Multi-region fusion intelligent tongue diagnosis system

Abstract

The multi-region fusion intelligent tongue diagnosis system comprises an intelligent terminal and a cloud server side, the intelligent terminal comprises a user side; the user side comprises a personal information management module, a tongue picture acquisition function module, an intelligent algorithm matcher module and a tongue picture report management module; the cloud server side comprises a central server and a tongue picture algorithm cluster system. According to the method, after the tongue picture data in the race-region-age database is subjected to feature extraction and fusion, traditional Chinese medicine constitution classification and viscera dialectical differentiation are performed through the tongue picture classification model, and the comprehensive evaluation result of the personal tongue picture is obtained in combination with the race skin color, the living region, the age and other basic data in the database; meanwhile, the model accuracy and generalization ability are improved by continuously increasing new data volume and adjusting model parameters, so that more accurate and intelligent tongue picture diagnosis can be carried out on the user, a personal tongue picture health library is generated, and the physique and disease change conditions of the user are dynamically monitored through the tongue picture.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of health diagnosis and treatment, and particularly relates to a multi-region fusion intelligent tongue diagnosis system. BACKGROUND

[0002] Tongue diagnosis is a method for a traditional Chinese medicine doctor to judge the health condition of a human body by observing tongue features such as tongue body, tongue texture, tongue fur and sublingual collaterals. Tongue features are also the most easily mastered and most easily objectivized item for constitution identification and disease diagnosis, and are one of the important bases for clinical diagnosis of traditional Chinese medicine. Clinical practice has proved that the deficiency and excess of viscera, the abundance and decline of blood and qi, the gain and loss of body fluid, and the depth of disease condition can be objectively reflected from tongue features. Therefore, taking tongue features as a starting point, reflecting the abundance and decline of yin and yang, cold and heat and deficiency and excess of the internal environment of the body through tongue features has great reference value for judging constitution characteristics and disease diagnosis. However, traditional tongue diagnosis relies more on the knowledge level and diagnosis experience of a traditional Chinese medicine doctor, and lacks objective means.

[0003] At present, with the development of big data, image processing and artificial intelligence technology, methods for judging diseases or constitutions by using tongue features are emerging in endlessly. Tongue features change dynamically with time, external environment and living habits, and tongue features of different regional populations are also different. Due to the difference of regionalization, intelligent and accurate tongue diagnosis cannot be realized. SUMMARY

[0004] The present application aims to provide a multi-region fusion intelligent tongue diagnosis system, which comprises an intelligent terminal and a cloud server end.

[0005] The intelligent terminal comprises a user end.

[0006] The user end comprises a personal information management module, a tongue feature collection function module, an intelligent algorithm matcher module and a tongue feature report management module.

[0007] The personal information management module is used for obtaining personal information of a user.

[0008] The tongue feature collection function module is used for collecting tongue feature images.

[0009] The intelligent algorithm matcher module matches corresponding algorithms based on the personal information of the user, and uses the matched algorithms to diagnose the tongue feature images to obtain tongue diagnosis results.

[0010] The tongue feature report management module is used for storing and recording a user tongue feature comprehensive evaluation report returned by the cloud server end.

[0011] The cloud server end comprises a center server and a tongue feature algorithm cluster system.

[0012] The center server is configured to receive and store data of the intelligent terminal and the tongue image algorithm cluster system.

[0013] The center server is configured to generate a user tongue image comprehensive evaluation report.

[0014] The tongue image algorithm cluster system stores an algorithm library and a tongue image database.

[0015] The tongue image algorithm cluster system uses algorithms in the algorithm library to diagnose tongue image, and weights and matches diagnosis results of different algorithms to obtain a tongue image comprehensive evaluation result.

[0016] Further, the intelligent terminal further comprises a consultant terminal.

[0017] The consultant terminal comprises a registration and login module, a professional certification module, a user consultation management module, a consultant labeling module, a consultant personal information management module, and a security certification module.

[0018] The consultant performs a registration and login operation through the registration and login module.

[0019] The professional certification module is configured to certify professional information and record a professional level of the consultant.

[0020] The user consultation management module comprises an online consultation service module, an offline appointment service module, and a user information search module.

[0021] The online consultation service module provides a platform for the consultant to communicate with the user online.

[0022] The content of the online communication includes but is not limited to traditional Chinese medicine health guidance suggestions, health knowledge popularization, and course learning.

[0023] The form of the online communication includes but is not limited to text and image communication, voice communication, and video communication.

[0024] The consultant provides offline appointment service and health management service for the user through the offline appointment service module.

[0025] The health management service includes but is not limited to traditional Chinese medicine conditioning, acupuncture and massage.

[0026] The consultant searches for relevant information of the user through the user information search module.

[0027] The relevant information of the user includes an ID and a contact method of the user.

[0028] The consultant performs constitution and zang-fu syndrome differentiation on the tongue image uploaded by the user terminal through the consultant labeling module, and verifies the user tongue image comprehensive evaluation report generated by the cloud server terminal.

[0029] The body constitution differentiation includes peaceful constitution, depressed constitution, deficient constitution, yin-deficient constitution, yang-deficient constitution, phlegm-damp constitution, damp-heat constitution, blood stasis constitution, and special constitution.

[0030] The organ differentiation includes heart and small intestine differentiation, lung and large intestine differentiation, spleen and stomach differentiation, liver and gallbladder differentiation, and kidney and bladder differentiation.

[0031] The personal information management module is used for managing and changing the personal basic information and professional information of the consultant.

[0032] The personal basic information of the consultant includes the name, gender, and age of the consultant.

[0033] The consultant performs secondary identity authentication when viewing and managing the user information through the security authentication module.

[0034] The secondary identity authentication mode includes one or more combinations of password authentication, fingerprint authentication, and face authentication.

[0035] Further, the intelligent terminal further includes a switching module.

[0036] The user terminal and the consultant terminal are switched through the switching module.

[0037] Further, the intelligent terminal further includes an integration module.

[0038] The user terminal, the consultant terminal, and the switching module are encapsulated in the integration module.

[0039] Further, the user terminal further includes a questionnaire filling module and a professional guidance suggestion module.

[0040] The questionnaire filling module obtains the life condition of the user through a questionnaire.

[0041] The life condition of the user includes dietary preference, sleep condition, emotional condition, exercise condition, and medication condition.

[0042] The professional guidance suggestion module is used for displaying the professional guidance suggestion provided by the consultant according to the tongue image comprehensive evaluation report.

[0043] The professional guidance suggestion includes but is not limited to emotional regulation, dietary regulation, living regulation, exercise health care, and acupoint health care.

[0044] Further, the personal information management module includes a user information input module, an automatic positioning module, and a user information history record module.

[0045] The user information input module includes a race information input module, a residence information input module, and a basic information input module.

[0046] The human race information input module is used to obtain the human race skin color information of the user.

[0047] The residence information input module includes a current residence information input module, a historical residence information input module, and a residence duration information input module.

[0048] The current residence information input module is used to obtain the location information of the current residence of the user.

[0049] The historical residence information input module is used to obtain the location information of the historical residence of the user.

[0050] The residence duration information input module is used to obtain the residence duration of the user in the current residence and the historical residence.

[0051] The basic information input module is used to obtain the basic information of the user.

[0052] The basic information includes but is not limited to ID, name, gender, age, address, and occupation.

[0053] The automatic positioning module is used to locate the location of the user and record the permanent residence of the user.

[0054] The user information history record module is used to store and manage the personal information of the user.

[0055] The personal information of the user includes the human race skin color information of the user, the location information of the current residence, the location information of the historical residence, the residence duration, the basic information, the location of the user, and the permanent residence information.

[0056] The tongue appearance collection function module includes a shooting module, a tongue appearance upload module, and a tongue appearance history record module.

[0057] The shooting module is used to shoot the tongue appearance image of the user.

[0058] The tongue appearance image includes a tongue surface image and a sublingual plexus image.

[0059] The tongue appearance upload module is used to transmit the tongue appearance image of the user to the cloud server side.

[0060] The tongue appearance history record module is used to store and record the shot tongue appearance image of the user.

[0061] The intelligent algorithm matcher module includes a single area algorithm matching module and a multiple area algorithm matching module.

[0062] The intelligent algorithm matcher module is matched by the residence duration of the user in the current residence and the historical residence.

[0063] When the user long-term lives in an area, the single area algorithm matching module selects the algorithm corresponding to the area to diagnose the tongue image and obtains the tongue diagnosis result.

[0064] When the user has multiple living areas, the multiple area algorithm matching module selects the algorithms corresponding to the multiple areas to comprehensively diagnose the tongue image and obtains the tongue diagnosis result.

[0065] Further, the center server comprises a data storage module, a tongue image report generation module, a data transmission module, and a database construction module.

[0066] The data storage module is configured to store data of the intelligent terminal and the tongue image algorithm cluster system.

[0067] The tongue image report generation module is configured to generate a user tongue image comprehensive evaluation report.

[0068] The data transmission module is configured to perform data interaction between the intelligent terminal and the cloud server.

[0069] The database construction module comprises a tongue image information module, a tongue image report module, a health maintenance opinion module, a user role table module, and a personal database module.

[0070] The tongue image information module is configured to store tongue diagnosis results of the user.

[0071] The tongue diagnosis result comprises tongue color, tongue fur, crack, and tongue shape category.

[0072] The tongue image report module is configured to analyze the tongue diagnosis result of the user, generate, and store a tongue image report of each user.

[0073] The tongue image report comprises a user constitution type and a tongue image category.

[0074] The health maintenance opinion module is configured to store a health maintenance scheme according to the tongue diagnosis result.

[0075] The health maintenance scheme comprises living health maintenance, diet health maintenance, exercise health maintenance, and sleep health maintenance.

[0076] The user role table module is configured to store login information of the user.

[0077] The personal database module comprises a personal information table module and a personal tongue image report library module.

[0078] The personal information table module is configured to store basic information of the user.

[0079] The basic information comprises, but is not limited to, ID, name, gender, age, address, and occupation.

[0080] The personal tongue image report library module is used for storing user tongue image comprehensive evaluation reports and displaying and analyzing the reports in the form of radar charts and line charts.

[0081] The user tongue image comprehensive evaluation report includes user basic information, user tongue image, user tongue diagnosis result and health preservation scheme.

[0082] Further, the tongue image algorithm cluster system includes a data acquisition module, a tongue image preprocessing module, a tongue image feature labeling module, a tongue image database construction module, an algorithm library module and a tongue image comprehensive evaluation module.

[0083] The data acquisition module includes a basic information acquisition module, a tongue image data acquisition module and a questionnaire acquisition module.

[0084] The basic information acquisition module is used for acquiring basic information of people of different skin colors, different living areas and different age groups, including but not limited to skin color, gender and occupation.

[0085] The tongue image data acquisition module includes a light source module, a camera module, a color card correction module and a tongue image storage module.

[0086] The light source module is used for providing an illumination environment.

[0087] The camera module provides a tongue image acquisition device for taking tongue image pictures.

[0088] The tongue image acquisition device includes but is not limited to a tongue diagnosis instrument and a smart phone.

[0089] The color card correction module is used for color correcting the tongue image pictures taken by the tongue image acquisition device to obtain corrected tongue image pictures.

[0090] The tongue image storage module is used for storing the corrected tongue image pictures.

[0091] The questionnaire acquisition module obtains corresponding traditional Chinese medicine constitution results by constitution and zang-fu syndrome differentiation of people of different skin colors, different living areas and different age groups through questionnaires.

[0092] The constitution differentiation includes gentle constitution, depressed constitution, deficient constitution, yin-deficient constitution, yang-deficient constitution, phlegm-damp constitution, damp-heat constitution, blood-stasis constitution and special constitution.

[0093] The zang-fu syndrome differentiation includes heart and small intestine syndrome differentiation, lung and large intestine syndrome differentiation, spleen and stomach syndrome differentiation, liver and gallbladder syndrome differentiation and kidney and bladder syndrome differentiation.

[0094] The tongue image preprocessing module is used for preprocessing the corrected tongue image pictures.

[0095] The preprocessing includes one or more of scaling, cropping, rotating and normalizing.

[0096] The tongue image feature labeling module is configured to label tongue image features of the preprocessed tongue image.

[0097] The tongue image features include tongue color, tongue shape, moss color, and moss quality.

[0098] The tongue image database construction module classifies the labeled tongue image to construct a race-region-age tongue image database.

[0099] The algorithm library module includes a race tongue diagnosis algorithm library module, a region tongue diagnosis algorithm library module, an age tongue diagnosis algorithm library module, and an algorithm library updating module.

[0100] The race tongue diagnosis algorithm library module stores tongue diagnosis algorithms for different races.

[0101] The region tongue diagnosis algorithm library module stores tongue diagnosis algorithms for different regions.

[0102] The age tongue diagnosis algorithm library module stores tongue diagnosis algorithms for different ages.

[0103] The algorithm library updating module updates the algorithm library by adjusting some parameters of the tongue diagnosis algorithm or increasing the amount of data.

[0104] The tongue image comprehensive evaluation module selects corresponding algorithms according to race, region, and age, and performs weight matching on the diagnosis results of different algorithms to obtain a tongue image comprehensive evaluation result.

[0105] Further, the tongue image comprehensive evaluation result is as follows:

[0106] TC score = T P · ω1 + T Q · ω2 + T A · ω3 (1)

[0107] In the formula, TC score is the tongue image comprehensive evaluation result. T P represents the race tongue image algorithm evaluation result, ω1 is the weight of the race, ω2 is the weight of the region, T A represents the age tongue image algorithm evaluation result, and ω3 is the weight of the age.

[0108] In the formula, i represents the region index, m represents the total number of regions, Ti represents the output result of the i-th region algorithm, and γ Q is as follows:

[0109]

[0110] In the formula, i represents the region index, m represents the total number of regions, Ti represents the output result of the i-th region algorithm, and γi Represents the weight of the algorithm for the i-th residential area.

[0111] Furthermore, the steps for constructing the algorithm library in the tongue image algorithm cluster system are as follows:

[0112] S1 collects basic information and corresponding tongue images of people of different races, skin colors, different residential areas and different age groups.

[0113] s2 labels tongue images with tongue features, race, skin color, residential area, and age group.

[0114] s3 constructs training set I with race and skin color as input and tongue image features as output.

[0115] The training set II is constructed with residential area as input and tongue image features as output.

[0116] The training set III is constructed with age stage as input and tongue image features as output.

[0117] s4 constructs ethnic tongue diagnosis algorithm models, regional tongue diagnosis algorithm models, and age tongue diagnosis algorithm models.

[0118] s5 trains the ethnic tongue diagnosis algorithm model with training set I to obtain a trained ethnic tongue diagnosis algorithm model.

[0119] The regional tongue diagnosis algorithm model is trained with training set II to obtain a trained regional tongue diagnosis algorithm model.

[0120] The age tongue diagnosis algorithm model is trained with training set III to obtain a trained age tongue diagnosis algorithm model.

[0121] S6 builds an ethnic tongue diagnosis algorithm library based on the trained ethnic tongue diagnosis algorithm model.

[0122] Based on the trained ethnic regional tongue diagnosis algorithm model, a regional tongue diagnosis algorithm library is constructed.

[0123] Based on the trained ethnicity and age tongue diagnosis algorithm model, an age tongue diagnosis algorithm library is constructed.

[0124] The technical effect of the present invention is unquestionable. After extracting and fusing the features of the tongue image data in the race-region-age database, the present invention uses the tongue image classification model to perform TCM constitution classification and viscera dialectics, and combines the basic data such as race, skin color, residential area and age in the database to obtain a comprehensive evaluation result of the individual tongue image. At the same time, by continuously adding new data and adjusting model parameters, the model accuracy and generalization ability are improved, thereby enabling more accurate intelligent tongue image diagnosis of users, and generating a personal tongue image health library, and dynamically monitoring the user's constitution and disease changes through tongue images. BRIEF DESCRIPTION OF THE DRAWINGS

[0125] Figure 1 System structure diagram of the embodiment of the present application Figure 1 ;

[0126] Figure 2 System structure diagram of the embodiment of the present application Figure 2 ;

[0127] Figure 3 System structure diagram of the embodiment of the present application Figure 3 ;

[0128] Figure 4 System structure diagram of the embodiment of the present application

[0129] Figure 5 System structure diagram of the embodiment of the present application

[0130] Figure 6 System structure diagram of the embodiment of the present application

[0131] Figure 7 System structure diagram of the embodiment of the present application

[0132] Figure 8 System structure diagram of the embodiment of the present application

[0133] Figure 9 System structure diagram of the embodiment of the present application

[0134] Figure 10 System structure diagram of the embodiment of the present application

[0135] Figure 11 System structure diagram of the embodiment of the present application

[0136] Figure 12 System structure diagram of the embodiment of the present application

[0137] Figure 13 System structure diagram of the embodiment of the present application

[0138] In the figure: intelligent terminal 1000, cloud server end 2000, center server 2100, data storage module 2110, tongue appearance report generation module 2120, data transmission module 2130, database construction module 2140, tongue appearance information module 2141, tongue appearance report module 2142, health preservation opinion module 2143, user role table module 2144, personal database module 2145, personal information table module 21451, personal tongue appearance report library module 21452, tongue appearance algorithm cluster system 2200, data acquisition module 2210, basic information acquisition module 2211, tongue appearance data acquisition module 2212, light source module 22121, camera module 22122, color card correction module 22123, tongue appearance storage module 22124, questionnaire acquisition module 2213, tongue appearance preprocessing module 2220, tongue appearance feature labeling module 2230, tongue appearance database construction module 2240, algorithm library module 2250, race tongue diagnosis algorithm library module 2251, region tongue diagnosis algorithm library module 2252, age tongue diagnosis algorithm library module 2253, algorithm library updating module 2254, tongue appearance comprehensive evaluation module 2260, user end 3000, personal information management module 3100, user information input module 3110, race information input module 3111, residence information input module 3112, current residence information input module 31121, historical residence information input module 31122, residence duration information input module 31123, basic information input module 3113, automatic positioning module 3120, user information history record module 3130, tongue appearance acquisition function module 3200, shooting module 3210, tongue appearance uploading module 3220, tongue appearance history record module 3230, intelligent algorithm matcher module 3300, single region algorithm matching module 3310, multiple region algorithm matching module 3320, questionnaire filling module 3400, tongue appearance report management module 3500, professional guidance suggestion module 3600, consultant end 4000, registration login module 4100, professional certification module 4200, user consultation management module 4300, online consultation service module 4310, offline appointment service module 4320, user information search module 4330, consultant labeling module 4400, consultant personal information management module 4500, security authentication module 4600, switching module 5000, integrated module 6000. DETAILED DESCRIPTION

[0139] The application will be further described in conjunction with the examples below, but should not be understood as limiting the above-mentioned subject matter of the application to the following examples. Various replacements and changes can be made according to ordinary technical knowledge and conventional means in the art without departing from the above-mentioned technical idea of the application, and all of them should be included in the protection scope of the application.

[0140] Example 1

[0141] Referring to Figures 1 to 13 A multi-region fusion intelligent tongue diagnosis system, comprising: an intelligent terminal 1000 and a cloud server end 2000.

[0142] The intelligent terminal 1000 comprises a user end 3000.

[0143] The user end 3000 comprises a personal information management module 3100, a tongue image collection function module 3200, an intelligent algorithm matcher module 3300, and a tongue image report management module 3500.

[0144] The personal information management module 3100 is configured to obtain personal information of a user.

[0145] The tongue image collection function module 3200 is configured to collect tongue image.

[0146] The intelligent algorithm matcher module 3300 is configured to match corresponding algorithms based on the personal information of the user, and diagnose the tongue image by using the matched algorithms to obtain a tongue diagnosis result.

[0147] The tongue image report management module 3500 is configured to store and record a user tongue image comprehensive evaluation report returned by the cloud server end 2000.

[0148] The cloud server end 2000 comprises a center server 2100 and a tongue image algorithm cluster system 2200.

[0149] The center server 2100 is configured to receive and store data of the intelligent terminal 1000 and the tongue image algorithm cluster system 2200.

[0150] The center server 2100 is configured to generate a user tongue image comprehensive evaluation report.

[0151] The tongue image algorithm cluster system 2200 stores an algorithm library and a tongue image database.

[0152] The tongue image algorithm cluster system 2200 is configured to diagnose the tongue image by using algorithms in the algorithm library, and weight and proportion diagnosis results of different algorithms to obtain a tongue image comprehensive evaluation result.

[0153] Embodiment 2

[0154] A multi-region fusion intelligent tongue diagnosis system, the main technical content of which is shown in Embodiment 1, further, the intelligent terminal 1000 further comprises a consultant end 4000.

[0155] The consultant end 4000 comprises a registration and login module 4100, a professional certification module 4200, a user consultation management module 4300, a consultant labeling module 4400, a consultant personal information management module 4500, and a security certification module 4600.

[0156] The consultant registers and logs in through the registration and login module 4100.

[0157] The professional certification module 4200 is used to certify the professional information and record the job level of the consultant.

[0158] The user consultation management module 4300 includes an online consultation service module 4310, an offline appointment service module 4320, and a user information search module 4330.

[0159] The online consultation service module 4310 provides a platform for the consultant to communicate with the user online.

[0160] The content of the online communication includes but is not limited to traditional Chinese medicine health guidance suggestions, health knowledge popularization, and course learning.

[0161] The form of online communication includes but is not limited to text and image communication, voice communication, and video communication.

[0162] The consultant provides offline appointment service and health management service for the user through the offline appointment service module 4320.

[0163] The health management service includes but is not limited to traditional Chinese medicine conditioning, acupuncture and massage.

[0164] The consultant searches for the relevant information of the user through the user information search module 4330.

[0165] The relevant information of the user includes the user's ID and contact information.

[0166] The consultant performs constitution and zang-fu syndrome differentiation on the tongue image uploaded by the user terminal 3000 through the consultant marking module 4400, and verifies the user tongue comprehensive evaluation report generated by the cloud server 2000.

[0167] The constitution differentiation includes gentle constitution, depressed constitution, deficient constitution, yin-deficient constitution, yang-deficient constitution, phlegm-damp constitution, damp-heat constitution, blood-stasis constitution, and special constitution.

[0168] The zang-fu differentiation includes heart and small intestine differentiation, lung and large intestine differentiation, spleen and stomach differentiation, liver and gallbladder differentiation, and kidney and bladder differentiation.

[0169] The consultant personal information management module 4500 is used to manage and change the personal basic information and professional information of the consultant.

[0170] The personal basic information of the consultant includes the consultant's name, gender, and age.

[0171] The consultant performs secondary identity authentication when viewing and managing user information through the security authentication module 4600.

[0172] The secondary identity authentication mode includes one or more combinations of password authentication, fingerprint authentication, and face authentication.

[0173] Embodiment 3:

[0174] A multi-region fusion intelligent tongue diagnosis system, the main technical content is seen in any one of embodiments 1 to 2, further, the intelligent terminal 1000 further includes a switching module 5000.

[0175] The user end 3000 and the consultant end 4000 are switched through the switching module 5000.

[0176] Embodiment 4:

[0177] A multi-region fusion intelligent tongue diagnosis system, the main technical content is seen in any one of embodiments 1 to 3, further, the intelligent terminal 1000 further includes an integrated module 6000.

[0178] The user end 3000, the consultant end 4000 and the switching module 5000 are encapsulated in the integrated module 6000.

[0179] Embodiment 5:

[0180] A multi-region fusion intelligent tongue diagnosis system, the main technical content is seen in any one of embodiments 1 to 4, further, the user end 3000 further includes a questionnaire filling module 3400 and a professional guidance suggestion module 3600.

[0181] The questionnaire filling module 3400 obtains the user's living conditions through a questionnaire form.

[0182] The user's living conditions include dietary preferences, sleep conditions, emotional conditions, exercise conditions, and medication conditions.

[0183] The professional guidance suggestion module 3600 is used to display professional guidance suggestions provided by consultants according to user tongue image comprehensive evaluation reports.

[0184] The professional guidance suggestions include, but are not limited to, emotional regulation, dietary care, living regulation, exercise health care, and acupoint health care.

[0185] Embodiment 6:

[0186] A multi-region fusion intelligent tongue diagnosis system, the main technical content is seen in any one of embodiments 1 to 5, further, the personal information management module 3100 includes a user information input module 3110, an automatic positioning module 3120, and a user information history record module 3130.

[0187] The user information input module 3110 includes a race information input module 3111, a residence information input module 3112, and a basic information input module 3113.

[0188] The race information input module 3111 is configured to obtain the race and skin color information of the user.

[0189] The residence information input module 3112 includes a current residence information input module 31121, a historical residence information input module 31122, and a residence duration information input module 31123.

[0190] The current residence information input module 31121 is configured to obtain the location information of the current residence of the user.

[0191] The historical residence information input module 31122 is configured to obtain the location information of the historical residence of the user.

[0192] The residence duration information input module 31123 is configured to obtain the residence duration of the user in the current residence and the historical residence.

[0193] The basic information input module 3113 is configured to obtain the basic information of the user.

[0194] The basic information includes, but is not limited to, ID, name, gender, age, address, and occupation.

[0195] The automatic positioning module 3120 is configured to locate the location of the user and record the permanent residence of the user.

[0196] The user information history record module 3130 is configured to store and manage the personal information of the user.

[0197] The personal information of the user includes the race and skin color information of the user, the location information of the current residence, the location information of the historical residence, the residence duration, the basic information, the location of the user, and the permanent residence information.

[0198] The tongue appearance collection function module 3200 includes a photographing module 3210, a tongue appearance uploading module 3220, and a tongue appearance history record module 3230.

[0199] The photographing module 3210 is configured to photograph the tongue appearance image of the user.

[0200] The tongue appearance image includes a tongue surface image and a sublingual plexus image.

[0201] The tongue appearance uploading module 3220 is configured to transmit the tongue appearance image of the user to the cloud server end 2000.

[0202] The tongue appearance history record module 3230 is configured to store and record the photographed tongue appearance image of the user.

[0203] The intelligent algorithm matcher module 3300 includes a single region algorithm matching module 3310 and a multiple region algorithm matching module 3320.

[0204] The intelligent algorithm matcher module 3300 is an algorithm matching based on the length of residence of the user in the current residence and the historical residence.

[0205] When the user has long-term residence in a region, the single region algorithm matching module 3310 selects an algorithm corresponding to the region to diagnose the tongue image and obtain a tongue diagnosis result.

[0206] When the user has multiple residence regions, the multiple region algorithm matching module 3320 selects multiple algorithms corresponding to the multiple regions to comprehensively diagnose the tongue image and obtain a tongue diagnosis result.

[0207] The comprehensive diagnosis here is a tongue diagnosis result obtained by assigning a corresponding weight to the diagnosis results of different regions, which can be regarded as a regional tongue diagnosis result for multiple regions.

[0208] Embodiment 7:

[0209] A multi-region fusion intelligent tongue diagnosis system, the main technical content of which is any one of embodiments 1 to 6, further, the center server 2100 includes a data storage module 2110, a tongue image report generation module 2120, a data transmission module 2130, and a database construction module 2140.

[0210] The data storage module 2110 is configured to store data of the intelligent terminal 1000 and the tongue image algorithm cluster system 2200.

[0211] The tongue image report generation module 2120 is configured to generate a user tongue image comprehensive evaluation report.

[0212] The tongue image report generation module 2120 is configured to generate a user tongue image comprehensive evaluation report according to the tongue image comprehensive evaluation result in the tongue image algorithm cluster system 2200 and the collected user related information (such as gender, age, etc.).

[0213] The data transmission module 2130 is configured to perform data interaction between the intelligent terminal 1000 and the cloud server 2000.

[0214] The database construction module 2140 includes a tongue image information module 2141, a tongue image report module 2142, a health preservation opinion module 2143, a user role table module 2144, and a personal database module 2145.

[0215] The tongue image information module 2141 is configured to store a user tongue diagnosis result.

[0216] The tongue diagnosis result includes tongue color, tongue fur, crack, and tongue shape category.

[0217] The tongue appearance report module 2142 analyzes the tongue diagnosis results of the user, generates and stores the tongue appearance report of each user.

[0218] The tongue appearance report module 2142 is mainly the tongue appearance report of each user in the database.

[0219] The tongue appearance report includes the user's physical type and tongue appearance category.

[0220] The health maintenance opinion module 2143 is used to store the health maintenance scheme according to the tongue diagnosis results.

[0221] The health maintenance scheme includes health maintenance of living, diet, exercise and sleep.

[0222] The user role table module 2144 is used to store the login information of the user.

[0223] The personal database module 2145 includes a personal information table module 21451 and a personal tongue appearance report library module 21452.

[0224] The personal information table module 21451 is used to store the basic information of the user.

[0225] The basic information includes but is not limited to ID, name, gender, age, address, occupation.

[0226] The personal tongue appearance report library module 21452 is used to store the user's tongue appearance comprehensive evaluation report and display and analyze it in the form of radar chart and line chart.

[0227] The user's tongue appearance comprehensive evaluation report includes the user's basic information, tongue image, tongue diagnosis results and health maintenance scheme.

[0228] The tongue appearance comprehensive evaluation includes:

[0229] 1. The user's personal information, such as nickname, race, gender, age, residence, occupation, etc.

[0230] 2. User tongue appearance information (tongue image uploaded by the user);

[0231] 3. User tongue diagnosis results: including tongue comprehensive evaluation results and each zang-fu tongue diagnosis result;

[0232] 4. Conditioning suggestions for constitution and tongue diagnosis results, including diet, living, exercise, and emotional aspects.

[0233] Embodiment 8:

[0234] The main technical content of a multi-region fusion intelligent tongue diagnosis system is shown in any one of embodiments 1 to 7. Further, the tongue image algorithm cluster system 2200 comprises a data acquisition module 2210, a tongue image preprocessing module 2220, a tongue image feature labeling module 2230, a tongue image database construction module 2240, an algorithm library module 2250, and a tongue image comprehensive evaluation module 2260.

[0235] The data acquisition module 2210 comprises a basic information acquisition module 2211, a tongue image data acquisition module 2212, and a questionnaire acquisition module 2213.

[0236] The basic information acquisition module 2211 is used to acquire basic information of different ethnic skin colors, different residential areas, and different age groups, including but not limited to skin color, gender, and occupation.

[0237] The tongue image data acquisition module 2212 comprises a light source module 22121, a camera module 22122, a color card correction module 22123, and a tongue image storage module 22124.

[0238] The light source module 22121 is used to provide an illumination environment.

[0239] The camera module 22122 provides a tongue image acquisition device for capturing tongue image pictures.

[0240] The tongue image acquisition device includes but is not limited to a tongue diagnosis instrument and a smart phone.

[0241] The color card correction module 22123 is used to correct the tongue image pictures captured by the tongue image acquisition device to obtain corrected tongue image pictures.

[0242] The tongue image storage module 22124 is used to store the corrected tongue image pictures.

[0243] The questionnaire acquisition module 2213 uses a questionnaire to determine the constitution and zang-fu syndrome of different ethnic skin colors, different residential areas, and different age groups, and obtains the corresponding traditional Chinese medicine constitution results.

[0244] The constitution syndrome includes gentle constitution, depressed constitution, deficient constitution, yin-deficient constitution, yang-deficient constitution, phlegm-damp constitution, damp-heat constitution, blood-stasis constitution, and special constitution.

[0245] The zang-fu syndrome includes heart and small intestine syndrome, lung and large intestine syndrome, spleen and stomach syndrome, liver and gallbladder syndrome, and kidney and bladder syndrome.

[0246] The tongue image preprocessing module 2220 is used to preprocess the corrected tongue image pictures.

[0247] The preprocessing includes one or more of scaling, cropping, rotating, and normalizing.

[0248] The tongue image feature labeling module 2230 is configured to label the tongue image features of the preprocessed tongue image.

[0249] The tongue image features include tongue color, tongue shape, moss color, and moss quality.

[0250] The tongue image database construction module 2240 classifies the labeled tongue image to construct a race-region-age tongue image database.

[0251] The algorithm library module 2250 includes a race tongue diagnosis algorithm library module 2251, a region tongue diagnosis algorithm library module 2252, an age tongue diagnosis algorithm library module 2253, and an algorithm library updating module 2254.

[0252] The race tongue diagnosis algorithm library module 2251 stores tongue diagnosis algorithms for different races.

[0253] The region tongue diagnosis algorithm library module 2252 stores tongue diagnosis algorithms for different regions.

[0254] The age tongue diagnosis algorithm library module 2253 stores tongue diagnosis algorithms for different ages.

[0255] The algorithm library updating module 2254 updates the algorithm library by adjusting some parameters of the tongue diagnosis algorithm or increasing the amount of data.

[0256] The tongue image comprehensive evaluation module 2260 selects corresponding algorithms according to race, region, and age, and performs weight matching on the diagnosis results of different algorithms to obtain a tongue image comprehensive evaluation result.

[0257] Embodiment 9:

[0258] A multi-region fusion intelligent tongue diagnosis system, the main technical content of which is shown in any one of embodiments 1 to 8, further, the tongue image comprehensive evaluation result is as follows:

[0259] TC score = T P · ω1 + T Q · ω2 + T A · ω3 (1)

[0260] In the formula, TC score is the tongue image comprehensive evaluation result. T P represents the race tongue image algorithm evaluation result, ω1 is the weight of the race, ω2 is the weight of the region, T A represents the age tongue image algorithm evaluation result, and ω3 is the weight of the age.

[0261] In the formula, the region tongue image algorithm evaluation result T Q is as follows:

[0262]

[0263] wherein i represents a residence area index, m represents a total number of residence areas, Ti represents an output result of the i-th residence area algorithm, γ i represents a weight of the i-th residence area algorithm.

[0264] Embodiment 10:

[0265] A multi-region fusion intelligent tongue diagnosis system, the main technical content is seen in any one of embodiments 1 to 9, further, the algorithm library in the tongue image algorithm cluster system 2200 is constructed as follows:

[0266] s1 collects basic information and corresponding tongue image of people of different skin colors, different residence areas and different age stages.

[0267] s2 annotates the tongue image with tongue features, skin color, residence area and age stage.

[0268] s3 constructs a training set I with skin color as input and tongue features as output.

[0269] A training set II is constructed with residence area as input and tongue features as output.

[0270] A training set III is constructed with age stage as input and tongue features as output.

[0271] s4 constructs a race tongue diagnosis algorithm model, a region tongue diagnosis algorithm model and an age tongue diagnosis algorithm model.

[0272] s5 trains the race tongue diagnosis algorithm model with the training set I to obtain a trained race tongue diagnosis algorithm model.

[0273] The region tongue diagnosis algorithm model is trained with the training set II to obtain a trained region tongue diagnosis algorithm model.

[0274] The age tongue diagnosis algorithm model is trained with the training set III to obtain a trained age tongue diagnosis algorithm model.

[0275] s6 constructs a race tongue diagnosis algorithm library based on the trained race tongue diagnosis algorithm model.

[0276] A region tongue diagnosis algorithm library is constructed based on the trained race region tongue diagnosis algorithm model.

[0277] An age tongue diagnosis algorithm library is constructed based on the trained race age tongue diagnosis algorithm model.

[0278] Embodiment 11:

[0279] A multi-region fusion intelligent tongue diagnosis system, the main technical content is shown in any one of embodiments 1 to 10, further, the system operation steps are as follows:

[0280] P1 user fills in the basic information of the individual, and uses the mobile terminal to take a picture of the tongue and adjust the tongue contour;

[0281] P2 user uploads the filled basic information and the photographed tongue image to the cloud server end;

[0282] P3 according to the information of user filled in, such as race, gender, age, place of residence and residence time, the corresponding algorithm is called in the cloud server end to extract the tongue image features and obtain the constitution identification and disease diagnosis results;

[0283] P4 the tongue image analysis report generated by the cloud server end is transmitted to the user end;

[0284] P5 user can get the corresponding professional conditioning suggestion and guidance according to the constitution result in the report, and can also get more detailed conditioning suggestions through consulting consultants.

[0285] Embodiment 12:

[0286] A method of applying the system of any one of embodiments 1 to 11, the main technical content includes:

[0287] 1) obtain the tongue image, which contains the tongue image of the face, lip and tooth background, etc.;

[0288] 2) use the labeling software to label the tongue contour of the tongue image;

[0289] 3) input the tongue image data set with completed tongue labeling into the tongue segmentation network model, and the output result is the segmented tongue image; the tongue segmentation model is constructed based on the ResNext network structure, mainly composed of an encoder and a decoder, in addition, an attention mechanism module is added between the convolution layers of the residual connection, which is used to improve the performance of the network for image segmentation;

[0290] 4) extract the features of the segmented tongue image; the tongue image features include but are not limited to tongue color, tongue shape, moss color, texture feature, target area, tongue pulse feature;

[0291] 5) according to the extracted tongue image features, form a feature vector, and carry out pretreatment such as dimension reduction;

[0292] 6) input the pretreated feature vector into the classifier to obtain the mapping relationship between the tongue image and the disease and constitution.

[0293] Embodiment 13:

[0294] Referring to Figures 1 to 13The application discloses a multi-region fusion intelligent tongue diagnosis system, and mainly relates to the following technical contents.

[0295] The intelligent terminal 1000 comprises a user terminal 3000 and a consultant terminal 4000.

[0296] The user terminal 3000 comprises a personal information management module 3100, a tongue image collection function module 3200, an intelligent algorithm matcher module 3300, a questionnaire filling module 3400, a tongue image report management module 3500 and a professional guidance suggestion module 3600.

[0297] The consultant terminal 4000 comprises a consultant registration and login module 4100, a professional certification module 4200, a user consultation management module 4300, a consultant labeling module 4400, a consultant personal information management module 4500 and a security certification module 4600.

[0298] The center server 2100 comprises a data storage module 2110, a tongue image report generation module 2120, a data transmission module 2130 and a database construction module 2140.

[0299] The data storage module 2110 is used for storing data of the intelligent terminal 1000 and the cloud server 2000.

[0300] The tongue image report generation module 2120 is used for generating a tongue image comprehensive evaluation report of a user and transmitting the report to the intelligent terminal 1000.

[0301] The data transmission module 2130 is used for data interaction with the intelligent terminal 1000.

[0302] The database construction module 2140 is used for constructing a database comprising but not limited to basic features of tongue images, physical features, personal information of users and the like.

[0303] The tongue image algorithm cluster system 2200 processes tongue image data of different races, different regions and different age stages through independent tongue diagnosis algorithms to obtain tongue diagnosis result data and returns the data to the center server 2100.

[0304] The user terminal 3000 comprises a personal information management module 3100, a tongue image collection function module 3200, an intelligent algorithm matcher module 3300, a questionnaire filling module 3400, a tongue image report management module 3500 and a professional guidance suggestion module 3600.

[0305] The personal information management module 3100 includes a user information input module 3110, an automatic positioning module 3120, and a user information history record module 3130.

[0306] The user information input module 3110 includes a race information input module 3111, a residence information input module 3112, and a basic information input module 3113.

[0307] The race information input module 3111 is configured to obtain the race and skin color information of the user and transmit the information to the cloud server 2000.

[0308] The residence information input module 3112 includes a current residence information input module 31121, a historical residence information input module 31122, and a residence duration information input module 31123, which are configured to obtain the region and residence duration of the current and historical residence of the user and transmit the information to the cloud server 2000.

[0309] The current residence information input module 31121 is configured to obtain the region and location of the current residence of the user.

[0310] The historical residence information input module 31122 is configured to obtain the region and location of the historical residence of the user.

[0311] The residence duration information input module 31123 is configured to obtain the residence duration of the current and historical residence of the user.

[0312] The residence duration T P0 is calculated according to the following formula:

[0313] T P0 = T P1 -T P2

[0314] wherein, T P1 is the time of leaving the current residence region, and T P2 is the time of arriving at the residence region.

[0315] The basic information input module 3113 is configured to obtain the basic information of the user and transmit the information to the cloud server 2000, wherein the basic information includes but is not limited to gender, age, height, weight, occupation type, and income status.

[0316] The automatic positioning module 3120 is configured to position the residence of the user and record the location of the residence of the user.

[0317] The user information history record module 3130 is configured to store and manage the basic information of the user.

[0318] The user information history record module includes a basic information history record unit, a race and skin color information history record unit, and a residence information history record unit.

[0319] The basic information history record unit is configured to display and count the basic information of the user.

[0320] The race and skin color information history record unit is configured to display and count the race and skin color information of the user.

[0321] The residence information history record unit is configured to display and count the location information of the residence area of the user and the residence duration information of the area.

[0322] The tongue image collection function module 3200 includes a photographing module 3211, a tongue image uploading module 3212, and a tongue image history record module 3213.

[0323] The photographing module 3211 photographs the tongue image of the user and transmits the tongue image to the tongue image uploading module 3212. The tongue image includes a tongue surface image and a sublingual plexus image.

[0324] The tongue image uploading module 3212 transmits the photographed tongue image to the cloud server end 2000.

[0325] The tongue image history record module 3213 is configured to store and record the tongue image uploading record and tongue image data of the user.

[0326] The intelligent algorithm selector module 3300 is configured to automatically select a corresponding algorithm model according to the race and skin color, residence area, and age of the user.

[0327] The questionnaire filling module 3400 is configured to enable the user to fill in a questionnaire according to the actual situation of the user and upload the filled data to the cloud server end 2000.

[0328] The tongue image report management module 3500 is configured to store and record the tongue image report of the user returned from the cloud server end 2000.

[0329] The professional guidance suggestion module 3600 is configured to enable the consultant to provide a corresponding professional guidance suggestion according to the tongue image report of the user.

[0330] The professional guidance suggestion includes but is not limited to the following scheme (taking a person with qi deficiency as an example):

[0331] Table 1 professional guidance suggestion

[0332]

[0333] The intelligent terminal 1000 further includes a consultant end 4000, a switching module 5000, and an integration module 6000.

[0334] The consultant end 4000 includes a consultant registration login module 4100, a professional certification module 4200, a user consultation management module 4300, a consultant annotation module 4400, a consultant personal information management module 4500, and a security certification module 4600.

[0335] The consultant registration login module 4100 is used for the registration and login verification of consultants.

[0336] The professional certification module 4200 is used for the entry of the job level and the certification of the professional information of consultants.

[0337] The user consultation management module 4300 includes an online consultation service module 4310, an offline appointment service module 4320, and a user information search module 4330.

[0338] The online consultation service module 4310 is used for online communication between users and consultants, including but not limited to traditional Chinese medicine health guidance suggestions, health knowledge popularization and course learning, and communication forms including but not limited to text and image communication, voice communication and video communication.

[0339] The offline appointment service module 4320 facilitates users to make offline appointments with consultants and perform health management services, including but not limited to traditional Chinese medicine conditioning and acupuncture and massage.

[0340] The user information search module 4330 facilitates consultants to quickly find out the information of users through user ID, contact information, etc., and view the relevant situation.

[0341] The consultant annotation module 4400 is used for the consultant to perform secondary constitution and zang-fu syndrome differentiation on the tongue image uploaded by the user, and verify the tongue image diagnosis report given by the system.

[0342] The consultant personal information management module 4500 is used for changing and managing the personal basic information and professional information of consultants.

[0343] The security certification module 4600 facilitates the secondary identity authentication of consultants when viewing and managing user information, including one or more combinations of a password authentication unit, a fingerprint authentication unit and a face authentication unit.

[0344] The user end 3000 and the consultant end 4000 are switched by the switching module 5000.

[0345] The user end 3000, the consultant end 4000 and the switching module 5000 are encapsulated in the integrated module 6000.

[0346] The user end 3000 and the consultant end 4000 exist on two different intelligent terminals 1000, respectively.

[0347] The user terminal 3000 and the consultant terminal 4000 are independently operated and respectively communicate with the cloud server 2000.

[0348] The cloud server 2000 includes a center server 2100 and a tongue image algorithm cluster system 2200.

[0349] The center server 2100 includes a data storage module 2110, a tongue image report generation module 2120, a data transmission module 2130, and a database construction module 2140.

[0350] The data storage module 2110 is used for storing data of the intelligent terminal 1000 and the cloud server 2000.

[0351] The tongue image report generation module 2120 is used for generating a comprehensive evaluation report of the tongue image of a user and transmitting the report to the intelligent terminal 1000.

[0352] The data transmission module 2130 is used for data interaction with the intelligent terminal 1000.

[0353] The database construction module 2140 is used for constructing a database including but not limited to basic features of tongue images, physical characteristics, personal information of users, etc.

[0354] The tongue image algorithm cluster system 2200 processes tongue image data of different races, different regions, and different age stages by independent tongue diagnosis algorithms, obtains tongue diagnosis result data, and returns the data to the center server 2100.

[0355] The tongue image algorithm cluster system 2200 includes a data acquisition module 2210, a tongue image preprocessing module 2220, a tongue image feature and physical result labeling module 2230, a tongue image database construction module 2240, an algorithm library module 2250, and a tongue image comprehensive evaluation module 2260.

[0356] The data acquisition module 2210 includes a basic information acquisition module 2211, a tongue image data acquisition module 2212, and a questionnaire acquisition module 2213.

[0357] The basic information acquisition module 2211 is used for acquiring basic information of different races, different regions, and different age stages, including but not limited to skin color, gender, occupation, etc.

[0358] The tongue image data acquisition module 2212 includes a light source module 22121, a camera module 22122, a color card correction module 22123, and a tongue image storage module 22124.

[0359] The light source module 22121 provides an illumination environment required for acquiring tongue images.

[0360] The camera module 22122 provides a tongue image collection device and takes pictures of the tongue image, which includes but is not limited to a tongue diagnosis instrument, a smart phone and other shooting devices.

[0361] The color card correction module 22123 performs color correction on the collected original tongue image to obtain a corrected tongue image.

[0362] The tongue image storage module 22124 stores the corrected tongue image.

[0363] The questionnaire collection module 2213 is used to further determine the constitution and zang-fu syndrome differentiation results of the tongue image collection population in combination with a questionnaire.

[0364] The constitution includes gentle constitution, depressed constitution, qi-deficient constitution, yin-deficient constitution, yang-deficient constitution, phlegm-damp constitution, damp-heat constitution, blood-stasis constitution and special constitution.

[0365] The zang-fu syndrome differentiation is shown in the following table.

[0366] Table 2. Heart and small intestine syndrome and tongue image characteristics

[0367]

[0368] Table 3. Lung and large intestine syndrome and tongue image characteristics

[0369]

[0370] Table 4. Spleen and stomach syndrome and tongue image characteristics

[0371]

[0372] Table 5. Liver and gallbladder syndrome and tongue image characteristics

[0373]

[0374] Table 6. Kidney and bladder syndrome and tongue image characteristics

[0375]

[0376] The tongue image preprocessing module 2220 performs at least one of scaling, cropping, rotating and normalizing on the corrected tongue image to obtain a preprocessed tongue image.

[0377] The tongue image feature labeling module 2230 labels the preprocessed tongue image data according to tongue color, tongue shape, moss color and moss quality by an expert team.

[0378] The tongue color features include pale white tongue, pale red tongue, red tongue, purple tongue, and cyan purple tongue; the moss color features include white moss, light yellow moss, yellow moss, dark yellow moss, gray black moss, and dark black moss; the moss quality features include thin moss, thick moss, greasy moss, rotten moss, peeling moss, no moss, and little moss; the tongue shape features include fat, old, tender, point prick, tooth mark, crack, and blood stasis spot; and the body fluid features include moist, smooth, and dry.

[0379] The tongue image database construction module 2240 classifies the tongue image with the label to obtain a race-region-age tongue image database.

[0380] The algorithm library module 2250 includes a race tongue diagnosis algorithm library module 2251, a region tongue diagnosis algorithm library module 2252, an age tongue diagnosis algorithm library module 2253, and an algorithm library updating module 2254.

[0381] The race tongue diagnosis algorithm library module 2251 includes algorithms R1, R2, and R3, which represent tongue diagnosis algorithms for different races, respectively. The tongue diagnosis algorithm R1 analyzes the tongue image of race A to obtain the return result data T P1 ; the tongue diagnosis algorithm R2 analyzes the tongue image of race B to obtain the return result data T P2 ; and the tongue diagnosis algorithm R3 analyzes the tongue image of race C to obtain the return result data T P3 .

[0382] The region tongue diagnosis algorithm library module 2252 includes algorithms B1, B2, B3, …, and BN, which represent tongue diagnosis algorithms for different regions, respectively. The algorithm B1 analyzes the tongue image of the region to obtain the return result data T Q1 ; the algorithm B2 analyzes the tongue image of the region to obtain the return result data T Q2 ; and the algorithm BN analyzes the tongue image of the region to obtain the return result data T Qn .

[0383] The age tongue diagnosis algorithm library module 2253 includes algorithms C1, C2, C3, and C4, which provide tongue diagnosis algorithms for different age stages, respectively. The algorithm C1 analyzes the tongue image of a user in the youth stage to obtain the return result data T A1 ; the algorithm C2 analyzes the tongue image of a user in the youth stage to obtain the return result data T A2 ; the algorithm C3 analyzes the tongue image of a user in the middle age stage to obtain the return result data T A3 ; and the algorithm C4 analyzes the tongue image of a user in the old age stage to obtain the return result data T A4 .

[0384] The algorithm library updating module 2254 can obtain more accurate diagnosis results by adjusting part of the parameters of the algorithm model or increasing the amount of data to train the tongue diagnosis algorithm model.

[0385] The tongue appearance comprehensive evaluation module 2260 is used to weight and match the diagnostic results of different algorithms selected according to the skin color, residence area and age stage of the user to obtain a tongue appearance comprehensive evaluation result.

[0386] The tongue appearance comprehensive evaluation algorithm used by the tongue appearance comprehensive evaluation module 2250 processes the diagnostic results of different algorithms selected according to the skin color, residence area and age stage to obtain a tongue appearance comprehensive evaluation result TCscore, and the specific formula is as follows:

[0387] TC score = T P · ω1+ T Q · ω2+ T A · ω3

[0388] Wherein, TC score is the tongue appearance comprehensive evaluation result, T P represents the race tongue appearance evaluation result, ω1 is the weight of the skin color, T Q represents the residence area tongue appearance evaluation result, ω2 is the weight of the residence area, T A represents the age stage tongue appearance evaluation result, and ω3 is the weight of the age stage.

[0389] Wherein, the residence area tongue appearance evaluation result T Q is related to the number of residence areas selected by the user and the residence time, and the calculation formula is as follows:

[0390]

[0391] Wherein, T Q is the residence area tongue appearance evaluation result, m represents that the user has selected m area algorithms, T i represents the output result of the area algorithm, ω i represents the weight corresponding to the area algorithm, and the weight is distributed according to the residence area time.

[0392] The intelligent algorithm matcher module 3300 can match algorithms according to the time length of the user's residence area, including a single area algorithm matching module 3310 and a plurality of area algorithm matching module 3320.

[0393] The single area algorithm matching module 3310 selects only one area tongue diagnosis algorithm to obtain a tongue diagnosis result, and this module is suitable for a user who has long-term residence in the area and has not left the residence area during the period.

[0394] The multiple regional algorithm matching module 3320 selects multiple regional tongue diagnosis algorithms to obtain a comprehensive diagnosis result. This module is suitable for users who do not live in a certain region for a long time and live in some regions for a short time. Therefore, the user can select the current regional algorithm that meets the conditions.

[0395] The tongue image algorithm cluster system includes the following steps:

[0396] S1: Collect basic information and tongue image data of different ethnic groups, different regions, and different age groups, and obtain the constitution results of the population by combining questionnaire surveys. The basic information includes but is not limited to ethnicity, age, and permanent residence, etc. The tongue image data is obtained through a tongue image collection device.

[0397] S2: The expert team performs tongue image evaluation according to the basic characteristics of tongue image, such as tongue color, tongue shape, moss color, and moss quality.

[0398] S3: Construct a race-region-age tongue image database: construct a race-region-age stage tongue image database according to the annotated tongue image data. The A, B, and C race labels in the database indicate the race; the 1st, 2nd, 3rd, …, Nth region labels indicate the residence area; and the youth, young, middle-aged, and old labels indicate the age stage.

[0399] S4: Use an artificial intelligence model to perform autonomous learning or deep learning on each tongue image data set in the race-region-age tongue image database to achieve intelligent identification of various features in the tongue image.

[0400] S5: Construct a race-region-age tongue image algorithm library: integrate tongue diagnosis algorithms trained according to different tongue image data sets to obtain a race-region-age tongue image algorithm library, which is represented as Mod = [Mod1, Mod2, …, Mod n ];

[0401] S6: Adjust the parameters of the tongue diagnosis algorithm model or expand the tongue image database to improve the accuracy of the tongue diagnosis algorithm model.

[0402] The database construction module 2140 mainly includes a tongue image information module 2141, a tongue image report module 2142, a health preservation opinion module 2143, a user role table module 2144, and a personal database module 2145.

[0403] The tongue image information module 2141 is used to store information parameters after tongue image analysis, mainly including user ID, tongue image number, tongue color, tongue moss, cracks, and tongue shape category information.

[0404] The tongue image report module 2142 is used to store tongue image analysis report information, mainly including user ID, constitution type, tongue image category information, and tongue image analysis report information.

[0405] The health maintenance opinion module 2143 is used to store health maintenance plans given according to body constitution types, mainly including information such as living, diet, exercise, and sleep.

[0406] The user role table module 2144 is used to store user login information, and the system user is a user role field such as a doctor and a patient, mainly for limiting the permission information of personnel.

[0407] The personal database module 2145 includes a personal information table module 21451 and a personal tongue appearance report library module 21452.

[0408] The personal information table module 21451 is used to store user personal information, mainly including basic information such as user ID, name, gender, age, address, occupation, etc.

[0409] The personal tongue appearance report library module 21452 is used to store the tongue appearance health report of the user, and the tongue appearance condition of the user is displayed and analyzed in the form of radar chart and line chart.

Claims

1. A multi-region fusion intelligent tongue diagnosis system, characterized by: include: Smart terminal (1000) and cloud server (2000); The intelligent terminal (1000) includes a user terminal (3000); The user terminal (3000) includes a personal information management module (3100), a tongue image collection function module (3200), an intelligent algorithm matching module (3300), and a tongue image report management module (3500); The personal information management module (3100) is used to obtain the user's personal information. The tongue image acquisition function module (3200) is used to acquire tongue image; The intelligent algorithm matcher module (3300) matches the corresponding algorithm based on the user's personal information, and uses the matched algorithm to diagnose the tongue image to obtain the tongue diagnosis result; The tongue image report management module (3500) is used to store and record the user tongue image comprehensive evaluation report returned by the cloud server (2000); The cloud server side (2000) includes a central server (2100) and a tongue image algorithm cluster system (2200); The central server (2100) is used to receive and store data from the intelligent terminal (1000) and the tongue image algorithm cluster system (2200); The central server (2100) is used to generate a comprehensive evaluation report of the user's tongue image; The tongue image algorithm cluster system (2200) stores an algorithm library and a tongue image database; The tongue image algorithm cluster system (2200) uses the algorithms in the algorithm library to diagnose tongue images, and performs weight matching on the diagnostic results of different algorithms to obtain a comprehensive tongue image evaluation result.

2. The multi-region fusion intelligent tongue diagnosis system according to claim 1, characterized in that: The intelligent terminal (1000) further includes a consultant terminal (4000); The consultant terminal (4000) includes a registration and login module (4100), a professional certification module (4200), a user consultation management module (4300), a consultant marking module (4400), a consultant personal information management module (4500), and a security authentication module (4600); The consultant performs registration and login operations through the registration and login module (4100); The professional certification module (4200) is used to authenticate the consultant's professional information and enter the job level; The user consultation management module (4300) includes an online consultation service module (4310), an offline reservation service module (4320), and a user information search module (4330); The online consulting service module (4310) provides a platform for consultants to communicate with users online; The content of such online communication shall include but not be limited to TCM health care guidance, health knowledge popularization and course learning; The forms of online communication include but are not limited to text communication, voice communication and video communication; The consultant provides users with offline appointment services and health management services through the offline appointment service module (4320); The health management services include but are not limited to Chinese medicine conditioning, acupuncture and massage; The consultant searches for relevant information of the user through the user information search module (4330); The user's relevant information includes the user's ID and contact information; The consultant uses the consultant annotation module (4400) to perform constitution and organ differentiation on the tongue image uploaded by the user terminal (3000), and verifies it with the user tongue image comprehensive assessment report generated by the cloud server terminal (2000); The constitution differentiation includes balanced constitution, qi stagnation constitution, qi deficiency constitution, yin deficiency constitution, yang deficiency constitution, phlegm-damp constitution, damp-heat constitution, blood stasis constitution, and special constitution; The viscera differentiation includes the differentiation of heart and small intestine, the differentiation of lung and large intestine, the differentiation of spleen and stomach, the differentiation of liver and gallbladder, and the differentiation of kidney and bladder; The consultant personal information management module (4500) is used to manage and modify the consultant's basic personal information and professional information; The consultant's basic personal information includes the consultant's name, gender, and age; The consultant performs secondary identity authentication through the security authentication module (4600) when viewing and managing user information; The secondary identity authentication method includes one or more combinations of password authentication, fingerprint authentication and face authentication.

3. The multi-region fusion intelligent tongue diagnosis system according to claim 2, characterized in that: The intelligent terminal (1000) further includes a switching module (5000); The user terminal (3000) and the consultant terminal (4000) are switched via a switching module (5000).

4. The multi-region fusion intelligent tongue diagnosis system according to claim 3, characterized in that: The intelligent terminal (1000) further includes an integrated module (6000); The user terminal (3000), the advisor terminal (4000) and the switching module (5000) are encapsulated in an integrated module (6000).

5. The multi-region fusion intelligent tongue diagnosis system according to claim 1, characterized in that: The user terminal (3000) also includes a questionnaire filling module (3400) and a professional guidance and suggestion module (3600); The questionnaire filling module (3400) obtains the user's living conditions through a questionnaire; The user's living conditions include dietary preferences, sleep conditions, mood conditions, exercise conditions, and medication conditions; The professional guidance and suggestion module (3600) is used to display the professional guidance and suggestions provided by the consultant based on the user's tongue image comprehensive assessment report; The professional guidance and suggestions include but are not limited to emotional regulation, diet regulation, daily life regulation, exercise health care, and acupoint health care.

6. The multi-region fusion intelligent tongue diagnosis system according to claim 1, characterized in that: The personal information management module (3100) includes a user information entry module (3110), an automatic positioning module (3120), and a user information history recording module (3130); The user information input module (3110) includes a race information input module (3111), a residence information input module (3112), and a basic information input module (3113); The ethnicity information entry module (3111) is used to obtain the ethnicity and skin color information of the user; The residence information input module (3112) includes a current residence information input module (31121), a historical residence information input module (31122), and a residence duration information input module (31123); The current residence information input module (31121) is used to obtain the location information of the user's current residence; The historical residence information input module (31122) is used to obtain the location information of the user's historical residence; The residence time information input module (31123) is used to obtain the residence time of the user at the current residence and the previous residence; The basic information entry module (3113) is used to obtain the basic information of the user; The basic information includes but is not limited to ID, name, gender, age, address, and occupation; The automatic positioning module (3120) is used to locate the user's location and record the user's permanent residence; The user information history record module (3130) is used to store and manage the user's personal information; The user's personal information includes the user's race and skin color, current residence location information, historical residence location information, length of residence, basic information, location and permanent residence information; The tongue image collection function module (3200) includes a shooting module (3210), a tongue image uploading module (3220), and a tongue image history recording module (3230); The shooting module (3210) is used to shoot the user's tongue image; The tongue image includes a tongue surface image and a sublingual vein image; The tongue image uploading module (3220) is used to transmit the user's tongue image to the cloud server (2000); The tongue image history recording module (3230) is used to store and record the tongue image of the user; The intelligent algorithm matching module (3300) includes a single region algorithm matching module (3310) and a plurality of region algorithm matching modules (3320); The intelligent algorithm matching module (3300) performs algorithm matching based on the length of time the user has lived in the current residence and the historical residence; When the user lives in a certain area for a long time, the single area algorithm matching module (3310) selects the algorithm of the corresponding area to diagnose the tongue image and obtain the tongue diagnosis result; When the user has multiple living areas, the multiple area algorithm matching module (3320) selects the corresponding multiple area algorithms to perform comprehensive diagnosis on the tongue image to obtain the tongue diagnosis result.

7. The multi-region fusion intelligent tongue diagnosis system according to claim 1, characterized in that: The central server (2100) includes a data storage module (2110), a tongue image report generation module (2120), a data transmission module (2130), and a database construction module (2140); The data storage module (2110) is used to store data of the intelligent terminal (1000) and the tongue image algorithm cluster system (2200); The tongue image report generating module (2120) is used to generate a comprehensive evaluation report of the user's tongue image; The data transmission module (2130) is used for data interaction between the intelligent terminal (1000) and the cloud server (2000); The database construction module (2140) includes a tongue image information module (2141), a tongue image report module (2142), a health advice module (2143), a user role table module (2144), and a personal database module (2145); The tongue image information module (2141) is used to store the user's tongue diagnosis results; The tongue diagnosis results include tongue color, tongue coating, cracks, and tongue shape; The tongue image reporting module (2142) analyzes the tongue diagnosis results of the user, generates and stores the tongue image report of each user; The tongue image report includes the user's constitution type and tongue image category; The health advice module (2143) is used to store the health plan given based on the tongue diagnosis results; The health regimen includes living regimen, diet regimen, exercise regimen and sleep regimen; The user role table module (2144) is used to store the user's login information; The personal database module (2145) includes a personal information table module (21451) and a personal tongue image report library module (21452); The personal information table module (21451) is used to store the user's basic information; The basic information includes but is not limited to ID, name, gender, age, address, and occupation; The personal tongue image report library module (21452) is used to store the user's tongue image comprehensive evaluation report and display and analyze it in the form of radar charts and line charts; The user tongue image comprehensive evaluation report includes user basic information, user tongue image, user tongue diagnosis results, and health care plan.

8. The multi-region fusion intelligent tongue diagnosis system according to claim 1, characterized in that: The tongue image algorithm cluster system (2200) includes a data acquisition module (2210), a tongue image preprocessing module (2220), a tongue image feature annotation module (2230), a tongue image database construction module (2240), an algorithm library module (2250), and a tongue image comprehensive evaluation module (2260); The data collection module (2210) includes a basic information collection module (2211), a tongue image data collection module (2212), and a questionnaire collection module (2213); The basic information collection module (2211) is used to collect basic information of people of different races, skin colors, different residential areas and different age groups, including but not limited to skin color, gender and occupation; The tongue image data acquisition module (2212) includes a light source module (22121), a camera module (22122), a colorimetric card calibration module (22123), and a tongue image storage module (22124); The light source module (22121) is used to provide a lighting environment; The camera module (22122) provides a tongue image acquisition device for capturing tongue images; The tongue image collection device includes but is not limited to a tongue diagnosis instrument and a smart phone; The colorimetric card correction module (22123) is used to perform color correction on the tongue image captured by the tongue image acquisition device to obtain a corrected tongue image; The tongue image storage module (22124) is used to store the corrected tongue image; The questionnaire collection module (2213) conducts physical constitution and organ differentiation on people of different races, skin colors, different residential areas and different age groups through questionnaires to obtain corresponding TCM physical constitution results; The constitution differentiation includes balanced constitution, qi stagnation constitution, qi deficiency constitution, yin deficiency constitution, yang deficiency constitution, phlegm-damp constitution, damp-heat constitution, blood stasis constitution, and special constitution; The viscera differentiation includes the differentiation of heart and small intestine, the differentiation of lung and large intestine, the differentiation of spleen and stomach, the differentiation of liver and gallbladder, and the differentiation of kidney and bladder; The tongue image pre-processing module (2220) is used to pre-process the corrected tongue image; The preprocessing includes one or more of scaling, cropping, rotation, and normalization; The tongue image feature annotation module (2230) is used to annotate the tongue image features after pre-processing; The tongue features include tongue color, tongue shape, tongue coating color and tongue coating quality; The tongue image database construction module (2240) classifies the annotated tongue image and constructs a race-region-age tongue image database; The algorithm library module (2250) includes a human race tongue diagnosis algorithm library module (2251), a regional tongue diagnosis algorithm library module (2252), an age tongue diagnosis algorithm library module (2253), and an algorithm library update module (2254); The ethnic tongue diagnosis algorithm library module (2251) stores tongue diagnosis algorithms for different ethnic groups and skin colors; The regional tongue diagnosis algorithm library module (2252) stores tongue diagnosis algorithms for different residential areas; The age tongue diagnosis algorithm library module (2253) stores tongue diagnosis algorithms for different age stages; The algorithm library updating module (2254) updates the algorithm library by adjusting some parameters of the tongue diagnosis algorithm or increasing the data volume; The tongue image comprehensive evaluation module (2260) selects the corresponding algorithm according to race, skin color, residential area and age stage, and performs weight matching on the diagnostic results of different algorithms to obtain the tongue image comprehensive evaluation result.

9. The multi-region fusion intelligent tongue diagnosis system according to claim 8, characterized in that: The comprehensive evaluation results of the tongue are as follows: TC score =T P ·ω1+T Q ·ω2+T A ·ω3 (1) Where TC score is the comprehensive evaluation result of tongue appearance; T P represents the evaluation result of tongue image algorithm based on race and skin color, ω1 is the weight of race and skin color, ω2 is the weight of residential area, T A represents the evaluation result of tongue image algorithm by age stage, ω3 is the weight of age stage; Among them, the evaluation result of tongue image algorithm in residential area T Q As shown below: Where i represents the residential area index, m represents the total number of residential areas, Ti represents the output of the algorithm for the i-th residential area, and γ i Represents the weight of the algorithm for the i-th residential area.

10. The multi-region fusion intelligent tongue diagnosis system according to claim 9, characterized in that: The steps for constructing the algorithm library in the tongue image algorithm cluster system (2200) are as follows: S1 collects basic information and corresponding tongue images of people of different races, skin colors, different residential areas and different age groups; S2 labels the tongue image with tongue features, race, skin color, residential area, and age stage; s3 uses race and skin color as input and tongue image features as output to construct training set I; The training set II is constructed with the residential area as input and tongue image features as output; The training set III is constructed with age stage as input and tongue image features as output; s4 builds tongue diagnosis algorithm models for race, region, and age; s5: training the ethnic tongue diagnosis algorithm model with the training set I to obtain a trained ethnic tongue diagnosis algorithm model; The regional tongue diagnosis algorithm model is trained with the training set II to obtain a trained regional tongue diagnosis algorithm model; The age tongue diagnosis algorithm model is trained with the training set III to obtain a trained age tongue diagnosis algorithm model; S6 builds an ethnic tongue diagnosis algorithm library based on the trained ethnic tongue diagnosis algorithm model; Build a regional tongue diagnosis algorithm library based on the trained ethnic regional tongue diagnosis algorithm model; Based on the trained ethnicity and age tongue diagnosis algorithm model, an age tongue diagnosis algorithm library is constructed.

Images