Skin detection method and skin detection device

Abstract

The invention provides a skin detection method and a skin detection device. The skin detection method comprises the following steps: (a) collecting skin spectrum information of skin to be detected; and (b) comparing the spectral distribution characteristic data of the collected skin spectral information with preset spectral distribution characteristic data to obtain a comparison result.

Description

Technical Field

[0001] The present invention relates to the field of optical detection technology, and in particular to a skin detection method and a skin detection device. Background Art

[0002] The skin is one of the most important organs of the human body. It has many functions. First of all, the skin separates the outside world from the internal organs and acts as the first barrier of the human body. The skin blocks the invasion of external bacteria, dust, etc. through the thick stratum corneum, ensuring a safe environment in the human body. Secondly, the skin is also an important secretory and excretory tissue. There are many sweat glands in the skin, through which sweat secretion and sebum excretion are carried out. Thirdly, the skin also has the function of regulating body temperature. It can be said that the skin is closely related to the health of the human body. The skin is vulnerable to many external and internal factors. External factors include ultraviolet radiation (from sunlight), environmental pollution, wind, high temperature or infrared radiation, low humidity, low humidity, harmful surfactants, abrasives, etc. Internal factors include chronological age and other chemical changes from the skin. Whether external or internal, these factors can lead to visible signs of skin aging, such as wrinkles and other forms of roughness (including increased pore size, flaking and wrinkles), as well as other histological changes related to skin aging or damage. Furthermore, the water content of the stratum corneum has a significant impact on the appearance, elasticity, texture and dryness of the skin, and also has a significant impact on the absorption of drugs and other molecules into and through the skin.

[0003] In recent years, with the improvement of living standards, people pay more and more attention to skin care, but the causes of skin problems are relatively complex. Due to the lack of professional equipment, it is difficult to understand the physiological characteristics of one's own skin, and therefore it is difficult to take the correct skin care method according to the characteristics of one's own skin. In response to the above problems, people have designed a skin detection device, which, as the name suggests, is a device that can identify and detect the skin.

[0004] Skin problems, which are usually not distinguishable by the human eye, can also be detected through spectral imaging, because different skin changes will cause changes in spectral energy distribution, but there will also be effects caused by makeup. Therefore, in the prior art, the detection of skin problems has the problem of low detection precision and accuracy. Summary of the invention

[0005] A major advantage of the present invention is that it provides a skin detection method and a skin detection device, which detect and identify skin problems through spectral imaging. Since the reflectance spectrum of each pixel corresponding to the spectral imaging corresponding to different skin problems is different, judgment is made based on different characteristic spectra, which is beneficial to improving the accuracy and precision of detection.

[0006] Another advantage of the present invention is to provide a skin detection method and a skin detection device, wherein the skin detection method uses set bands corresponding to different resolutions, so that by improving the resolution, more details can be obtained and the accuracy is higher.

[0007] Another advantage of the present invention is that it provides a skin detection method and a skin detection device, wherein the skin detection method uses skin spectrum information of skin color to determine changes in spectrum information of a location, which is generally compared with a preset value. When it is lower than or higher than a set threshold, it is determined that the location has a corresponding skin problem. The accuracy of judging skin problems based on this method is high.

[0008] According to one aspect of the present invention, a skin detection method of the present invention that can achieve the aforementioned objects and other objects and advantages includes:

[0009] (a) collecting skin spectrum information of the skin to be detected; and

[0010] (b) comparing the collected spectral distribution characteristic data of the skin spectral information with preset spectral distribution characteristic data to obtain a comparison result.

[0011] According to an embodiment of the present invention, in step (b), comparing the current spectral distribution characteristic data with conventional spectral distribution characteristic data to obtain a comparison result may include:

[0012] (b.1) determining the preset spectral distribution characteristic data corresponding to the target skin; and

[0013] (b.2) Comparing the current spectral distribution characteristic data with the preset spectral characteristic distribution data, determining the health state corresponding to the current spectral distribution characteristic data in the preset spectral characteristic distribution data, so as to obtain the comparison result.

[0014] According to one embodiment of the present invention, a calibration step is further included before step (a), wherein the step is: using a standard reference object with known reflectivity at different wavelengths to calibrate the spectral response of the imaging channel at different wavelengths to obtain true reflection characteristics.

[0015] According to one embodiment of the present invention, the step (b) further includes a preset step, wherein the step is: pre-storing the spectral energy distribution of the skin state under different scenes for multiple times, and using the spectral energy distribution as preset spectral characteristic distribution data of the corresponding skin.

[0016] According to an embodiment of the present invention, the step (b) further comprises the following steps: determining that the predicted state of the target skin is a preset state according to the spectral energy variation trend of the target skin, and then issuing a prompt warning.

[0017] According to one embodiment of the present invention, the step (a) further comprises the following steps:

[0018] (a.1) obtaining a light energy response signal matrix and a standard spectrum output by a spectral image sensor chip of the spectral imaging device;

[0019] (a.2) determining a basic element restoration function and a response signal vector of the basic element restoration function based on a light energy response signal matrix, wherein the basic element restoration function uses a predetermined pixel value of the spectral image sensor chip and a pixel value in its vicinity to restore a spectral image value of a corresponding predetermined channel;

[0020] (a.3) obtaining a recovery tensor, wherein the product of the recovery tensor and the response signal vector is equal to the output of the basic element recovery function based on the response signal vector; and

[0021] (a.4) Obtaining a restored spectral image based on the product of the restored tensor and the response signal vector.

[0022] According to one embodiment of the present invention, in the step (a.1), the light energy response signal matrix is ​​represented as a matrix B including two dimensions of image width w and image height h, the dimension of the standard spectrum is l, and is set so that the distance between the product of the true value tensor of the spectral image received by the spectral imaging device and the standard spectrum and the spectral image tensor to be restored is minimized.

[0023] According to one embodiment of the present invention, the standard spectrum is represented as s, and the channel standard spectrum corresponding to the kth channel of the standard spectrum is represented as sk, so that: xk→O(i,j)sk, wherein xk is the spectral image value of the kth channel of a spectral pixel, O(i,j) is the true value tensor of the spectral curve of a spectral pixel, and → indicates that the Euclidean distance between the tensors is the minimum.

[0024] According to one embodiment of the present invention, the step (a) further comprises the following steps:

[0025] Acquiring transmission spectrum data of the spectral imaging chip and output signal data of the spectral imaging chip;

[0026] Acquire local transmission spectrum data of the transmission spectrum data and local output signal data of the output signal data based on the pixel used to reconstruct the spectrum image;

[0027] Inputting the local output signal data into an attention model to obtain attention local data; and

[0028] The local transmission spectrum data, the local output signal data and the attention local data are input into a neural network model to obtain the pixels for reconstructing a spectral image.

[0029] According to one embodiment of the present invention, in the step (a), obtaining local transmission spectrum data of the transmission spectrum data and local output signal data of the output signal data based on the pixel used to reconstruct the spectral image includes: based on the position of the pixel used to reconstruct the spectral image, obtaining local transmission spectrum data of the transmission spectrum data and local output signal data of the output signal data in an area near the position with a side length of a predetermined number of pixels.

[0030] According to one embodiment of the present invention, in the step (a), inputting the local output signal data into the attention model to obtain the attention local data includes: dividing the local output signal data into a plurality of predetermined areas, each predetermined area including output signal data corresponding to a plurality of pixels of the spectral imaging chip; and performing matrix multiplication for each of the predetermined areas to obtain the attention local data.

[0031] According to another aspect of the present application, the present application further provides a skin detection device, comprising:

[0032] Spectral imaging equipment; and

[0033] A processor is communicatively connected to the spectral imaging device, the spectral imaging device is used to collect spectral image information of the skin to be tested, wherein the processor compares the collected spectral distribution characteristic data of the skin spectral information with preset spectral distribution characteristic data to obtain a comparison result.

[0034] Further objects and advantages of the present invention will be fully apparent from an understanding of the following description and the accompanying drawings.

[0035] These and other objects, features and advantages of the present invention will be more fully understood from the following detailed description and accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. In the accompanying drawings, unless otherwise specified, the same reference numerals are used to represent the same components. Among them:

[0037] Figure 1 is a schematic diagram of a spectral image according to a first preferred embodiment of the present invention.

[0038] Figure 2is a schematic diagram of a continuous spectrum image according to a first preferred embodiment of the present invention.

[0039] Figure 3 is a schematic configuration diagram of a spectral imaging device according to a first preferred embodiment of the present invention.

[0040] Figure 4 is a schematic diagram of a spectral imaging chip for calculating a reconstructed spectrum according to a first preferred embodiment of the present invention.

[0041] Figure 5A It is a schematic diagram of the surface structure of a CMOS chip in the prior art.

[0042] Figure 5B It is a schematic diagram of the surface structure of a spectral imaging chip according to a preferred embodiment of the present invention.

[0043] Figure 6 It is a schematic diagram of different pixels corresponding to different structures of the spectral imaging device according to the first preferred embodiment of the present invention.

[0044] Figure 7 is a schematic diagram of facial skin detection according to the first preferred embodiment of the present invention.

[0045] Figure 8 According to the first preferred embodiment of the present invention, the facial spectrum image can correspond various positions of the face to different pixels, and obtain the spectrum information corresponding to the different positions.

[0046] Fig. 9 4 is a spectrum distribution diagram of healthy skin preset on the face in the skin detection method according to the first preferred embodiment of the present invention.

[0047] Fig.10 It is a distribution diagram of the actual corresponding optical disc reflection spectrum in the skin detection method according to the first preferred embodiment of the present invention.

[0048] Fig.11 It is a detection effect diagram obtained by the skin detection method according to the first preferred embodiment of the present invention.

[0049] Fig.12 Schematic diagram of the steps of the skin detection method according to the first preferred embodiment of the present invention.

[0050] Fig.13 FIG. 4 is a schematic diagram of a skin detection device according to another preferred embodiment of the present invention. DETAILED DESCRIPTION

[0051] It should be pointed out that the embodiments shown in the drawings are only used as examples to specifically and vividly explain and illustrate the concept of the present invention. The size and structure thereof are not necessarily drawn to scale, nor do they constitute a limitation to the concept of the present invention.

[0052] The directional terms such as up, down, left, right, front, back, front, back, top, bottom, etc. mentioned or may be mentioned in this specification are defined relative to the structures shown in the various drawings. They are relative concepts and may change accordingly according to different positions and different usage conditions. Therefore, these or other directional terms should not be interpreted as restrictive terms.

[0053] Those skilled in the art should understand that, in the disclosure of the present invention, the terms "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicating the orientation or position relationship are based on the orientation or position relationship shown in the drawings, which are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the above terms should not be understood as limiting the present invention.

[0054] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0055] Referring to the drawings of this application specification Figures 1 to 13 As shown, a skin detection method and a skin detection device according to the first preferred embodiment of the present application are explained in the following description. The skin detection method of the preferred embodiment of the present application is to obtain a spectral image of the skin to be detected, and judge whether there is a problem with the skin to be detected based on the corresponding spectral image. Since the reflectance spectrum of each pixel corresponding to the spectral image corresponding to different skin problems is different, the judgment is made based on different characteristic spectra. Figure 1 As shown, the spectral image in the skin detection method of the preferred embodiment of the present application is a three-dimensional matrix, with X and Y as the spatial dimensions and the wavelength of the reflection spectrum of each pixel as the third axis. Since different resolutions correspond to different set bands, the higher the resolution, the more details can be obtained.

[0056] For example, in Figure 1 In the spectral imaging, the image was collected with 70nm as a band, where the spectral image captures the spectrum of each pixel in a certain spectral band, and the bandwidth of each spectral point is consistent. Figure 2As shown, according to another optional embodiment of the present application, a continuous spectral image can also be acquired.

[0057] In this preferred embodiment of the present application, the skin detection device is a detection device based on the skin detection method, wherein the skin detection device includes a spectral imaging device 10, and the method for restoring the spectrum obtained in the spectral imaging image according to the embodiment of the present application is applied to the spectral imaging device 10. Figure 3 The schematic configuration diagram of the spectral imaging device 10 according to the embodiment of the present application is shown, and the spectral imaging device 10 includes an image sensor 11 and a filter structure 12 arranged on the photosensitive path of the image sensor, wherein the filter structure 12 is a broadband filter structure in the frequency domain or wavelength domain. The pass spectra of different wavelengths of the filter structures at different locations are not exactly the same. The filter structure 12 can be a structure or material with filtering characteristics such as a super surface, a photonic crystal, a nanocolumn, a multilayer film, a dye, a quantum dot, a MEMS (micro-electromechanical system), an FP etalon (FP standard tool), a cavity layer (resonant cavity layer), a waveguide layer (waveguide layer), a diffraction element, etc. For example, in an embodiment of the present application, the filter structure 12 can be a light modulation layer in Chinese patent CN201921223201.2. The image sensor 11 is used to convert an optical signal into an electrical signal, wherein the image sensor (i.e., a light detector array) can be a CMOS image sensor (CIS), a CCD, an array light detector, etc.

[0058] The spectral imaging device 10 further includes an optical system 13 and a data processing unit 14, wherein the optical system 13 is located at the front end of the light incident direction of the filter structure 12, and in this preferred embodiment of the present application, the optical system 13 and the data processing unit 14 are optional. According to the spectral imaging device 10 of the embodiment of the present application, the optical system 13 can be an optical system such as a lens assembly, a uniform light assembly, etc. In addition, in the present application, the optional data processing unit 14 can be a processing unit such as MCU, CPU, GPU, FPGA, NPU, ASIC, etc., which can export the data generated by the image sensor 11 to the outside for processing. In this preferred embodiment of the present application, the spectral imaging device 10 is used to obtain a spectral image of the skin to be tested.

[0059] The working principle of the spectral imaging device 10 is further explained below:

[0060] For example, after the image sensor 11 measures the light intensity information, it is transmitted to the data processing unit 14 for restoration calculation. The process is described in detail as follows:

[0061] The intensity signal of the incident light at different wavelengths λ is recorded as x(λ), and the transmission spectrum curve of the filter structure 12 is recorded as T(λ). The filter (the filter structure 12) has m groups of structural units, and the transmission spectra of each group of structural units are different. Overall, the filter structure 12 can be recorded as Ti(λ) (i=1,2,3,…,m). There are corresponding physical pixels under each group of structural units to detect the light intensity bi modulated by the filter structure 12. In a specific embodiment of the present application, a modulation unit is formed with one physical pixel, that is, one physical pixel corresponds to a group of structural units, but it is not limited to this. In other embodiments, multiple physical pixels may correspond to one group of structural units. Therefore, in the spectral imaging device 10 according to the embodiment of the present application, multiple groups of structural units constitute a "spectral pixel". Further, the present invention can use at least one spectral pixel to restore the image.

[0062] It should be noted that the number of effective transmission spectra (the transmission spectrum used for spectral restoration is called effective transmission spectrum) Ti(λ) of the filter structure 12 may be inconsistent with the number of structural units. The transmission spectrum of the filter structure 12 is artificially set, tested, or calculated according to certain rules according to the needs of identification or restoration (for example, the transmission spectrum of each structural unit tested is the effective transmission spectrum). Therefore, the number of effective transmission spectra of the filter structure 12 may be less than the number of structural units, or even more than the number of structural units. In this variant embodiment, a certain transmission spectrum curve is not necessarily determined by a group of structural units.

[0063] The relationship between the spectral distribution of the incident light and the measurement value of the image sensor 11 can be expressed by the following equation:

[0064] bi=∫x(λ)*Ti(λ)*R(λ)dλ

[0065] Discretize it again and get:

[0066] bi=Σ(x(λ)*Ti(λ)*R(λ))

[0067] Where R(λ) is the response of the image sensor, expressed as:

[0068] Ai(λ)=Ti(λ)*R(λ),

[0069] The above formula can be expanded into matrix form:

[0070]

[0071] Among them, bi (i = 1, 2, 3, ..., m) is the response of the image sensor after the light to be measured passes through the filter structure 12, which corresponds to the light intensity measurement values ​​of the image sensor corresponding to m structural units. When a physical pixel corresponds to a structural unit, it can be understood as the light intensity measurement values ​​corresponding to m "physical pixels", which is a vector with a length of m. A is the system's response to light of different wavelengths, which is determined by two factors: the transmittance of the filter structure 12 and the quantum efficiency of the image sensor. A is a matrix, and each row vector corresponds to the response of a group of structural units to incident light of different wavelengths. Here, the incident light is discretely and uniformly sampled, with a total of n sampling points. The number of columns of A is the same as the number of sampling points of the incident light. Here, x (λ) is the light intensity of the incident light at different wavelengths λ, that is, the incident light spectrum to be measured.

[0072] In another optional embodiment of the present application, the filter structure 12 is directly formed on the upper surface of the image sensor 11, such as quantum dots, nanowires, etc., which directly form the filter structure 12 or material (nanowires, quantum dots, etc.) on the photosensitive area of ​​the sensor. Taking the filter structure 12 as an example, at this time, it can be understood that when the raw material of the image sensor 11 is processed to form the image sensor, the filter structure 12 is processed on the upper surface of the raw material, and the transmission spectrum and the response of the image sensor 11 are integrated, that is, it can be understood that the response of the detector and the transmission spectrum are the same curve. At this time, the relationship between the spectral distribution of the incident light and the light intensity measurement value of the image sensor 11 can be expressed by the following formula:

[0073] bi=Σ(x(λ)*Ri(λ))

[0074] That is, in this embodiment, the transmission spectrum Ai(λ)=Ri(λ)

[0075] Furthermore, it can also be a combination of the above two embodiments, that is, at least one filter structure 12 for modulating incident light is provided on the image sensor 11 having the filter structure. It can be understood that the image sensor (i.e., the light detector array) can be a CMOS image sensor (CIS), CCD, array light detector, etc., or an image sensor integrated with the filter structure 12.

[0076] At this time, the relationship between the spectral distribution of the incident light and the light intensity measurement value of the image sensor 11 can be expressed by the following formula:

[0077] bi=∫x(λ)*Ti(λ)*Ri(λ)dλ

[0078] Discretize it again and get:

[0079] bi=Σ(x(λ)*Ti(λ)*Ri(λ))

[0080] That is, in this embodiment, Ai(λ)=Ti(λ)*Ri(λ)

[0081] If multi-channel spectral imaging is considered, the imaging principle of each spectral pixel of the spectral imaging device 10, for example, a snapshot spectral camera, can be expressed by the following equation:

[0082] TX=Y

[0083] By reconstructing the spectral information of the incident light through the above calculation, the spectral information X of the image acquired by the spectral imaging sensor can be obtained. Among them, T is the modulation capability of the response of each chip calibrated in advance by the known input spectral information X1 and the acquired response Y1. The change of this T corresponding to different spectral chips is affected by their metasurface structure.

[0084] Figure 4 FIG. 1 is a schematic diagram showing a spectral image sensor chip 100 for calculating and reconstructing a spectrum of the spectral imaging device 10 of the skin detection device according to an embodiment of the present application. Figure 4 The spectral image sensor chip shown is an image acquisition chip and object imaging recognition device disclosed by the inventor of the present application in Chinese patent CN201921222710.3. The spectral image sensor chip 100 includes a light modulation layer 110, an image sensing layer 120 and at least two groups of pixel confirmation modules 130. The light modulation layer 110 is located above the image sensing layer 120. Each group of pixel confirmation modules 130 includes a structural unit and a sensing unit 132. Each of the structural units and each of the sensing units 132 are respectively arranged on the light modulation layer 110 and the image sensing layer 120 in a corresponding manner. The structural unit and the sensing unit form a modulation unit correspondingly.

[0085] It is worth mentioning that in the preferred embodiment provided by the present application, the spectral image sensor chip 100 of the skin detection device can be but is not limited to a computationally reconstructed spectral image chip. As an example, the image sensor layer 120 of the spectral image sensor chip is configured to obtain the optical signal modulated by the optical modulation layer 110.

[0086] In this preferred embodiment of the present invention, each of the structural units is provided with a plurality of modulation holes or modulation columns penetrating the light modulation layer, and the modulation holes and modulation columns in the same structural unit are arranged into a two-dimensional graphic structure with a specific arrangement rule. The spectral chip includes at least one modulation unit 131 and at least one non-modulation unit. The structural unit corresponding to the modulation unit 131 is arranged corresponding to the sensing unit, while the non-modulation unit is a sensing unit without a structural unit arranged above the sensing unit.

[0087] In a specific embodiment, each modulation unit 131 is arranged corresponding to a sensing unit 132, so that the spectrum received by the modulation unit 131 is formed into a modulation pixel by using the image sensor in the sensing unit 132, and the intensity distribution at a pixel point can be obtained through the algorithm. The multiple modulation pixels corresponding to different groups of pixel confirmation modules can be integrated to form an image containing multiple spectrum information.

[0088] In another embodiment, different from the above structural embodiment, at least four different dyes are formed above the sensing unit 132, and the dyes and the sensing unit constitute a modulation unit, that is, the difference from the above embodiment is that the structural unit is replaced by the dye. Figure 5A As shown in FIG. 1 , the surface of an existing CMOS chip is generally provided with an R, G, B (three dyes) Bayer array, that is, R, G, B dyes and physical pixels are provided in correspondence. Figure 5B As shown, the number and types of dyes of the spectral image sensor chip 100 of this preferred embodiment of the present application are different from those of the CMOS chip of the prior art, wherein the number of dyes is at least 4, so as to obtain at least 4 modulation units, that is, different modulation units have different transmittances, and a plurality of modulation units (which can be repeatedly arranged in a periodic manner) constitute the corresponding filter structure 12. Preferably, in order to make the spectral curve have a better recovery effect, the number of dyes is preferably 9, or 16, that is, 9 or 16 modulation units can be arranged to form a spectral pixel composed of a 3*3 modulation unit arrangement or a 4*4 modulation unit arrangement. In this embodiment, the transmission spectrum curve corresponding to the dye preferably has only one peak. It should be noted that in some embodiments, at least 4 modulation units constitute a spectral pixel, and the corresponding transmission spectrum curve can be understood to be similar to the above-mentioned embodiment of the filter structure, that is, wide spectrum calculation can be realized, for example, 9 modulation units or 16 modulation units constitute a spectral pixel (the same modulation unit composed of the same dye may exist), such as Fig. 9 It can be understood that the modulation units corresponding to the nine different dyes constitute one spectral pixel. The spectral chip can be understood to be composed of a plurality of spectral pixels.

[0089] like Figure 6 The figure shows schematic diagrams of different pixels corresponding to different structures. In this embodiment, different structures are arranged on each pixel, or arranged periodically.

[0090] It is understandable that when light reaches a pixel through different metasurface structures and is imaged by an image sensor, pixels at different positions corresponding to different bands will have different spectral information. Therefore, in this preferred embodiment of the present application, in the image of the skin captured by spectral imaging, the spectral information corresponding to each pixel position of the skin can be restored.

[0091] In the preferred embodiment of the present application, the skin detection method and the skin detection device use the skin spectrum information of the skin color to determine the change of the spectrum information of the position; by comparing with the preset value; when the detection value is lower than or exceeds the set threshold, it is determined that the position has a corresponding skin problem; or when the detection value is within the set threshold range, it is determined that the skin position has a corresponding skin problem.

[0092] like Fig.12 As shown, in this preferred embodiment of the present application, the skin detection method includes the following steps:

[0093] (a) collecting skin spectrum information of the skin to be detected; and

[0094] (b) comparing the collected spectral distribution characteristic data of the skin spectral information with preset spectral distribution characteristic data to obtain a comparison result.

[0095] Optionally, in the step (b) of the skin detection method, comparing the current spectral distribution characteristic data with conventional spectral distribution characteristic data to obtain a comparison result may include:

[0096] (b.1) determining the preset spectral distribution characteristic data corresponding to the target skin; and

[0097] (b.2) Comparing the current spectral distribution characteristic data with the preset spectral characteristic distribution data, determining the health state corresponding to the current spectral distribution characteristic data in the preset spectral characteristic distribution data, so as to obtain the comparison result.

[0098] Further, the spectral distribution of the spectral images corresponding to different bands is obtained, and if a certain band is not within a set threshold range, it is judged as unhealthy. Optionally, in another optional embodiment of the present application, the spectral distribution of the spectral images corresponding to different bands is obtained, and if a certain band is within a set threshold range, the corresponding symptom is judged.

[0099] like Figure 8 As shown, according to the facial spectrum image, various positions of the face can be mapped to different pixels, so as to obtain spectrum information corresponding to different positions. Fig. 9 The spectral distribution of healthy skin preset for this face. As an example, Figure 7 There are three black spots in the image, which are set as unhealthy locations. Figure 8 The corresponding spectral image can be obtained for each pixel point in the image, thereby obtaining the actual distribution diagram of the corresponding optical disc reflectance spectrum.

[0100] like Figures 9 to 11As shown, the brightness values ​​at three positions of the user's skin area in the image captured at the positions of bands Dλ1, Dλ2, and Dλ3 are obviously changed, which are lower than the spectral distribution energy value in a healthy state. The corresponding unhealthy reasons can be found according to the pre-stored changes.

[0101] Furthermore, in the present application, the reflection characteristics of the skin at each wavelength are obtained, and whether the skin is healthy is determined based on the characteristics of the obtained spectral information. Therefore, in order to eliminate the difference in spectral response of the imaging path (including the light source, the shooting device and the filter) at different wavelengths, a standard reference object with known reflectivity at different wavelengths is used to calibrate the spectral response of the imaging path at different wavelengths, thereby obtaining the true reflection characteristics. The calibration method is related to the reflection characteristics to be obtained, where the reflection characteristics can refer to the gray value, reflectivity and deformation of the pixel, for example, the deformation of the gray value can make the normalized gray value and the difference between the gray values.

[0102] Accordingly, in this preferred embodiment of the present application, a calibration step is further included before step (a) of the skin detection method, wherein the step is: using a standard reference object with known reflectivity at different wavelengths to calibrate the spectral response of the imaging channel at different wavelengths, so as to obtain the true reflection characteristics.

[0103] like Fig.13 As shown, the skin detection device further includes a light source 20 and a processor 30, wherein the light source 20 irradiates the surface of the skin to be tested to provide the required light for the spectral imaging device 10, and the processor 30 is connected to the spectral imaging device 10. The spectral imaging device 10 transmits the detected spectral data information of the skin to be tested to the processor 30, and the processor 30 performs comparison based on the preset spectral characteristic distribution data to determine the health status corresponding to the current spectral distribution characteristic data in the preset spectral characteristic distribution data to obtain the comparison result.

[0104] It is worth mentioning that the light source 20 of the skin detection device must cover the selected multi-spectral band (natural light can also be used when conditions permit). For example, if the selected band is 380nm to 1000nm, then the light source 20 must include these two bands.

[0105] It is understandable that, since skin problems are usually difficult for the human eye to distinguish, they can also be detected through spectral imaging, because different skin changes will cause changes in spectral energy distribution, but there will also be influences caused by makeup, so it is still necessary to pre-store the spectral energy distribution of skin conditions in different scenes multiple times in different scenes. It is worth mentioning that the step (b) of the skin detection method further includes a preset step, wherein the step is: pre-store the spectral energy distribution of skin conditions in different scenes multiple times, and use the spectral energy distribution as the preset spectral feature distribution data of the corresponding skin.

[0106] The step (b) of the skin detection method further comprises the following steps: determining that the predicted state of the target skin is a preset state according to the spectral energy variation trend of the target skin, and then issuing a prompt warning.

[0107] Furthermore, the spectral imaging device 10 recovers the spectral image data of the skin, wherein the step (a) of the skin detection method described in the present application further includes the following steps:

[0108] (a.1) obtaining a light energy response signal matrix and a standard spectrum output by a spectral image sensor chip of the spectral imaging device;

[0109] (a.2) determining a basic element restoration function and a response signal vector of the basic element restoration function based on a light energy response signal matrix, wherein the basic element restoration function uses a predetermined pixel value of the spectral image sensor chip and a pixel value in its vicinity to restore a spectral image value of a corresponding predetermined channel;

[0110] (a.3) obtaining a recovery tensor, wherein the product of the recovery tensor and the response signal vector is equal to the output of the basic element recovery function based on the response signal vector; and

[0111] (a.4) Obtaining a restored spectral image based on the product of the restored tensor and the response signal vector.

[0112] For ease of understanding, this application description introduces the entire content of the Chinese invention with application number CN2021111546565.

[0113] It should be noted that in the step (a.1) of the skin detection method of the present application, the light energy response signal matrix is ​​represented as a matrix B including two dimensions of image width w and image height h, the dimension of the standard spectrum is l, and is set so that the distance between the product of the true value tensor of the spectral image received by the spectral imaging device and the standard spectrum and the spectral image tensor to be restored is minimized.

[0114] Furthermore, the standard spectrum is represented by s, and the channel standard spectrum corresponding to the kth channel of the standard spectrum is represented by sk, so that: xk→O(i,j)sk

[0115] Among them, xk is the spectral image value of the kth channel of a spectral pixel, O(i,j) is the true value tensor of the spectral curve of a spectral pixel, and → indicates that the Euclidean distance between the tensors is the minimum.

[0116] For ease of understanding, this application description introduces the entire content of the Chinese invention with application number CN2021111516729.

[0117] Furthermore, the spectral imaging device 10 reconstructs spectral image data of the skin, wherein the step (a) of the skin detection method described in the present application further includes the following steps:

[0118] Acquiring transmission spectrum data of the spectral imaging chip and output signal data of the spectral imaging chip;

[0119] Acquire local transmission spectrum data of the transmission spectrum data and local output signal data of the output signal data based on the pixel used to reconstruct the spectrum image;

[0120] Inputting the local output signal data into an attention model to obtain attention local data; and

[0121] The local transmission spectrum data, the local output signal data and the attention local data are input into a neural network model to obtain the pixels for reconstructing a spectral image.

[0122] In the step (a) of the preferred embodiment of the present application, obtaining local transmission spectrum data of the transmission spectrum data and local output signal data of the output signal data based on the pixel used to reconstruct the spectral image includes: based on the position of the pixel used to reconstruct the spectral image, obtaining local transmission spectrum data of the transmission spectrum data and local output signal data of the output signal data in an area near the position with a side length of a predetermined number of pixels.

[0123] In the step (a) of the preferred embodiment of the present application, inputting the local output signal data into the attention model to obtain the attention local data includes: dividing the local output signal data into a plurality of predetermined areas, each predetermined area including output signal data corresponding to a plurality of pixels of the spectral imaging chip; and performing matrix multiplication for each of the predetermined areas to obtain the attention local data.

[0124] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.

[0125] The technical scope of the present invention is not limited to the contents in the above description. Those skilled in the art can make various deformations and modifications to the above embodiments without departing from the technical idea of ​​the present invention, and these deformations and modifications all belong to the protection scope of the present invention.

Claims

1. A skin detection method, characterized in that: include: (a) collecting skin spectrum information of the skin to be detected; and (b) comparing the collected spectral distribution characteristic data of the skin spectral information with preset spectral distribution characteristic data to obtain a comparison result.

2. The skin detection method according to claim 1, wherein in the step (b), comparing the current spectral distribution characteristic data with conventional spectral distribution characteristic data to obtain a comparison result may include: (b.1) determining the preset spectral distribution characteristic data corresponding to the target skin; and (b.2) Comparing the current spectral distribution characteristic data with the preset spectral characteristic distribution data, determining the health state corresponding to the current spectral distribution characteristic data in the preset spectral characteristic distribution data, so as to obtain the comparison result.

3. The skin detection method according to claim 2, further comprising a calibration step before step (a), wherein the step is: using a standard reference object with known reflectivity at different wavelengths to calibrate the spectral response of the imaging channel at different wavelengths to obtain true reflection characteristics.

4. The skin detection method according to claim 2, wherein the step (b) further includes a preset step, wherein the step is: pre-storing the spectral energy distribution of the skin state under different scenes for multiple times, and using the spectral energy distribution as the preset spectral feature distribution data of the corresponding skin.

5. The skin detection method according to claim 4, wherein the step (b) further comprises the following steps: determining that the predicted state of the target skin is a preset state according to the spectral energy change trend of the target skin, and then issuing a prompt warning.

6. The skin detection method according to claim 4, wherein the step (a) further comprises the following steps: (a.1) obtaining a light energy response signal matrix and a standard spectrum output by a spectral image sensor chip of the spectral imaging device; (a.2) determining a basic element restoration function and a response signal vector of the basic element restoration function based on a light energy response signal matrix, wherein the basic element restoration function uses a predetermined pixel value of the spectral image sensor chip and a pixel value in its vicinity to restore a spectral image value of a corresponding predetermined channel; (a.3) obtaining a recovery tensor, wherein the product of the recovery tensor and the response signal vector is equal to the output of the basic element recovery function based on the response signal vector; and (a.4) Obtaining a restored spectral image based on the product of the restored tensor and the response signal vector.

7. The skin detection method according to claim 6, wherein in the step (a.1), the light energy response signal matrix is ​​represented as a matrix B including two dimensions of image width w and image height h, the dimension of the standard spectrum is l, and is set so that the distance between the product of the true value tensor of the spectral image received by the spectral imaging device and the standard spectrum and the spectral image tensor to be restored is minimized.

8. The skin detection method according to claim 7, wherein the standard spectrum is represented by s, and the channel standard spectrum corresponding to the kth channel of the standard spectrum is represented by s k , so that: x k →O(i,j)s k ,in, x k is the spectral image value of the kth channel of a spectral pixel, O(i,j) is the true value tensor of the spectral curve of a spectral pixel, and → indicates that the Euclidean distance between the tensors is the minimum.

9. The skin detection method according to claim 4, wherein the step (a) further comprises the following steps: Acquiring transmission spectrum data of the spectral imaging chip and output signal data of the spectral imaging chip; Acquire local transmission spectrum data of the transmission spectrum data and local output signal data of the output signal data based on the pixel used to reconstruct the spectrum image; Inputting the local output signal data into an attention model to obtain attention local data; and The local transmission spectrum data, the local output signal data and the attention local data are input into a neural network model to obtain the pixels for reconstructing a spectral image.

10. The skin detection method according to claim 9, wherein in the step (a), obtaining local transmission spectrum data of the transmission spectrum data and local output signal data of the output signal data based on the pixels used to reconstruct the spectrum image comprises: Based on the position of the pixel used to reconstruct the spectral image, local transmission spectrum data of the transmission spectrum data and local output signal data of the output signal data in an area near the position with a side length of a predetermined number of pixels are acquired.

11. The skin detection method according to claim 10, wherein in the step (a), inputting the local output signal data into the attention model to obtain the attention local data comprises: Dividing the local output signal data into a plurality of predetermined regions, each predetermined region including output signal data corresponding to a plurality of pixels of the spectral imaging chip; And, matrix multiplication is performed for each of the predetermined areas to obtain the local attention data.

12. A skin detection device, characterized in that: include: Spectral imaging equipment; and A processor is communicatively connected to the spectral imaging device, the spectral imaging device is used to collect spectral image information of the skin to be tested, wherein the processor compares the collected spectral distribution characteristic data of the skin spectral information with preset spectral distribution characteristic data to obtain a comparison result.

Images