarization degree difference corresponding to the three color lasers is calculated to obtain a comprehensive polarization degree difference, and it is judged whether the comprehensive polarization degree difference is greater than the preset second polarization degree difference threshold (which can be set according to actual needs, such as 10%, but not limited to this), if so, the pixel point is judged to be an abnormal pixel point, otherwise the pixel point is judged to be a normal pixel point; or

[0158] Wherein, when the positioning module 3 determines the abnormal area of ​​the skin according to the position of the abnormal pixel point, it executes:

[0159] If there are isolated abnormal pixels, the isolated abnormal pixels are reclassified as normal pixels; an isolated abnormal pixel means that all the pixels adjacent to the abnormal pixel are normal pixels;

[0160] If there are isolated normal pixels, the isolated normal pixels are reclassified as abnormal pixels; an isolated normal pixel means that all the pixels adjacent to the normal pixel are abnormal pixels;

[0161] Obtaining each connected area of ​​abnormal pixel distribution (the method of obtaining the connected area is a prior art and will not be described in detail here);

[0162] If there is a hole in the connected area whose area is smaller than the preset area threshold (set according to actual needs) (the hole is actually one of the normal pixel distribution areas), then fill the hole as part of the corresponding connected area;

[0163] The set of the above-mentioned connected areas is regarded as the abnormal area of ​​the skin.

[0164] In some other embodiments, the positioning module 3 is used to perform the following when comparing the polarization degree of each pixel in the polarization degree image with the average polarization degree of the preliminary normal area to determine the abnormal area of ​​the skin:

[0165] Calculate the ratio of the polarization degree of each pixel in the polarization degree image to the average polarization degree of the preliminary normal area;

[0166] The type of each pixel is determined according to the ratio; the type is a normal pixel or an abnormal pixel;

[0167] The abnormal area of ​​the skin is determined based on the location of the abnormal pixel points (refer to the specific process in the previous article).

[0168] Among them, when judging the type of each pixel point according to the ratio, it can be judged whether the ratios corresponding to the blue laser and the green laser of the pixel point are both greater than a preset first ratio threshold (which can be set according to actual needs). If so, the pixel point is judged to be an abnormal pixel point, otherwise the pixel point is judged to be a normal pixel point; or, the average value or weighted average value of the ratios corresponding to the three color lasers is calculated to obtain a comprehensive ratio, and it is judged whether the comprehensive ratio is greater than a preset second ratio threshold (which can be set according to actual needs). If so, the pixel point is judged to be an abnormal pixel point, otherwise the pixel point is judged to be a normal pixel point.

[0169] The calculation module 5 can calculate the structural similarity (SSIM) between each pair of abnormal region images corresponding to various color lasers, but is not limited to structural similarity. The calculation method of structural similarity is prior art and will not be described in detail here.

[0170] It can be seen from Tables 4 and 5 that for the pigmentation area, the structural similarity between the two images of the abnormal area corresponding to the three colors of lasers is relatively large, and for the inflammatory deposition area, the structural similarity between the two images of the abnormal area corresponding to the three colors of lasers is relatively small. Based on this characteristic, the pigmentation area and the inflammatory area can be effectively distinguished according to the structural similarity.

[0171] In some embodiments, the judgment module 6 is used to judge, when judging the abnormal area as a pigmentation area or an inflammation area according to the similarity:

[0172] Calculate the average or weighted average of each similarity to obtain equivalent similarity;

[0173] The equivalent similarity is compared with the preset similarity threshold to determine whether the abnormal area is a pigmentation area or an inflammatory area.

[0174] Specifically, if the ratio of equivalent similarity is greater than a preset similarity threshold, the abnormal area is determined to be a pigmentation area, otherwise, the abnormal area is determined to be an inflammation area.

[0175] The preset similarity threshold may be set according to actual needs. For example, if the similarity is structural similarity, it may be set to 0.1, but is not limited thereto.

[0176] As can be seen from the above, the skin inflammation detection device obtains an LED white light image of the skin under LED white light irradiation, and obtains polarization images of the skin at four polarization angles under red laser, green laser, and blue laser irradiation, respectively; according to the polarization image, polarization images corresponding to lasers of various colors are obtained; the abnormal area of ​​the skin is determined according to the LED white light image and the polarization image; the image of the abnormal area is extracted from each polarization image to obtain the corresponding abnormal area image; the similarity between the abnormal area images corresponding to lasers of various colors is calculated; and the abnormal area is judged to be a pigmentation area or an inflammation area according to the similarity; thereby, the pigmentation area and the inflammation area can be effectively distinguished, thereby improving the accuracy of skin inflammation area detection.

[0177] Please refer to Figure 3 , Figure 3A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The present application provides an electronic device, including: a processor 301 and a memory 302. The processor 301 and the memory 302 are interconnected and communicate with each other through a communication bus 303 and / or other forms of connection mechanisms (not shown). The memory 302 stores a computer program executable by the processor 301. When the electronic device is running, the processor 301 executes the computer program to execute the skin inflammation detection method in any optional implementation of the above embodiment to achieve the following functions: obtaining an LED white light image of the skin under LED white light irradiation, and obtaining polarization images of the skin at four polarization angles under red laser, green laser, and blue laser irradiation, respectively; according to the polarization image, obtaining polarization images corresponding to lasers of various colors; determining abnormal areas of the skin according to the LED white light image and the polarization image; extracting images of abnormal areas in each polarization image to obtain corresponding abnormal area images; calculating the similarity between the abnormal area images corresponding to lasers of various colors; judging whether the abnormal area is a pigmentation area or an inflammation area according to the similarity.

[0178] An embodiment of the present application provides a storage medium having a computer program stored thereon. When the computer program is executed by a processor, the skin inflammation detection method in any optional implementation of the above-mentioned embodiment is executed to achieve the following functions: obtaining an LED white light image of the skin under LED white light irradiation, and obtaining polarization images of the skin at four polarization angles under red laser, green laser, and blue laser irradiation, respectively; obtaining polarization images corresponding to lasers of various colors based on the polarization images; determining abnormal areas of the skin based on the LED white light images and the polarization images; extracting images of abnormal areas in each polarization image to obtain corresponding abnormal area images; calculating the similarity between each of the abnormal area images corresponding to lasers of various colors; and judging whether the abnormal area is a pigmentation area or an inflammation area based on the similarity. Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0179] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0180] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0181] Furthermore, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0182] In this document, relational terms such as first and second, etc. are used merely to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0183] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting skin inflammation, It is characterized in that Includes steps: A1. Acquire an LED white light image of the skin under LED white light, and acquire polarization images of the skin at four polarization angles under red laser, green laser, and blue laser, respectively; A2. According to the polarization image, obtain polarization images corresponding to various color lasers; A3. Determine the abnormal area of ​​the skin according to the LED white light image and the polarization image; A4. Extracting the image of the abnormal region from each of the polarization images to obtain a corresponding abnormal region image; A5. Calculate the similarity between the two images of the abnormal area corresponding to the various color lasers; A6. judging the abnormal area to be a pigmentation area or an inflammation area according to the similarity; Step A3 includes: Identifying a preliminary normal area of ​​the skin according to the LED white light image; calculating an average polarization degree of the preliminary normal area in the polarization degree image; Comparing the polarization degree of each pixel in the polarization degree image with the average polarization degree of the preliminary normal area to determine the abnormal area of ​​the skin; The step of comparing the polarization degree of each pixel in the polarization degree image with the average polarization degree of the preliminary normal area to determine the abnormal area of ​​the skin comprises: Calculating the polarization degree difference between the polarization degree of each pixel in the polarization degree image and the average polarization degree of the preliminary normal area; Determine the type of each pixel point according to the polarization degree difference; the type is a normal pixel point or an abnormal pixel point; The abnormal area of ​​the skin is determined according to the position of the abnormal pixel point.

2. The method for detecting skin inflammation according to claim 1, It is characterized in that The wavelength of the red laser is 620nm-690nm; The wavelength of the green laser is 500 nm-560 nm; The wavelength of the blue laser is 380 nm-490 nm.

3. The method for detecting skin inflammation according to claim 1, It is characterized in that Step A1 includes: Polarization images of the skin at four polarization angles were acquired when the skin was illuminated by red laser, green laser, and blue laser at an illumination angle of 30°-50°, respectively.

4. The method for detecting skin inflammation according to claim 1, It is characterized in that The four polarization angles include 0°, 45°, 90° and 135°; The LED white light image, the polarization image and the degree of polarization image have the same pixel size; Step A2 includes: Calculating the polarization degree of each pixel point of the polarization degree image according to the light intensity of each pixel point of the polarization image at the four polarization angles corresponding to the same color laser; The polarization degree image is generated according to the polarization degree of each pixel point.

5. The method for detecting skin inflammation according to claim 1, It is characterized in that Step A6 includes: Calculating an average or weighted average of the similarities to obtain an equivalent similarity; The equivalent similarity is compared with a preset similarity threshold to determine whether the abnormal area is a pigmentation area or an inflammation area.

6. A skin inflammation detection device, It is characterized in that include: A first acquisition module is used to acquire an LED white light image of the skin under the irradiation of LED white light, and acquire polarization images of the skin at four polarization angles under the irradiation of red laser, green laser and blue laser respectively; A second acquisition module is used to acquire polarization degree images corresponding to lasers of various colors according to the polarization image; A positioning module, used for determining the abnormal area of ​​the skin according to the LED white light image and the polarization image; An extraction module, used for extracting the image of the abnormal region from each of the polarization degree images to obtain a corresponding abnormal region image; A calculation module, used for calculating the similarity between the images of the abnormal region corresponding to the lasers of various colors; A judgment module, used for judging whether the abnormal area is a pigmentation area or an inflammation area according to the similarity; When determining the abnormal area of ​​the skin according to the LED white light image and the polarization image, the positioning module executes: Identifying a preliminary normal area of ​​the skin according to the LED white light image; calculating an average polarization degree of the preliminary normal area in the polarization degree image; Comparing the polarization degree of each pixel in the polarization degree image with the average polarization degree of the preliminary normal area to determine the abnormal area of ​​the skin; When comparing the polarization degree of each pixel in the polarization degree image with the average polarization degree of the preliminary normal area to determine the abnormal area of ​​the skin, the positioning module executes: Calculating the polarization degree difference between the polarization degree of each pixel in the polarization degree image and the average polarization degree of the preliminary normal area; Determine the type of each pixel according to the polarization degree difference; The type is a normal pixel or an abnormal pixel; The abnormal area of ​​the skin is determined according to the position of the abnormal pixel point.

7. An electronic device, It is characterized in that It comprises a processor and a memory, wherein the memory stores a computer program executable by the processor, and when the processor executes the computer program, it runs the steps in the skin inflammation detection method as described in any one of claims 1 to 5.

8. A storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the skin inflammation detection method according to any one of claims 1 to 5 are executed.

Images