Detection system and detection method

Abstract

This application relates to the field of artificial intelligence technology and provides a detection system and method, including: an image acquisition module for acquiring images of a user's tongue and face; an image processing module for dividing the tongue image into a nine-grid layout to obtain multiple first detection images, and dividing the face image into a fifteen-grid layout to obtain multiple second detection images; a tongue detection module for inputting each of the first detection images into a tongue detection model for processing to obtain tongue detection results in different tongue dimensions; a face detection module for inputting each of the second detection images into a face detection model for processing to obtain face detection results in different face dimensions; and a result analysis module for determining the user's vital signs based on the tongue and face detection results. This system, through image segmentation processing, can obtain images free from background interference, thereby improving detection accuracy.

Description

Technical Field

[0001] This application belongs to the field of artificial intelligence technology, and particularly relates to a detection system and a detection method. Background Art

[0002] In recent years, as people's attention to health has been continuously increasing, the value of traditional Chinese medicine has been increasingly emphasized. Among the four diagnostic methods in traditional Chinese medicine, inspection can determine the location and nature of diseases by observing the patient's body shape, complexion, tongue body, and tongue coating according to the changes in form and color. At the same time, with the remarkable progress of artificial intelligence globally, machine learning and deep learning algorithms have been continuously optimized, showing powerful capabilities in the fields of image recognition, image preprocessing technology, data analysis, etc. Therefore, by learning a large amount of image data of the tongue and face, a model that can automatically identify tongue image and complexion features is trained to improve the accuracy and efficiency of traditional Chinese medicine diagnosis. Thus, the research and development of a health detector for the tongue and face in traditional Chinese medicine is a product that conforms to the overall development trend of the technical field and meets the needs of the modernization of traditional Chinese medicine.

[0003] However, the detection accuracy of the current related technologies for the patient's condition is often low and difficult to meet the actual needs. Summary of the Invention

[0004] The embodiments of this application provide a detection system and a detection method, which improve the detection accuracy of the patient's condition.

[0005] In a first aspect, the embodiments of this application provide a detection system, including:

[0006] An image acquisition module, configured to acquire the tongue image and face image of a user;

[0007] An image processing module, configured to divide the tongue image into a nine-grid to obtain a first detection image corresponding to each tongue detection area of the user, and divide the face image into a fifteen-grid to obtain a second detection image corresponding to each face detection area of the user;

[0008] A tongue detection module, configured to input each of the first detection images into a tongue detection model for processing to obtain the tongue detection results of the user's tongue in different tongue dimensions;

[0009] A face detection module, configured to input each of the second detection images into a face detection model for processing to obtain the face detection results of the user's face in different face dimensions;

[0010] A result analysis module, configured to determine the vital sign detection result of the user based on each of the tongue detection results and each of the face detection results.

[0011] Optionally, the image processing module is specifically used for:

[0012] A first analysis result is obtained by performing color analysis on the tongue image based on a standard tongue image and a first standard text; a second analysis result is obtained by performing color analysis on the face image based on a standard face image and a second standard text.

[0013] Based on the first analysis result, a first correction method is determined, and the tongue image is corrected based on the first correction method to obtain a first corrected image;

[0014] Based on the second analysis result, a second correction method is determined, and the facial image is corrected based on the second correction method to obtain a second corrected image;

[0015] The first corrected image is divided into a nine-grid structure to obtain each first detection image, and the second corrected image is divided into a fifteen-grid structure to obtain each second detection image.

[0016] Optionally, the tongue detection model includes: a tongue color detection sub-model, a tongue coating color detection sub-model, a tongue texture detection sub-model, a tongue coating texture detection sub-model, a saliva detection sub-model, a crack detection sub-model, and a teeth mark detection sub-model. Correspondingly, the tongue detection module specifically includes:

[0017] The tongue color detection unit is used to input each of the first detection images into the tongue color detection sub-model for processing, and to obtain the first result of the user's tongue in the tongue color dimension.

[0018] The moss color detection unit is used to input each of the first detection images into the moss color detection sub-model for processing, and obtain a second result of the user's tongue in the moss color dimension;

[0019] The tongue detection unit is used to input each of the first detection images into the tongue detection sub-model for processing, and obtain the third result of the user's tongue in the tongue dimension.

[0020] The moss detection unit is used to input each of the first detection images into the moss detection sub-model for processing, and to obtain the fourth result of the user's tongue in the moss dimension.

[0021] The body fluid detection unit is used to input each of the first detection images into the body fluid detection sub-model for processing, and obtain the fifth result of the user's tongue in the body fluid dimension;

[0022] The crack detection unit is used to input each of the first detection images into the crack detection sub-model for processing, and obtain the sixth result of the user's tongue in the crack dimension;

[0023] The teeth mark detection unit is used to input each of the first detection images into the teeth mark detection sub-model for processing, and obtain the seventh result of the user's tongue in the teeth mark dimension.

[0024] Optionally, the facial detection model includes: a facial luster detection sub-model, an eye puffiness detection sub-model, a facial pigmentation detection sub-model, a facial acne detection sub-model, a lip thickness detection sub-model, and an eyebrow density detection sub-model. Correspondingly, the facial detection module specifically includes:

[0025] The facial gloss detection unit is used to input each of the second detection images into the facial gloss detection sub-model for processing, and obtain the eighth result of the user's face in the facial gloss dimension.

[0026] The eye puffiness detection unit is used to input each of the second detection images into the eye puffiness detection sub-model for processing, and obtain the ninth result of the user's face in the eye puffiness dimension;

[0027] The facial pigmentation detection unit is used to input each of the second detection images into the facial pigmentation detection sub-model for processing, and obtain the tenth result of the user's face in the facial pigmentation dimension;

[0028] The facial acne detection unit is used to input each of the second detection images into the facial acne detection sub-model for processing, and to obtain the eleventh result of the user's face in the facial acne dimension.

[0029] The lip thickness detection unit is used to input each of the second detection images into the lip thickness detection sub-model for processing, and obtain the twelfth result of the user's face in the lip thickness dimension.

[0030] The eyebrow density detection unit is used to input each of the second detection images into the eyebrow density detection sub-model for processing, and obtain the thirteenth result of the user's face in the eyebrow density dimension.

[0031] Optionally, the tongue texture dimensions include tongue area, ecchymosis, and prickles; the tongue texture detection unit is specifically used for:

[0032] Each of the first detection images is segmented to obtain the tongue region; the number of pixels in the tongue region is counted to obtain the pixel area corresponding to the tongue region; the pixel area is converted based on the device parameters of the image acquisition module to obtain the tongue area.

[0033] Color clustering algorithm is used to perform color analysis on each of the first detection images to obtain ecchymosis regions; morphological processing is used to analyze the ecchymosis regions to obtain the number of ecchymosis and the total area of ​​ecchymosis.

[0034] Feature extraction is performed on the tongue region to obtain the dotted region; the dotted region is analyzed based on morphological processing to obtain the number of dots and the total area of ​​dots.

[0035] Optionally, the crack detection unit is specifically used for:

[0036] Image segmentation processing is performed on each of the first detection images to obtain the tongue region;

[0037] Determine the characteristics of the crack;

[0038] Based on the crack features, edge detection is performed on the tongue region to obtain the edges of each crack.

[0039] The number of cracks is obtained by performing connected region analysis on each crack edge.

[0040] Each crack is subjected to contour detection to obtain the length of each crack.

[0041] Optionally, the facial gloss detection unit is specifically used for:

[0042] Texture features are extracted from each of the second detection images to obtain the texture features of the face;

[0043] Each of the second detection images is subjected to color space conversion processing to obtain each color image under the set color model;

[0044] Each color image is analyzed using set parameters to obtain gloss analysis results; the set parameters are determined by the set color model.

[0045] Based on the texture features and the gloss analysis results, the eighth result is determined.

[0046] Optionally, the eye puffiness detection unit is specifically used for:

[0047] Facial key point detection processing is performed on each of the second detection images to determine the eye region;

[0048] Based on the eye region and a standard eye image, determine the range of shape variation of the eye in the eye region;

[0049] The result of eye puffiness is determined based on the magnitude of the shape change;

[0050] If the result of the eye swelling is that swelling exists, then the swollen area in the eye region is determined, and the color distribution and wrinkle depth of the swollen area are determined;

[0051] The ninth result is determined based on the color distribution and the wrinkle depth.

[0052] Optionally, the detection system further includes:

[0053] An information storage module is used to obtain the user's identity information when a detection command is detected; obtain the user's historical detection records and body information based on the identity information; and associate and store the identity information with the historical detection records and body information in an information database.

[0054] The information storage module is also used to associate and store the tongue detection results, facial detection results, and vital sign detection results with the identity information in the information database.

[0055] Optionally, the detection system further includes:

[0056] The information output module is used to determine the user's constitution type based on the vital sign detection results; determine the user's symptom information based on the constitution type, each of the tongue detection results, and each of the facial detection results; determine the user's conditioning information in different conditioning dimensions based on the symptom information, and output the conditioning information.

[0057] Secondly, embodiments of this application provide a detection method, including:

[0058] Collect images of the user's tongue and face;

[0059] The tongue image is divided into a nine-grid structure to obtain the first detection image corresponding to each tongue detection area of ​​the user, and the facial image is divided into a fifteen-grid structure to obtain the second detection image corresponding to each facial detection area of ​​the user.

[0060] Each of the first detection images is input into the tongue detection model for processing to obtain the tongue detection results of the user in different tongue dimensions;

[0061] Each of the second detection images is input into the face detection model for processing to obtain the face detection results of the user's face in different facial dimensions;

[0062] The results analysis module is used to determine the user's vital signs detection results based on the tongue detection results and the facial detection results.

[0063] The beneficial effects of the embodiments in this application compared with the prior art are:

[0064] This application provides a detection system including an image acquisition module for acquiring images of a user's tongue and face; an image processing module for dividing the tongue image into a nine-grid layout to obtain first detection images corresponding to each tongue detection area, and dividing the face image into a fifteen-grid layout to obtain second detection images corresponding to each face detection area; a tongue detection module for inputting each of the first detection images into a tongue detection model for processing to obtain tongue detection results for the user in different tongue dimensions; a face detection module for inputting each of the second detection images into a face detection model for processing to obtain face detection results for the user's face in different face dimensions; and a result analysis module for determining the user's vital signs detection results based on the tongue detection results and the face detection results. Compared with existing technologies, this system can obtain images of each tongue detection region without background interference through a nine-grid division, thus improving the accuracy of tongue detection; and can obtain images of each face detection region without background interference through a fifteen-grid division, thus improving the accuracy of face detection; at the same time, this system can obtain results in different tongue dimensions through the tongue detection model and results in different face dimensions through the face detection model, thereby improving the accuracy of user detection. Attached Figure Description

[0065] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0066] Figure 1 This is a schematic diagram of the structure of a detection system provided in an embodiment of this application;

[0067] Figure 2 This is a flowchart illustrating the implementation of a detection method provided in an embodiment of this application;

[0068] Figure 3 This is a schematic diagram of the specific structure of the tongue detection module in the detection system provided in one embodiment of this application;

[0069] Figure 4 This is a schematic diagram of the specific structure of the face detection module in a detection system provided in an embodiment of this application;

[0070] Figure 5 This is a schematic diagram of the structure of a detection system provided in another embodiment of this application;

[0071] Figure 6 This is a schematic diagram of the detection system provided in another embodiment of this application. Detailed Implementation

[0072] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0073] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0074] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0075] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0076] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0077] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0078] In recent years, with the continuous increase in people's attention to health, the value of traditional Chinese medicine has been increasingly emphasized. Among the four diagnostic methods in traditional Chinese medicine, inspection can determine the location and nature of diseases by observing the patient's body shape, complexion, tongue body, and tongue coating according to the changes in form and color. At the same time, with the remarkable progress of artificial intelligence globally and the continuous optimization of machine learning and deep learning algorithms, powerful capabilities have been demonstrated in the fields of image recognition, image preprocessing technology, data analysis, etc. Therefore, by learning a large amount of image data of the tongue and face, a model capable of automatically recognizing tongue and complexion features is trained to improve the accuracy and efficiency of traditional Chinese medicine diagnosis. Thus, the research and development of a health detector for the tongue and face in traditional Chinese medicine is a product that conforms to the overall development trend of the technical field and meets the needs of the modernization of traditional Chinese medicine.

[0079] However, the detection accuracy of the current related technologies for the patient's condition is often low and difficult to meet the actual needs. Therefore, in all embodiments of this application, a detection system is provided to improve the detection accuracy of the patient's condition.

[0080] Please refer to Figure 1 , Figure 1 which is a schematic structural diagram of a detection system provided by an embodiment of this application.

[0081] For the sake of convenience of description, only the parts related to this embodiment are shown and are detailed as follows:

[0082] As Figure 1 shown, the detection system 1 includes: an image acquisition module 10, an image processing module 20, a tongue detection module 30, a face detection module 40, and a result analysis module 50. Among them, the image acquisition module 10 is communicatively connected to the image processing module 20, the image processing module 20 is communicatively connected to the tongue detection module 30 and the face detection module 40 respectively, and the result analysis module 50 is communicatively connected to the tongue detection module 30 and the face detection module 40 respectively. It should be noted that the above communication connection method can be a wired communication connection or a wireless communication connection.

[0083] In some possible embodiments, the image acquisition module 10 can be a camera, a camera or other devices for taking pictures.

[0084] In an embodiment of this application, the image acquisition module 10 is used to acquire the tongue image and face image of the user (such as Figure 2 shown in step S101).

[0085] The image processing module 20 is used to divide the tongue image into a nine-grid to obtain the first detection image corresponding to each tongue detection area of the user, and divide the face image into a fifteen-grid to obtain the second detection image corresponding to each face detection area of the user (such as Figure 2 shown in step S102).

[0086] The tongue detection module 30 is used to input each first detection image into the tongue detection model for processing, and obtain the tongue detection results of the user's tongue in different tongue dimensions (e.g., Figure 2 (Step S103 shown).

[0087] The face detection module 40 is used to input each second detection image into the face detection model for processing, and obtain the face detection results of the user's face in different facial dimensions (such as...). Figure 2 Step S104 is shown.

[0088] The results analysis module 50 is used to determine the user's vital signs detection results (such as...) based on the results of each tongue detection and each facial detection. Figure 2 (Step S105 shown).

[0089] In practical applications, when operators need to understand a user's health status, or when a user needs to understand their own health status, the user can enter the shooting range of the image acquisition module 10 so that the image acquisition module 10 can acquire images of the user's tongue and face. The operators may include medical personnel or the user's guardians.

[0090] In some possible embodiments, in order to improve image acquisition efficiency and accuracy, the image acquisition module 10 can output various prompts to the user in real time during the process of acquiring tongue images. For example, at the beginning of acquiring tongue images, it can output the prompt "Please move to XX and stick out your tongue" to remind the user to stick out their tongue in time; or at the middle of acquiring tongue images, it can output prompts such as "Please stick out your tongue" to obtain a more accurate tongue image; or after successfully acquiring tongue images, it can output the prompt "Tongue image acquisition completed" to remind the user that the tongue acquisition stage has ended.

[0091] During the process of acquiring facial images, the image acquisition module 10 can output various prompts to the user in real time. For example, at the beginning of facial image acquisition, it can output the prompt "Please move to XX and stay still" to remind the user to perform facial detection in time; or in the middle of facial image acquisition, it can output prompts such as "Please move left / right XXX" to obtain a more accurate facial image; or after successfully acquiring facial images, it can output the prompt "Facial image acquisition completed" to remind the user that the facial acquisition stage has ended.

[0092] In one embodiment of this application, to avoid mutual interference between tongue image acquisition and facial image acquisition, the image acquisition module 10 may include a tongue acquisition unit (not shown in the figure) and a facial acquisition unit (not shown in the figure). Furthermore, the acquisition time of the tongue acquisition unit differs from the acquisition time of the facial acquisition unit.

[0093] It should be noted that the nine-grid division specifically refers to dividing the tongue image into nine first detection images according to nine tongue detection regions, in order to obtain images of each tongue detection region without background interference, thereby improving the accuracy of subsequent tongue detection. The sizes of the various first detection images are not entirely the same.

[0094] In this embodiment of the application, each tongue detection area may specifically include: the root area of ​​the tongue (including the right side area, the middle area, and the left side area), the middle area of ​​the tongue (including the right side area, the middle area, and the left side area), and the tip area of ​​the tongue (including the right side area, the middle area, and the left side area).

[0095] In practical applications, the root area of ​​the tongue corresponds to the large intestine, kidneys, and small intestine in traditional Chinese medicine, the middle area of ​​the tongue corresponds to the gallbladder, liver, and spleen in traditional Chinese medicine, and the tip area of ​​the tongue corresponds to the lungs, stomach, and heart in traditional Chinese medicine.

[0096] The 15-grid division specifically refers to dividing the facial image into 15 second detection images based on 15 facial detection regions. This results in facial detection region images free from background interference, thereby improving the accuracy of subsequent facial detection. The sizes of these second detection images are not entirely identical.

[0097] In this embodiment of the application, each facial detection area may specifically include: hairline area, forehead area, left temple area, right temple area, center of forehead area, area between eyebrows, area between eyes, left and right eye socket areas, left cheekbone area, right cheekbone area, nasal wing area, nose tip area, bridge of nose area, philtrum and mouth area, and chin area.

[0098] In practical applications, the hairline corresponds to the psychological stress area in Traditional Chinese Medicine (TCM); the forehead corresponds to the cardiovascular system; the temples correspond to the liver; the center of the forehead corresponds to the lungs; the area between the eyebrows corresponds to the lungs; the area between the eyes corresponds to the heart; the eye sockets correspond to the chest; the areas on either side of the frontal bone correspond to the small and large intestines (e.g., the area from the lower cheekbone to the outer corner of the eye is the large intestine area, and the area from the lower cheekbone to the inner corner of the eye is the small intestine area); the nasal wing area corresponds to the stomach; the tip of the nose corresponds to the spleen; the bridge of the nose corresponds to the liver; the philtrum area corresponds to the urinary and reproductive system (e.g., the bladder and uterus); and the chin area corresponds to the kidneys and areas of general soreness, also known as fatigue and soreness areas.

[0099] In one embodiment of this application, since the color intensity of the tongue affects the tongue detection result, and the color intensity of the face affects the face detection result, in order to further improve the detection accuracy of the tongue and the detection accuracy of the face, the image processing module 20 can be specifically used to:

[0100] A first analysis result is obtained by performing color analysis on the tongue image based on a standard tongue image and a first standard text; a second analysis result is obtained by performing color analysis on the face image based on a standard face image and a second standard text.

[0101] Based on the first analysis result, a first correction method is determined, and the tongue image is corrected based on the first correction method to obtain a first corrected image;

[0102] Based on the second analysis result, a second correction method is determined, and the facial image is corrected based on the second correction method to obtain a second corrected image;

[0103] The first corrected image is divided into a nine-grid structure to obtain each first detection image, and the second corrected image is divided into a fifteen-grid structure to obtain each second detection image.

[0104] It should be noted that the standard tongue image specifically refers to an image containing the standard tongue color, and the first standard text specifically refers to text containing a description of the tongue color, such as traditional Chinese medicine books.

[0105] The standard facial image specifically refers to an image containing standard facial colors, while the second standard text specifically refers to text containing descriptions of facial colors, such as traditional Chinese medicine books.

[0106] In practical applications, color analysis methods include, but are not limited to, histogram analysis, color distribution analysis, and color balance analysis.

[0107] A histogram is a tool used to represent the brightness levels of an image. Its horizontal axis represents brightness, and its vertical axis represents the number of pixels. A histogram provides a visual representation of the number of pixels at each brightness level, allowing for analysis of the image's exposure. The left side of the histogram represents the shadows, the right side represents the highlights, and the middle section represents the midtones.

[0108] Color distribution refers to the distribution of different colors in an image. By analyzing the number of pixels of different colors in a histogram, we can understand the color richness and color uniformity of an image.

[0109] Color balance refers to the proportion and contrast between different colors in an image.

[0110] In this embodiment, the first analysis result and the second analysis result include, but are not limited to, the results corresponding to parameters such as brightness, contrast, saturation and color temperature, as well as whether there is color cast.

[0111] Color correction methods include, but are not limited to: color balance adjustment (which corrects color cast by adjusting the proportions of the red, green, and blue color channels in an image), curve adjustment (which uses curve tools to adjust the shape of curves, changing the color of different brightness areas in an image, and distributing the image's brightness, contrast, and color), saturation adjustment (when the image's saturation is insufficient or excessive, appropriately adjusting the saturation parameters to make the colors more vivid or natural), and color temperature adjustment (adjusting the color temperature parameters to make the image's colors closer to actual lighting conditions).

[0112] Based on this, in this embodiment, the image processing module 20 can determine an accurate first correction method based on the first analysis result, and perform correction processing on the tongue image based on the first correction method to obtain a first corrected image.

[0113] The image processing module 20 can also determine an accurate second correction method based on the second analysis result, and perform correction processing on the facial image based on the second correction method to obtain a second corrected image.

[0114] Subsequently, the image processing module 20 can further divide the first corrected image into a nine-grid structure to obtain various first detection images, and divide the second corrected image into a fifteen-grid structure to obtain various second detection images.

[0115] In another embodiment of this application, the image processing module 20 can also perform preprocessing operations such as illumination restoration, blur detection, and high-frequency information filtering on the tongue image and face image to reduce or remove the influence of adverse factors such as illumination, other interference information, and noise on the quality of the tongue image and face image.

[0116] In this embodiment, the tongue detection model is used to detect different tongue dimensions in each first detection image. These different tongue dimensions include, but are not limited to: tongue color dimension, coating color dimension, tongue texture dimension, coating texture dimension, saliva dimension, crack dimension, and teeth mark dimension.

[0117] The tongue detection model can be obtained by training a pre-built deep learning model based on a first preset sample set. Each sample data in the first preset sample set includes a sample tongue image and corresponding tongue detection results. When training the pre-built deep learning model, the sample tongue image in each sample is used as the input to the deep learning model, and the corresponding tongue detection results in each sample are used as the output. Through training, the deep learning model can learn the correspondence between all possible tongue images and their respective detection results. The trained deep learning model is then used as the tongue detection model.

[0118] The facial detection model is used to detect different facial dimensions in each of the second detection images. These different facial dimensions include, but are not limited to: facial radiance, eye puffiness, facial pigmentation, facial acne, lip thickness, and eyebrow density.

[0119] The face detection model can be obtained by training a pre-built second deep learning model based on a second preset sample set. Each sample data in the second preset sample set includes a sample face image and corresponding face detection results. When training the pre-built second deep learning model, the sample face image in each sample is used as the input to the second deep learning model, and the corresponding face detection results in each sample are used as the output. Through training, the second deep learning model can learn the correspondence between all possible face images and their respective face detection results. The trained second deep learning model is then used as the face detection model.

[0120] In this embodiment, after obtaining the various tongue and facial detection results, the result analysis module 50 can generate corresponding vital sign detection results based on these results. The vital sign detection results are used to describe the user's health status.

[0121] Specifically, the results analysis module 50 can determine the state of each part of the user's body based on the results of each tongue detection and each facial detection, and determine the state as the aforementioned vital signs detection results.

[0122] As can be seen from the above, the detection system provided in this application includes an image acquisition module for acquiring images of a user's tongue and face; an image processing module for dividing the tongue image into a nine-grid layout to obtain first detection images corresponding to each tongue detection area, and dividing the face image into a fifteen-grid layout to obtain second detection images corresponding to each face detection area of ​​the user; a tongue detection module for inputting each of the first detection images into a tongue detection model for processing to obtain tongue detection results of the user's tongue in different tongue dimensions; a face detection module for inputting each of the second detection images into a face detection model for processing to obtain face detection results of the user's face in different face dimensions; and a result analysis module for determining the user's vital sign detection results based on each of the tongue detection results and each of the face detection results. Compared with existing technologies, this system can obtain images of each tongue detection region without background interference through a nine-grid division, thus improving the accuracy of tongue detection; and can obtain images of each face detection region without background interference through a fifteen-grid division, thus improving the accuracy of face detection; at the same time, this system can obtain results in different tongue dimensions through the tongue detection model and results in different face dimensions through the face detection model, thereby improving the accuracy of user detection.

[0123] Please see Figure 3 , Figure 3 This is a schematic diagram of the specific structure of the tongue detection module in the detection system provided in one embodiment of this application. It should be noted that, in this embodiment, the tongue detection model may include: a tongue color detection sub-model, a tongue coating color detection sub-model, a tongue texture detection sub-model, a tongue coating texture detection sub-model, a saliva detection sub-model, a crack detection sub-model, and a teeth mark detection sub-model.

[0124] Therefore, as Figure 3 As shown, the tongue detection module 30 specifically includes:

[0125] The tongue color detection unit 31 is used to input each of the first detection images into the tongue color detection sub-model for processing, so as to obtain the first result of the user's tongue in the tongue color dimension.

[0126] The moss color detection unit 32 is used to input each of the first detection images into the moss color detection sub-model for processing, so as to obtain the second result of the user's tongue in the moss color dimension.

[0127] The tongue detection unit 33 is used to input each of the first detection images into the tongue detection sub-model for processing, and obtain the third result of the user's tongue in the tongue dimension.

[0128] The moss detection unit 34 is used to input each of the first detection images into the moss detection sub-model for processing, and obtain the fourth result of the user's tongue in the moss dimension.

[0129] The saliva detection unit 35 is used to input each of the first detection images into the saliva detection sub-model for processing, and obtain the fifth result of the user's tongue in the saliva dimension.

[0130] The crack detection unit 36 ​​is used to input each of the first detection images into the crack detection sub-model for processing, and obtain the sixth result of the user's tongue in the crack dimension.

[0131] The teeth mark detection unit 37 is used to input each of the first detection images into the teeth mark detection sub-model for processing, and obtain the seventh result of the user's tongue in the teeth mark dimension.

[0132] The first result is used to describe the color of the tongue.

[0133] In practical applications, the tongue body, also known as the tongue tissue, is the muscular and vascular tissue of the tongue.

[0134] In this embodiment, the tongue color detection unit 31 can specifically perform image segmentation processing on each first detection image to obtain the tongue body region; then, the tongue color detection unit 31 can perform color analysis on the tongue body region through a color histogram to obtain the average color and color distribution of the tongue body region; then, the tongue color detection unit 31 can determine the first result based on the above average color and color distribution.

[0135] The second result is used to describe the color of the tongue coating.

[0136] In practical applications, tongue coating is a term in Traditional Chinese Medicine, referring to a thin, white, and moist coating on the back of the tongue, composed of shed keratinized epithelium, saliva, bacteria, food debris, and exudated white blood cells.

[0137] In this embodiment, the tongue coating detection unit 32 can specifically perform image segmentation processing on each first detection image to obtain the tongue coating region; then, the tongue coating detection unit 32 can perform color analysis on the tongue coating region through a color histogram to obtain the average color and color distribution of the tongue coating region; then, the tongue coating detection unit 32 can determine the second result based on the above average color and color distribution.

[0138] Since the tongue dimension includes the tongue area, ecchymosis, and pinpoints, the third result includes, but is not limited to, the tongue area, the number of ecchymosis, the total area of ​​ecchymosis, the number of pinpoints, and the total area of ​​pinpoints.

[0139] Among them, pinpoints refer to the clusters of red prickles on the tongue that protrude from the tongue surface.

[0140] In this embodiment, the tongue detection unit 33 can specifically perform image segmentation processing on each first detection image to obtain the tongue region. Then, the tongue detection unit 33 can count pixels in the tongue region to obtain the pixel area corresponding to the tongue region. Afterwards, the tongue detection unit 33 can convert the pixel area based on the device parameters of the image acquisition module 10 to obtain the actual tongue area.

[0141] Specifically, the tongue detection unit 33 can perform color analysis processing on each first detection image based on a color clustering algorithm to determine the ecchymosis region. Then, the tongue detection unit 33 can analyze the ecchymosis region based on morphological processing to obtain the number of ecchymosis spots and the total area of ​​ecchymosis pixels. Afterwards, the tongue detection unit 33 can convert the total area of ​​ecchymosis pixels based on the device parameters of the image acquisition module 10 to obtain the actual total area of ​​ecchymosis spots.

[0142] It should be noted that since the color of the ecchymosis is not the same as the color of the tongue, the tongue detection unit 33 can perform color clustering processing on each first detection image based on the color threshold to obtain the region that meets the color threshold and identify the region as the ecchymosis region.

[0143] In practical applications, morphological processing is used to extract useful image components that represent and depict the shape of regions, such as boundaries or skeletons. Morphological processing includes, but is not limited to, erosion, dilation, opening, closing, gradient operations, top-hat and bottom-hat operations, etc.

[0144] In this embodiment, the tongue detection unit 33 can specifically perform feature extraction processing on the tongue region to obtain the dotted area. Then, the tongue detection unit 33 can analyze the dotted area based on morphological processing to obtain the number of dots and the total area of ​​the dots.

[0145] In practical applications, lichen is used to describe the texture and shape of the tongue coating.

[0146] In this embodiment, the fourth result includes, but is not limited to, the area of ​​the tongue coating and the thickness of the tongue coating.

[0147] It should be noted that the thickness of the tongue coating is mainly examined on the middle and posterior two-thirds of the tongue dorsum.

[0148] Specifically, the tongue coating detection unit 34 can perform image segmentation processing on each of the first detection images to obtain the tongue coating region. Then, the tongue coating detection unit 34 can count pixels in the tongue coating region to obtain the pixel area corresponding to the tongue coating region. Afterwards, the tongue coating detection unit 34 can convert the aforementioned pixel area based on the device parameters of the image acquisition module 10 to obtain the actual tongue coating area.

[0149] Specifically, the tongue coating detection unit 34 can compare the segmented tongue coating area with a standard tongue coating image to determine the thickness of the tongue coating. The standard tongue coating image specifically refers to a tongue coating image with a standard tongue coating thickness.

[0150] In some possible embodiments, the thickness of the tongue coating can be described qualitatively.

[0151] In practical applications, qualitative description refers to summarizing and explaining the nature, characteristics, state, and relationships of things through textual expressions, rather than using quantitative values ​​or data to depict them.

[0152] For example, the thickness of the tongue coating can be described as no thickness, thin, relatively thin, medium, and relatively thick.

[0153] In this embodiment, the fifth result is used to describe the saliva state of the tongue.

[0154] The saliva detection unit 35 can specifically compare the distribution of each first detection image with a standard tongue image to determine the saliva state of the tongue coating. The standard tongue image specifically refers to a tongue image exhibiting a standard saliva state.

[0155] The state of body fluids includes, but is not limited to: insufficient body fluids, normal body fluids, and excessive body fluids.

[0156] In practical applications, cracks on the tongue refer to cracks or grooves of varying numbers, depths, and shapes that appear on the surface of the tongue.

[0157] In this embodiment, the sixth result includes, but is not limited to, the number of cracks and the length of each crack.

[0158] Specifically, the crack detection unit 36 ​​can perform image segmentation processing on each of the first detection images to obtain the tongue region. Then, the crack detection unit 36 ​​can determine crack features (such as elongation, linearity, etc.). Next, the crack detection unit 36 ​​can perform edge detection on the tongue region based on these crack features to detect the edges of each crack. Then, the crack detection unit 36 ​​can perform connected component analysis on each crack edge to determine the number of cracks. Finally, the crack detection unit 36 ​​can perform contour detection on each crack to determine the contour region where each crack is located, and calculate the length of each crack based on each contour region.

[0159] In practical applications, a connected region refers to an image region composed of foreground pixels with the same pixel value and adjacent positions. Connected region analysis refers to identifying and labeling each connected region in an image.

[0160] Contour detection is a computer vision technique used to identify and extract the boundaries of objects in an image. Through contour detection, the shapes and structures of different objects in an image can be found.

[0161] In this embodiment, the seventh result includes, but is not limited to, the number of tooth marks and the total area of ​​tooth marks.

[0162] Specifically, the tooth mark detection unit 37 can perform image segmentation processing on each of the first detection images to obtain the tongue region. Then, the tooth mark detection unit 37 can perform edge detection on the tongue region based on tooth mark features to detect the edges of each tooth mark. Next, the tooth mark detection unit 37 can perform morphological processing and connected component analysis on each tooth mark edge to determine the number of tooth marks and the area of ​​each tooth mark. Then, the tooth mark detection unit 37 can sum the areas of each tooth mark to obtain the total area of ​​the tooth mark pixels. Finally, the tooth mark detection unit 37 can convert the total area of ​​the tooth mark pixels based on the device parameters of the image acquisition module 10 to obtain the actual total area of ​​the tooth marks.

[0163] As can be seen from the above, the detection system provided in this embodiment can perform comprehensive detection of the user's tongue, improving the accuracy of tongue detection. Through the results corresponding to the different dimensions of the tongue, the physiological and pathological state of the human body can be more comprehensively reflected.

[0164] Please see Figure 4 , Figure 4 This is a schematic diagram of the specific structure of the facial detection module in the detection system provided in one embodiment of this application. It should be noted that, in this embodiment, the facial detection model includes: a facial gloss detection sub-model, an eye puffiness detection sub-model, a facial pigmentation detection sub-model, a facial acne detection sub-model, a lip thickness detection sub-model, and an eyebrow density detection sub-model.

[0165] Therefore, as Figure 4 As shown, the face detection module 40 specifically includes:

[0166] The facial gloss detection unit 41 is used to input each of the second detection images into the facial gloss detection sub-model for processing, and obtain the eighth result of the user's face in the facial gloss dimension.

[0167] The eye puffiness detection unit 42 is used to input each of the second detection images into the eye puffiness detection sub-model for processing, and obtain the ninth result of the user's face in the eye puffiness dimension.

[0168] The facial pigmentation detection unit 43 is used to input each of the second detection images into the facial pigmentation detection sub-model for processing, and obtain the tenth result of the user's face in the facial pigmentation dimension.

[0169] The facial acne detection unit 44 is used to input each of the second detection images into the facial acne detection sub-model for processing, so as to obtain the eleventh result of the user's face in the facial acne dimension.

[0170] The lip thickness detection unit 45 is used to input each of the second detection images into the lip thickness detection sub-model for processing, and obtain the twelfth result of the user's face in the lip thickness dimension.

[0171] The eyebrow density detection unit 46 is used to input each of the second detection images into the eyebrow density detection sub-model for processing, so as to obtain the thirteenth result of the user's face in the eyebrow density dimension.

[0172] In this embodiment, the eighth result is used to describe the degree of facial gloss (such as low gloss, medium gloss, and high gloss).

[0173] Specifically, the facial gloss detection unit 41 can extract texture features from each of the second detection images to obtain the facial texture features. Then, the facial gloss detection unit 41 can perform color space conversion processing on each of the second detection images to obtain various color images under a set color model. Next, the facial gloss detection unit 41 can perform set parameter analysis on each color image to obtain gloss analysis results. Finally, the facial gloss detection unit 41 can determine the eighth result based on the aforementioned texture features and gloss analysis results.

[0174] Among them, the color model includes, but is not limited to: Hue-Saturation-Value (HSV) model and Luminance-a-channel-b-channel (Lab) model.

[0175] Correspondingly, the parameters for the HSV model are hue, saturation, and lightness; the parameters for the Lab model include brightness value, etc.

[0176] In this embodiment, the ninth result is used to describe the degree of eye swelling (e.g., mild swelling, moderate swelling, and severe swelling).

[0177] Specifically, the eye puffiness detection unit 42 can perform facial key point detection processing on each second detection image to determine the position and contour of the eyes, thereby obtaining the eye region. Then, based on the eye region and a standard eye image, the eye puffiness detection unit 42 can determine the degree of shape change of the eyes within the eye region. Subsequently, the eye puffiness detection unit 42 can determine the eye puffiness result based on the degree of shape change. The eye puffiness result describes whether eye puffiness exists.

[0178] A standard eye image specifically refers to an image of the eyes when there is no puffiness.

[0179] In this embodiment, when the eye puffiness detection unit 42 detects the presence of eye puffiness, it can identify the puffy area within the eye region and determine the color distribution and fold depth of the puffy area. Subsequently, the eye puffiness detection unit 42 can determine a ninth result based on the color distribution and fold depth.

[0180] In this embodiment, the tenth result is used to describe the degree of pigmentation on the face (e.g., a small number of pigmentation spots, a moderate number of pigmentation spots, and a large number of pigmentation spots).

[0181] Since the color of the pigmented spots is different from the normal color of the face, the facial pigmented spot detection unit 43 can detect each second detection image based on a set color range, thereby obtaining the area within that color range and identifying that area as a pigmented spot area. Then, the facial pigmented spot detection unit 43 can determine the severity of the pigmentation based on the number of pigmented spots in each pigmented spot area.

[0182] In this embodiment, the eleventh result is used to describe the degree of acne on the face (e.g., a small number of acne, a moderate number of acne, and a large number of acne).

[0183] The facial acne detection unit 44 can specifically detect whether there are acne symptoms such as red papules, pustules, and blackheads in each of the second detection images to determine whether there is acne on the face. After detecting the presence of acne on the face, the facial acne detection unit 44 can determine the number and size of each acne based on morphological processing and connected region analysis, and obtain the eleventh result based on the number and size of each acne.

[0184] In this embodiment, the twelfth result is used to describe the thickness of the lips (e.g., thin lips, medium-thickness lips, and thick lips).

[0185] Specifically, the lip thickness detection unit 45 can perform facial key point detection processing on each of the second detection images to determine the position and contour of the lips, thereby obtaining the lip region. Then, the lip thickness detection unit 45 can measure the lip thickness from the highest point to the lowest point of the lip in the lip region. Afterwards, the lip thickness detection unit 45 can compare this thickness with a standard lip thickness to obtain a twelfth result.

[0186] In this embodiment, the thirteenth result is used to describe the density of the eyebrows (e.g., light eyebrows, medium-density eyebrows, and thick eyebrows).

[0187] The eyebrow density detection unit 46 can specifically perform facial key point detection processing on each second detection image to determine the position and outline of the eyebrows, thereby obtaining the eyebrow region. Then, the eyebrow density detection unit 46 can compare this eyebrow region with a standard eyebrow image to obtain the thirteenth result.

[0188] As can be seen from the above, the detection system provided in this embodiment can perform comprehensive detection of the user's face, improving the accuracy of facial detection. Through the results corresponding to the different facial dimensions, the physiological and pathological state of the human body can be more comprehensively reflected.

[0189] Please see Figure 5 , Figure 5 This is a schematic diagram of the detection system provided in another embodiment of this application. For example... Figure 5 As shown, the detection system 1 may further include:

[0190] The information storage module 60 is used to obtain the user's identity information when a detection command is detected; obtain the user's historical detection records and body information based on the identity information; and associate and store the identity information with the historical detection records and body information in the information database.

[0191] In one implementation of this embodiment, the detection system 1 is equipped with a detection control, which allows the user to trigger a detection command.

[0192] In another implementation of this embodiment, the detection system 1 determines that a detection command has been detected after detecting that the user is within the detection range of the detection system for a set period of time. The set period of time can be set according to actual needs and is not limited here.

[0193] In this embodiment, after detecting a detection command, the information storage module 60 in the detection system 1 can obtain the user's identity information to establish a user-generated electronic health record, facilitating the user's self-health management, timely understanding of their physical condition, and enabling doctors to continuously monitor the user's health status. This identity information includes, but is not limited to, basic information such as name, gender, and age.

[0194] In some possible embodiments, the user can input the aforementioned identity information on the display interface of the detection system 1, based on which the information storage module 60 can obtain the user's identity information.

[0195] In this embodiment, the information storage module 60 can obtain historical detection records and body information corresponding to the aforementioned identity information in real time through a server wirelessly connected to it. The server can be a cloud server.

[0196] Historical detection records are used to describe the historical tongue detection results, historical facial detection results, and historical vital sign detection results obtained by the user using the detection system 1 for health detection within a historical time period.

[0197] Body information is used to describe a user's basic medical conditions.

[0198] Subsequently, the information storage module 60 can associate and store the identity information with historical test records and physical information in the information database to build the user's electronic health record.

[0199] In one embodiment of this application, the information storage module 60 is further configured to associate and store the various tongue detection results, various facial detection results, and vital sign detection results with identity information in an information database to construct the user's latest electronic health record.

[0200] As can be seen from the above, the detection system provided in this embodiment stores various user information in association, which facilitates users' self-health management, timely understanding of their physical condition, and also facilitates doctors' continuous monitoring of users' health status.

[0201] Please see Figure 6 , Figure 6 This is a schematic diagram of the detection system provided in another embodiment of this application. For example... Figure 6 As shown, the detection system 1 may further include:

[0202] The information output module 70 is used to determine the user's constitution type based on the vital sign detection results; determine the user's symptom information based on the constitution type, each of the tongue detection results, and each of the facial detection results; determine the user's conditioning information in different conditioning dimensions based on the symptom information, and output the conditioning information.

[0203] In this embodiment, the results of the vital constitution test include, but are not limited to: the color of the tongue, the color of the tongue coating, the type of coating, the type of tongue shape, the state of body fluids, the overall color of the face, the state under the eyes, the location of chloasma, the location of freckles, and the location of acne.

[0204] Body constitution types include, but are not limited to, Yin deficiency constitution, Yang deficiency constitution, Qi deficiency constitution, Phlegm-dampness constitution, Blood stasis constitution, Damp-heat constitution, Special constitution, Damp-heat constitution, and Qi stagnation constitution.

[0205] Symptom information describes the user's condition, its common manifestations, and predisposing diseases. Symptoms include, but are not limited to, excess heat syndrome, cold-dampness syndrome, spleen deficiency, qi deficiency, and deficiency-cold syndrome.

[0206] Common symptoms include, but are not limited to: diarrhea, edema, stomach pain, dysmenorrhea, delayed menstruation (women), premature ejaculation and nocturnal emission (men), cough and wheezing, sore throat, irritability, insomnia, and restlessness.

[0207] Susceptible diseases include, but are not limited to: oral ulcers, sore throat, bleeding gums, constipation, gastroenteritis, acne, and high blood pressure.

[0208] Different dimensions of conditioning include, but are not limited to: traditional Chinese medicine, diet, exercise, daily life, acupoints, and psychology.

[0209] Based on this, the conditioning information in the TCM dimension can include edible Chinese medicinal herbs; the conditioning information in the dietary dimension includes edible and unsuitable foods; the conditioning information in the exercise dimension includes the exercise environment, exercise time, and exercise methods; the conditioning information in the daily life dimension includes the living environment, clothing suggestions, work environment, and living environment; the conditioning information in the acupoint dimension includes the acupoints that can be treated and the methods of treating those acupoints; and the conditioning information in the psychological dimension includes the mindset that can be maintained.

[0210] For example, taking the following vital sign test results as follows: tongue color is pale white, tongue coating color is white, coating type is thick and greasy, tongue shape is swollen and scalloped, body fluid status is slippery, overall facial color is red, under-eye condition is cold and dampness, chloasma is located around the eyes, freckles are located on the nasal wings, and acne is located on the chin, the information output module 70 can determine that the user's constitution type is Yang deficiency and cold syndrome, and determine the user's symptom information: deficiency and cold syndrome. Common manifestations include diarrhea, edema, stomach pain, dysmenorrhea and delayed menstruation (women), premature ejaculation and seminal emission (men), cough and asthma, sore throat, irritability and insomnia, and easy anger. Common diseases include oral ulcers, sore throat, bleeding gums, constipation, gastroenteritis, acne, and hypertension.

[0211] Therefore, the treatment information from the perspective of traditional Chinese medicine includes: for the treatment of spleen and stomach deficiency and cold, it is necessary to strengthen the spleen and replenish qi. You can take Fuzi Lizhong Wan, Xiaojianzhong Tang or Liangfu Wan, etc. The Chinese medicinal materials taken include ginseng, codonopsis, atractylodes macrocephala, dried ginger, aconite, cinnamon, cassia twig, galangal, evodia, pepper, Sichuan pepper, licorice, etc.

[0212] Dietary recommendations include avoiding cold and cooling foods that can damage the spleen's yang energy, and consuming foods that are warm in nature and sweet and spicy in taste, which have the effects of strengthening the spleen and replenishing qi, warming the stomach and intestines, and dispelling cold.

[0213] In terms of exercise, the conditioning information includes that people with deficiency-cold syndrome should choose to exercise in a warm or indoor environment, and try to avoid outdoor exercise during the lower temperature periods in the morning and evening. They can choose gentle exercise methods.

[0214] In terms of lifestyle adjustments, individuals with deficiency-cold syndrome should live in an environment with warm, gentle colors. They should pay attention to keeping warm, especially the abdomen, back, and limbs, and adjust clothing according to weather changes to avoid catching a cold. Prolonged work and living in dark, damp, or cold environments is not advisable.

[0215] Treatment information at the acupoint level includes moxibustion, acupuncture, foot baths, or massage at acupoints such as Shenque, Mingmen, Shenshu, Qihai, or Yongquan.

[0216] Psychological conditioning information includes maintaining emotional stability, regulating emotions and mindset, reducing the impact of negative emotions on the body, especially avoiding excessive worry, anxiety and fear, and cultivating an optimistic mindset, which can enhance the body's resistance and adaptability.

[0217] As can be seen from the above, the detection system provided in this embodiment can provide users with personalized treatment plans in different dimensions based on the user's vital sign detection results, which is targeted and practical.

[0218] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0219] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0220] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A detection system, characterized in that, include: The image acquisition module is used to acquire images of the user's tongue and face; The image processing module is used to divide the tongue image into a nine-grid structure to obtain a first detection image corresponding to each tongue detection area of ​​the user, and to divide the facial image into a fifteen-grid structure to obtain a second detection image corresponding to each facial detection area of ​​the user. The nine-square grid division specifically refers to dividing the tongue image into nine first detection images according to nine tongue detection areas. Each tongue detection area includes: the root area, the middle area, and the tip area. The root area includes the right side, the middle area, and the left side; the middle area includes the right side, the middle area, and the left side; and the tip area includes the right side, the middle area, and the left side. The root area corresponds to the large intestine, kidney, and small intestine in traditional Chinese medicine; the middle area corresponds to the gallbladder, liver, and spleen in traditional Chinese medicine; and the tip area corresponds to the lungs, stomach, and heart in traditional Chinese medicine. The tongue detection module is used to input each of the first detection images into the tongue detection model for processing, thereby obtaining the tongue detection results of the user's tongue in different tongue dimensions. The tongue detection model includes: a tongue color detection sub-model, a tongue coating color detection sub-model, a tongue texture detection sub-model, a tongue coating texture detection sub-model, a saliva detection sub-model, a crack detection sub-model, and a teeth mark detection sub-model. The tongue color detection sub-model, the tongue coating color detection sub-model, the tongue texture detection sub-model, the tongue coating texture detection sub-model, the saliva detection sub-model, the crack detection sub-model, and the teeth mark detection sub-model are all used to process each of the first detection images. The face detection module is used to input each of the second detection images into the face detection model for processing, and to obtain the face detection results of the user's face in different facial dimensions. The results analysis module is used to determine the user's vital signs detection results based on the tongue detection results and the facial detection results. The result analysis module is specifically used to: determine the state of each part of the user's body based on the tongue detection results and the facial detection results, and determine the state as the vital signs detection results.

2. The detection system as described in claim 1, characterized in that, The image processing module is specifically used for: A first analysis result is obtained by performing color analysis on the tongue image based on a standard tongue image and a first standard text; a second analysis result is obtained by performing color analysis on the face image based on a standard face image and a second standard text. Based on the first analysis result, a first correction method is determined, and the tongue image is corrected based on the first correction method to obtain a first corrected image; Based on the second analysis result, a second correction method is determined, and the facial image is corrected based on the second correction method to obtain a second corrected image; The first corrected image is divided into a nine-grid structure to obtain each first detection image, and the second corrected image is divided into a fifteen-grid structure to obtain each second detection image.

3. The detection system as described in claim 1, characterized in that, The tongue detection module specifically includes: The tongue color detection unit is used to input each of the first detection images into the tongue color detection sub-model for processing, and to obtain the first result of the user's tongue in the tongue color dimension. The moss color detection unit is used to input each of the first detection images into the moss color detection sub-model for processing, and obtain a second result of the user's tongue in the moss color dimension; The tongue detection unit is used to input each of the first detection images into the tongue detection sub-model for processing, and obtain the third result of the user's tongue in the tongue dimension. The moss detection unit is used to input each of the first detection images into the moss detection sub-model for processing, and to obtain a fourth result of the user's tongue in the moss dimension. The body fluid detection unit is used to input each of the first detection images into the body fluid detection sub-model for processing, and obtain the fifth result of the user's tongue in the body fluid dimension; A crack detection unit is used to input each of the first detection images into the crack detection sub-model for processing, and obtain the sixth result of the user's tongue in the crack dimension; The teeth mark detection unit is used to input each of the first detection images into the teeth mark detection sub-model for processing, and obtain the seventh result of the user's tongue in the teeth mark dimension.

4. The detection system as described in claim 1, characterized in that, The facial detection model includes: a facial luster detection sub-model, an eye puffiness detection sub-model, a facial pigmentation detection sub-model, a facial acne detection sub-model, a lip thickness detection sub-model, and an eyebrow density detection sub-model. Correspondingly, the facial detection module specifically includes: The facial gloss detection unit is used to input each of the second detection images into the facial gloss detection sub-model for processing, and obtain the eighth result of the user's face in the facial gloss dimension. An eye puffiness detection unit is used to input each of the second detection images into the eye puffiness detection sub-model for processing, and obtain the ninth result of the user's face in the eye puffiness dimension; The facial pigmentation detection unit is used to input each of the second detection images into the facial pigmentation detection sub-model for processing, and obtain the tenth result of the user's face in the facial pigmentation dimension; The facial acne detection unit is used to input each of the second detection images into the facial acne detection sub-model for processing, and to obtain the eleventh result of the user's face in the facial acne dimension. The lip thickness detection unit is used to input each of the second detection images into the lip thickness detection sub-model for processing, and obtain the twelfth result of the user's face in the lip thickness dimension. The eyebrow density detection unit is used to input each of the second detection images into the eyebrow density detection sub-model for processing, and obtain the thirteenth result of the user's face in the eyebrow density dimension.

5. The detection system as described in claim 3, characterized in that, The tongue body dimensions include tongue body area, ecchymosis, and pinpoints; the tongue body detection unit is specifically used for: Each of the first detection images is segmented to obtain the tongue region; the number of pixels in the tongue region is counted to obtain the pixel area corresponding to the tongue region; the pixel area is converted based on the device parameters of the image acquisition module to obtain the tongue area. Color clustering algorithm is used to perform color analysis on each of the first detection images to obtain ecchymosis regions; morphological processing is used to analyze the ecchymosis regions to obtain the number of ecchymosis and the total area of ​​ecchymosis. Feature extraction is performed on the tongue region to obtain the dotted region; the dotted region is analyzed based on morphological processing to obtain the number of dots and the total area of ​​dots.

6. The detection system as described in claim 3, characterized in that, The crack detection unit is specifically used for: Image segmentation processing is performed on each of the first detection images to obtain the tongue region; Determine the characteristics of the crack; Based on the crack features, edge detection is performed on the tongue region to obtain the edges of each crack. The number of cracks is obtained by performing connected region analysis on each crack edge. Each crack is subjected to contour detection to obtain the length of each crack.

7. The detection system as described in claim 4, characterized in that, The facial gloss detection unit is specifically used for: Texture features are extracted from each of the second detection images to obtain the texture features of the face; Each of the second detection images is subjected to color space conversion processing to obtain each color image under the set color model; Each color image is analyzed using set parameters to obtain gloss analysis results; the set parameters are determined by the set color model. Based on the texture features and the gloss analysis results, the eighth result is determined.

8. The detection system as described in claim 4, characterized in that, The eye puffiness detection unit is specifically used for: Facial key point detection processing is performed on each of the second detection images to determine the eye region; Based on the eye region and a standard eye image, determine the range of shape variation of the eye in the eye region; The result of eye puffiness is determined based on the magnitude of the shape change; If the result of the eye swelling is that swelling exists, then the swollen area in the eye region is determined, and the color distribution and wrinkle depth of the swollen area are determined; The ninth result is determined based on the color distribution and the wrinkle depth.

9. The detection system according to any one of claims 1-8, characterized in that, The detection system also includes: An information storage module is used to obtain the user's identity information when a detection command is detected; obtain the user's historical detection records and body information based on the identity information; and associate and store the identity information with the historical detection records and body information in an information database. The information storage module is also used to associate and store the tongue detection results, facial detection results, and vital sign detection results with the identity information in the information database.

10. The detection system according to any one of claims 1-8, characterized in that, The detection system also includes: The information output module is used to determine the user's constitution type based on the vital sign detection results; determine the user's symptom information based on the constitution type, each of the tongue detection results, and each of the facial detection results; determine the user's conditioning information in different conditioning dimensions based on the symptom information, and output the conditioning information.

11. A detection method, characterized in that, include: Collect images of the user's tongue and face; The tongue image is divided into a nine-grid structure to obtain the first detection image corresponding to each tongue detection area of ​​the user, and the facial image is divided into a fifteen-grid structure to obtain the second detection image corresponding to each facial detection area of ​​the user. The nine-square grid division specifically refers to dividing the tongue image into nine first detection images according to nine tongue detection areas. Each tongue detection area includes: the root area, the middle area, and the tip area. The root area includes the right side, the middle area, and the left side; the middle area includes the right side, the middle area, and the left side; and the tip area includes the right side, the middle area, and the left side. The root area corresponds to the large intestine, kidney, and small intestine in traditional Chinese medicine; the middle area corresponds to the gallbladder, liver, and spleen in traditional Chinese medicine; and the tip area corresponds to the lungs, stomach, and heart in traditional Chinese medicine. Each of the first detection images is input into the tongue detection model for processing to obtain the tongue detection results of the user's tongue in different tongue dimensions; the tongue detection model includes: tongue color detection sub-model, tongue coating color detection sub-model, tongue texture detection sub-model, tongue coating texture detection sub-model, saliva detection sub-model, crack detection sub-model, and teeth mark detection sub-model; the tongue color detection sub-model, the tongue coating color detection sub-model, the tongue texture detection sub-model, the tongue coating texture detection sub-model, the saliva detection sub-model, the crack detection sub-model, and the teeth mark detection sub-model are all used to process each of the first detection images; Each of the second detection images is input into the face detection model for processing to obtain the face detection results of the user's face in different facial dimensions; Based on the tongue detection results and the facial detection results, the user's vital signs detection results are determined; The step of determining the user's vital signs detection results based on the tongue detection results and the facial detection results includes: determining the state of each body part of the user based on the tongue detection results and the facial detection results, and determining the state as the vital signs detection results.

Images