An image intelligent processing system and method for tongue surface diagnosis

Abstract

The present invention relates to the field of image processing technology, and specifically to an image intelligent processing system and method for tongue diagnosis, in which a label object analysis module in the system obtains the binding relationship between face labels and tongue labels in historical label data, and based on the analysis results in the face image feature adaptation analysis module and the tongue image feature adaptation analysis module, locks the historical reference label object set corresponding to the user's tongue feature information. The present invention not only takes into account the collaborative analysis between the user's face image information and tongue image information, but also takes into account the device status information when collecting face image information and tongue image information. By analyzing the difference in device collection status, the face image feature adaptation adjustment coefficient and the tongue image feature adaptation adjustment coefficient are dynamically acquired, and then the historical reference label object set corresponding to the user's tongue feature information is accurately locked, providing data support for the subsequent generation of a diagnostic auxiliary reference information recommendation sequence corresponding to the user's tongue feature information.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to an intelligent image processing system and method for tongue surface diagnosis. Background Art

[0002] In the field of medical diagnosis, sublingual observation is an important auxiliary means that can provide useful information about the patient's health status. Traditionally, doctors diagnose diseases by observing the color, shape, texture and other characteristics of the sublingual area. The judgment process requires doctors to spend more time and energy. Therefore, using tongue and facial diagnostic instruments to assist in obtaining the user's condition has become the choice of many doctors.

[0003] However, during the use of the tongue and facial diagnosis instrument, the components of the tongue and facial diagnosis instrument will change with the change of usage time and the length of time interval between instrument maintenance, and the clarity and accuracy of the collected images will deviate, thereby affecting the analysis results of the collected images in subsequent steps and the doctor's diagnosis of the patient's condition. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent image processing system and method for tongue surface diagnosis to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: an image intelligent processing method for tongue surface diagnosis, the method comprising the following steps:

[0006] S1. Collecting the user's facial image and tongue image through a tongue diagnosis instrument, and extracting features from the collected images based on image recognition technology to generate user tongue feature information, wherein the user tongue feature information includes multi-dimensional facial feature information, multi-dimensional tongue feature information, and device collection status information;

[0007] S2, comparing the multi-dimensional facial feature information in the user's tongue and face feature information with the historical label data preset in the database, generating the facial feature adaptation degree of the user based on the corresponding historical label data and the facial feature adaptation adjustment coefficient based on the corresponding historical label data;

[0008] S3, comparing the tongue image feature multi-dimensional information in the user's tongue surface feature information with the historical label data preset in the database, generating the user's tongue image feature adaptation degree based on the corresponding historical label data and the tongue image feature adaptation adjustment coefficient based on the corresponding historical label data;

[0009] S4, obtaining the binding relationship between the face label and the tongue label in the historical label data, and based on the analysis results in step S2 and step S3, locking the historical reference label object set corresponding to the user's tongue and face feature information;

[0010] S5. Generate a recommended sequence of diagnostic auxiliary reference information corresponding to the user's tongue surface feature information based on the historical reference label object set corresponding to the user's tongue surface feature information and the disease diagnosis information corresponding to each label data in the historical data.

[0011] Furthermore, the multidimensional facial feature information includes the color level of each viscera partition of the face, the color level of the viscera partition is equal to the color level corresponding to the average pixel value of the corresponding pixel point in the facial image area where the corresponding viscera partition is located, and the color level corresponding to the average pixel value of the pixel point is a constant preset in the database; the number of viscera partitions included in the multidimensional facial feature information is a preset constant;

[0012] The multi-dimensional information of tongue image features includes the color grade of the tongue, the tongue coating area, the tongue coating thickness, and the position distribution of tongue cracks; the color grade of the tongue is equal to the color grade corresponding to the average value of the corresponding pixel points in the tongue image area where the corresponding tongue is located;

[0013] The device collects status information including the usage time of the tongue and facial diagnosis instrument and the interval between the last maintenance time of the tongue and facial diagnosis instrument and the current time.

[0014] When obtaining the user's tongue and face feature information in the present invention, not only the multi-dimensional information of facial features and the multi-dimensional information of tongue features are considered, but also the equipment collection status information is considered. This is because the degree and risk of deviation in the collection screen of the tongue and face diagnosis instrument vary depending on the usage status of the instrument. Therefore, taking into account the equipment collection status information, the multi-dimensional information of facial features (multi-dimensional information of tongue features) can be analyzed in subsequent steps to provide certain data support based on the facial feature adaptation adjustment coefficient of the face label in the historical label data (the tongue feature adaptation adjustment coefficient of the tongue label).

[0015] Furthermore, the calculation formula for generating the user's facial feature adaptability based on the corresponding historical tag data in S2 is as follows:

[0016] ,

[0017] Among them, MS i represents the facial feature adaptation degree of the user based on the i-th face label in the corresponding historical label data; the face label represents the multi-dimensional facial feature information of the user's tongue feature information in the historical data; C （i，j） represents the absolute value of the difference between the jth parameter in the i-th face label in the corresponding historical label data and the jth parameter in the multidimensional information of the user's facial image feature; j1 represents the number of parameters in the i-th face label in the corresponding historical label data; MG represents the tolerance for the deviation fluctuation of the internal organs color level preset in the database; DBM{} represents the facial image feature adaptation comparison function, when C （i，j） ≥MG, then judge Otherwise, it is judged ;

[0018] The calculation formula for generating the face feature adaptation adjustment coefficient based on the corresponding historical label data is as follows:

[0019] ,

[0020] Among them, PM i represents the facial feature adaptation adjustment coefficient of the i-th face label in the corresponding historical label data; TW represents the usage time of the tongue-face diagnosis instrument in the device collection status information; TW i Indicates the usage time of the tongue-face diagnostic instrument corresponding to the i-th face label in the corresponding historical label data; Indicates that the usage time of the same tag object model interval in the historical data is |TW-TW i | respectively corresponds to the average value of the color level difference of the facial viscera partition in the multi-dimensional information of the monitored facial features; TG represents the interval between the last maintenance time of the tongue and face diagnostic instrument and the current time in the equipment collection status information; Indicates that the interval between the last maintenance time of the tongue and facial diagnosis instrument for the same labeled object model in the historical data is |TG-TG i |is the average value of the difference in color levels of the corresponding facial viscera partitions in the multi-dimensional information of facial features monitored respectively; the same labeled object model represents a facial model preset in the database and the color of each facial viscera partition in the preset facial model is fixed; max{} represents the operation of finding the maximum value.

[0021] The present invention not only takes into account the differences in the adaptability of users' facial features based on corresponding historical label data, but also takes into account the impact of differences in device acquisition status information on the image feature adaptation adjustment coefficient. When acquiring the facial feature adaptation adjustment coefficient, the usage time of the corresponding tongue and facial diagnostic instrument when acquiring the image and the interval between the last maintenance time of the tongue and facial diagnostic instrument and the current time are taken into account. These two factors can reflect the impact of differences in device acquisition status on the deviation of image acquisition results, thereby providing a basis for locking the historical reference label object set corresponding to the user's tongue and facial feature information in subsequent steps.

[0022] Furthermore, the calculation formula for generating the tongue image feature adaptability of the user based on the corresponding historical label data in S3 is as follows:

[0023] ,

[0024] Among them, SS irepresents the adaptability of the user's tongue image feature based on the i-th tongue label in the corresponding historical label data; the tongue label represents the multi-dimensional information of the tongue image feature of the user's tongue surface feature information in the historical data; CS i represents the absolute value of the difference between the color level of the tongue in the multidimensional information of the tongue image feature and the color level of the tongue corresponding to the i-th tongue tag in the corresponding historical tag data; MG1 represents the tolerance of the tongue color level deviation fluctuation preset in the database; DBS{} represents the tongue image feature adaptation comparison function, when CS i ≥MG1, then judge Otherwise, it is judged SR i Represents the intersection area of ​​the tongue coating area in the multidimensional information of tongue image features and the tongue coating area corresponding to the i-th tongue label in the corresponding historical label data; SRB i represents the union area of ​​the tongue coating area in the multidimensional information of tongue image features and the tongue coating area corresponding to the i-th tongue label in the corresponding historical label data; SDi represents the absolute value of the difference between the tongue coating thickness in the multidimensional information of tongue image features and the tongue coating thickness corresponding to the i-th tongue label in the corresponding historical label data; SHi max Represents the maximum value between the tongue coating thickness in the multidimensional information of tongue image features and the tongue coating thickness corresponding to the i-th tongue label in the corresponding historical label data;

[0025] SL i It indicates the ratio of the maximum overlap area between the position distribution of the tongue crack corresponding to the corresponding tongue label and the position distribution of the tongue crack in the multidimensional information of tongue image features to the area of ​​the tongue crack in the multidimensional information of tongue image features after adjusting the angle of the image belonging to the i-th tongue label in the corresponding historical label data; if there is no tongue crack in the multidimensional information of tongue image features, then SL i =0;

[0026] The calculation formula for generating the tongue image feature adaptation adjustment coefficient based on the corresponding historical label data is as follows:

[0027] ,

[0028] Among them, PS i represents the tongue image feature adaptation adjustment coefficient of the i-th face label in the corresponding historical label data; μ represents the preset tongue face image feature conversion factor.

[0029] Furthermore, the S4 includes:

[0030] S41, obtaining the binding relationship between the face label and the tongue label in the historical label data, and binding the face label and the tongue label that have the same collection time and the corresponding collection device are the same tongue and face diagnosis instrument;

[0031] S42, obtaining the user's face feature adaptation degree MS based on the i-th face label in the corresponding historical label data i , and the user's tongue image feature adaptation degree SS based on the i-th tongue label in the corresponding historical label data i ;

[0032] S43, obtaining a set of all face labels whose face image feature adaptation degree between the face image feature multi-dimensional information in the user's tongue face feature information is greater than or equal to the face image adaptation threshold and whose tongue image feature adaptation degree between the bound tongue label and the tongue image feature multi-dimensional information in the user's tongue face feature information is greater than or equal to the tongue image adaptation threshold as a first reference label object set; the face image adaptation threshold and the tongue image adaptation threshold are both constants preset in the database;

[0033] S44, obtaining a facial feature adaptation adjustment coefficient based on the corresponding historical label data, and a tongue feature adaptation adjustment coefficient based on the corresponding historical label data; and recording a set of all face labels whose facial feature adaptation degree between the facial feature multidimensional information in the user's tongue face feature information is greater than or equal to the facial feature adaptation threshold based on the mapping value of the corresponding face label, and whose tongue feature adaptation degree between the bound tongue label and the tongue feature multidimensional information in the user's tongue face feature information is greater than or equal to the tongue image adaptation threshold based on the mapping value of the corresponding tongue label as a second reference label object set;

[0034] The reference face labels are different, and the corresponding face image adaptation thresholds are different based on the mapping values ​​of the corresponding face labels; the mapping values ​​of the face image adaptation thresholds based on the corresponding face labels are equal to the quotient obtained by dividing the corresponding face image adaptation threshold by the sum of the face image feature adaptation adjustment coefficient of the corresponding face label and 1; the reference tongue labels are different, and the corresponding tongue image adaptation thresholds are different based on the mapping values ​​of the corresponding tongue labels; the mapping values ​​of the tongue image adaptation thresholds based on the corresponding tongue labels are equal to the quotient obtained by dividing the corresponding tongue image adaptation threshold by the sum of the tongue image feature adaptation adjustment coefficient of the corresponding tongue label and 1;

[0035] S45. Record the union of the first reference label object set and the second reference label object set as the historical reference label object set corresponding to the user's tongue feature information, and mark the elements in the obtained historical reference label object set that belong to the first reference label object set.

[0036] Furthermore, the S5 includes:

[0037] Obtain the disease diagnosis information corresponding to the face label corresponding to the kth element in the historical reference label object set corresponding to the user's tongue surface feature information and the tongue label bound to it in the historical data; summarize the obtained disease diagnosis information, and sort the disease diagnosis information in descending order according to the risk assessment value corresponding to each disease diagnosis information, and generate a recommended sequence of diagnosis auxiliary reference information corresponding to the user's tongue surface feature information; mark the corresponding sequence elements in the recommended sequence of diagnosis auxiliary reference information corresponding to the user's tongue surface feature information for the disease diagnosis information corresponding to the marked element in the historical reference label object set corresponding to the user's tongue surface feature information;

[0038] In the process of obtaining the risk assessment value corresponding to the disease diagnosis information, the risk assessment value corresponding to each disease diagnosis information is equal to the maximum value of the facial image feature fitness corresponding to each face label corresponding to the corresponding disease diagnosis information in the historical reference label object set corresponding to the user's tongue surface feature information and the average value of the facial image feature fitness corresponding to the tongue label bound to it; each disease diagnosis information has one or more face labels corresponding to it in the historical reference label object set corresponding to the user's tongue surface feature information.

[0039] An image intelligent processing system for tongue surface diagnosis, the system comprising the following modules:

[0040] A tongue surface feature acquisition module, which acquires the user's facial image and tongue image through a tongue surface diagnostic instrument, and extracts features from the acquired images based on image recognition technology to generate user tongue surface feature information, wherein the user tongue surface feature information includes multi-dimensional facial feature information, multi-dimensional tongue feature information, and device acquisition status information;

[0041] A facial feature adaptation analysis module, which compares the facial feature multi-dimensional information in the user's tongue and face feature information with the historical label data preset in the database, and generates the facial feature adaptation degree of the user based on the corresponding historical label data and the facial feature adaptation adjustment coefficient based on the corresponding historical label data;

[0042] A tongue image feature adaptation analysis module, which compares the tongue image feature multi-dimensional information in the user's tongue surface feature information with the historical label data preset in the database, and generates the user's tongue image feature adaptation degree based on the corresponding historical label data and the tongue image feature adaptation adjustment coefficient based on the corresponding historical label data;

[0043] A label object analysis module, which obtains the binding relationship between the face label and the tongue label in the historical label data, and locks the historical reference label object set corresponding to the user's tongue and face feature information based on the analysis results in the face feature adaptation analysis module and the tongue feature adaptation analysis module;

[0044] An image auxiliary information management module generates a recommended sequence of diagnostic auxiliary reference information corresponding to the user's tongue surface feature information based on a historical reference label object set corresponding to the user's tongue surface feature information and disease diagnosis information corresponding to each label data in the historical data.

[0045] Furthermore, the facial feature adaptation analysis module includes a facial feature adaptation degree analysis unit and a facial feature adaptation adjustment coefficient analysis unit.

[0046] The facial feature adaptability analysis unit compares the facial feature multi-dimensional information in the user's tongue feature information with the historical label data preset in the database to generate the facial feature adaptability of the user based on the corresponding historical label data;

[0047] The facial feature adaptation adjustment coefficient analysis unit compares the facial feature multi-dimensional information in the user's tongue feature information with the historical label data preset in the database to generate the facial feature adaptation adjustment coefficient of the user based on the corresponding historical label data.

[0048] Compared with the prior art, the beneficial effect achieved by the present invention is that in the process of analyzing and processing the images collected by the tongue and face diagnosis instrument, the present invention not only takes into account the collaborative analysis between the user's facial image information and the tongue image information (the facial image information and the tongue image information at the same time are jointly used as the basis for determining the user's disease diagnosis information), but also takes into account the device status information when collecting the facial image information and the tongue image information. By analyzing the differences in the device collection status, the dynamic acquisition of the facial feature adaptation adjustment coefficient and the tongue feature adaptation adjustment coefficient is achieved, and then the historical reference label object set corresponding to the user's tongue and face feature information is accurately locked, which provides data support for the subsequent generation of the diagnostic auxiliary reference information recommendation sequence corresponding to the user's tongue and face feature information, thereby achieving effective processing of the images collected by the tongue and face diagnosis instrument. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0050] Figure 1 It is a structural schematic diagram of an image intelligent processing system for tongue surface diagnosis according to the present invention;

[0051] Figure 2 It is a flow chart of an intelligent image processing method for tongue surface diagnosis according to the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] See also Figure 1 The present invention provides a technical solution: an image intelligent processing system for tongue surface diagnosis, the system comprising the following modules:

[0054] A tongue surface feature acquisition module, which acquires the user's facial image and tongue image through a tongue surface diagnostic instrument, and extracts features from the acquired images based on image recognition technology to generate user tongue surface feature information, wherein the user tongue surface feature information includes multi-dimensional facial feature information, multi-dimensional tongue feature information, and device acquisition status information;

[0055] A facial feature adaptation analysis module, the facial feature adaptation analysis module includes a facial feature adaptation degree analysis unit and a facial feature adaptation adjustment coefficient analysis unit,

[0056] The facial feature adaptability analysis unit compares the facial feature multi-dimensional information in the user's tongue feature information with the historical label data preset in the database to generate the facial feature adaptability of the user based on the corresponding historical label data;

[0057] The facial feature adaptation adjustment coefficient analysis unit compares the facial feature multi-dimensional information in the user's tongue feature information with the historical label data preset in the database to generate the facial feature adaptation adjustment coefficient of the user based on the corresponding historical label data;

[0058] A tongue image feature adaptation analysis module, which compares the tongue image feature multi-dimensional information in the user's tongue surface feature information with the historical label data preset in the database, and generates the user's tongue image feature adaptation degree based on the corresponding historical label data and the tongue image feature adaptation adjustment coefficient based on the corresponding historical label data;

[0059] A label object analysis module, which obtains the binding relationship between the face label and the tongue label in the historical label data, and locks the historical reference label object set corresponding to the user's tongue and face feature information based on the analysis results in the face feature adaptation analysis module and the tongue feature adaptation analysis module;

[0060] An image auxiliary information management module generates a recommended sequence of diagnostic auxiliary reference information corresponding to the user's tongue surface feature information based on a historical reference label object set corresponding to the user's tongue surface feature information and disease diagnosis information corresponding to each label data in the historical data.

[0061] like Figure 2 As shown, an image intelligent processing method for tongue surface diagnosis comprises the following steps:

[0062] S1. Collecting the user's facial image and tongue image through a tongue diagnosis instrument, and extracting features from the collected images based on image recognition technology to generate user tongue feature information, wherein the user tongue feature information includes multi-dimensional facial feature information, multi-dimensional tongue feature information, and device collection status information;

[0063] The multi-dimensional facial feature information includes the color level of each viscera partition of the face, the color level of the viscera partition is equal to the color level corresponding to the average pixel value of the corresponding pixel point in the facial image area where the corresponding viscera partition is located, and the color level corresponding to the average pixel value of the pixel point is a constant preset in the database; the number of viscera partitions included in the multi-dimensional facial feature information is a preset constant;

[0064] The multi-dimensional information of tongue image features includes the color grade of the tongue, the tongue coating area, the tongue coating thickness, and the position distribution of tongue cracks; the color grade of the tongue is equal to the color grade corresponding to the average value of the corresponding pixel points in the tongue image area where the corresponding tongue is located;

[0065] The device collects status information including the usage time of the tongue and facial diagnosis instrument and the interval between the last maintenance time of the tongue and facial diagnosis instrument and the current time.

[0066] In this embodiment, both the color level of the internal organs and the color level of the tongue are distinguished by pixel values, the pixel values ​​are grayed, and the pixel gray value interval (0-255) is equally divided according to the preset number of color levels, and each equally divided pixel gray value interval is bound to a color level.

[0067] S2, comparing the multi-dimensional facial feature information in the user's tongue and face feature information with the historical label data preset in the database, generating the facial feature adaptation degree of the user based on the corresponding historical label data and the facial feature adaptation adjustment coefficient based on the corresponding historical label data;

[0068] The calculation formula for generating the user's facial feature adaptability based on the corresponding historical label data in S2 is as follows:

[0069] ,

[0070] Among them, MS i represents the facial feature adaptation degree of the user based on the i-th face label in the corresponding historical label data; the face label represents the multi-dimensional facial feature information of the user's tongue feature information in the historical data; C （i，j） represents the absolute value of the difference between the jth parameter in the i-th face label in the corresponding historical label data and the jth parameter in the multidimensional information of the user's facial image feature; j1 represents the number of parameters in the i-th face label in the corresponding historical label data; MG represents the tolerance for the deviation fluctuation of the internal organs color level preset in the database; DBM{} represents the facial image feature adaptation comparison function, when C （i，j） ≥MG, then judge Otherwise, it is judged ;

[0071] The calculation formula for generating the face feature adaptation adjustment coefficient based on the corresponding historical label data is as follows:

[0072]

[0073] Among them, PM i represents the facial feature adaptation adjustment coefficient of the i-th face label in the corresponding historical label data; TW represents the usage time of the tongue-face diagnosis instrument in the device collection status information; TW i Indicates the usage time of the tongue-face diagnostic instrument corresponding to the i-th face label in the corresponding historical label data; Indicates that the usage time of the same tag object model interval in the historical data is |TW-TW i | respectively corresponds to the average value of the color level difference of the facial viscera partition in the multi-dimensional information of the monitored facial features; TG represents the interval between the last maintenance time of the tongue and face diagnostic instrument and the current time in the equipment collection status information; Indicates that the interval between the last maintenance time of the tongue and facial diagnosis instrument for the same labeled object model in the historical data is |TG-TG i |is the average value of the difference in color levels of the corresponding facial viscera partitions in the multi-dimensional information of facial features monitored respectively; the same labeled object model represents a facial model preset in the database and the color of each facial viscera partition in the preset facial model is fixed; max{} represents the operation of finding the maximum value.

[0074] S3, comparing the tongue image feature multi-dimensional information in the user's tongue surface feature information with the historical label data preset in the database, generating the user's tongue image feature adaptation degree based on the corresponding historical label data and the tongue image feature adaptation adjustment coefficient based on the corresponding historical label data;

[0075] The calculation formula for generating the tongue image feature adaptability of the user based on the corresponding historical label data in S3 is as follows:

[0076]

[0077] Among them, SS i represents the adaptability of the user's tongue image feature based on the i-th tongue label in the corresponding historical label data; the tongue label represents the multi-dimensional information of the tongue image feature of the user's tongue surface feature information in the historical data; CS i represents the absolute value of the difference between the color level of the tongue in the multidimensional information of the tongue image feature and the color level of the tongue corresponding to the i-th tongue tag in the corresponding historical tag data; MG1 represents the tolerance of the tongue color level deviation fluctuation preset in the database; DBS{} represents the tongue image feature adaptation comparison function, when CS i ≥MG1, then judge Otherwise, it is judged SR i Represents the intersection area of ​​the tongue coating area in the multidimensional information of tongue image features and the tongue coating area corresponding to the i-th tongue label in the corresponding historical label data; SRB i represents the union area of ​​the tongue coating area in the multidimensional information of tongue image features and the tongue coating area corresponding to the i-th tongue label in the corresponding historical label data; SDi represents the absolute value of the difference between the tongue coating thickness in the multidimensional information of tongue image features and the tongue coating thickness corresponding to the i-th tongue label in the corresponding historical label data; SHi max Represents the maximum value between the tongue coating thickness in the multidimensional information of tongue image features and the tongue coating thickness corresponding to the i-th tongue label in the corresponding historical label data;

[0078] SL i It indicates the ratio of the maximum overlap area between the position distribution of the tongue crack corresponding to the corresponding tongue label and the position distribution of the tongue crack in the multidimensional information of tongue image features to the area of ​​the tongue crack in the multidimensional information of tongue image features after adjusting the angle of the image belonging to the i-th tongue label in the corresponding historical label data; if there is no tongue crack in the multidimensional information of tongue image features, then SL i =0;

[0079] The calculation formula for generating the tongue image feature adaptation adjustment coefficient based on the corresponding historical label data is as follows:

[0080] ,

[0081] Among them, PS i represents the tongue image feature adaptation adjustment coefficient of the i-th face label in the corresponding historical label data; μ represents the preset tongue face image feature conversion factor.

[0082] S4, obtaining the binding relationship between the face label and the tongue label in the historical label data, and based on the analysis results in step S2 and step S3, locking the historical reference label object set corresponding to the user's tongue and face feature information;

[0083] The S4 includes:

[0084] S41, obtaining the binding relationship between the face label and the tongue label in the historical label data, and binding the face label and the tongue label that have the same collection time and the corresponding collection device are the same tongue and face diagnosis instrument;

[0085] S42, obtaining the user's face feature adaptation degree MS based on the i-th face label in the corresponding historical label data i , and the user's tongue image feature adaptation degree SS based on the i-th tongue label in the corresponding historical label data i ;

[0086] S43, obtaining a set of all face labels whose face image feature adaptation degree between the face image feature multi-dimensional information in the user's tongue face feature information is greater than or equal to the face image adaptation threshold and whose tongue image feature adaptation degree between the bound tongue label and the tongue image feature multi-dimensional information in the user's tongue face feature information is greater than or equal to the tongue image adaptation threshold as a first reference label object set; the face image adaptation threshold and the tongue image adaptation threshold are both constants preset in the database;

[0087] S44, obtaining a facial feature adaptation adjustment coefficient based on the corresponding historical label data, and a tongue feature adaptation adjustment coefficient based on the corresponding historical label data; and recording a set of all face labels whose facial feature adaptation degree between the facial feature multidimensional information in the user's tongue face feature information is greater than or equal to the facial feature adaptation threshold based on the mapping value of the corresponding face label, and whose tongue feature adaptation degree between the bound tongue label and the tongue feature multidimensional information in the user's tongue face feature information is greater than or equal to the tongue image adaptation threshold based on the mapping value of the corresponding tongue label as a second reference label object set;

[0088] The reference face labels are different, and the corresponding face image adaptation thresholds are different based on the mapping values ​​of the corresponding face labels; the mapping values ​​of the face image adaptation thresholds based on the corresponding face labels are equal to the quotient obtained by dividing the corresponding face image adaptation threshold by the sum of the face image feature adaptation adjustment coefficient of the corresponding face label and 1; the reference tongue labels are different, and the corresponding tongue image adaptation thresholds are different based on the mapping values ​​of the corresponding tongue labels; the mapping values ​​of the tongue image adaptation thresholds based on the corresponding tongue labels are equal to the quotient obtained by dividing the corresponding tongue image adaptation threshold by the sum of the tongue image feature adaptation adjustment coefficient of the corresponding tongue label and 1;

[0089] S45. Record the union of the first reference label object set and the second reference label object set as the historical reference label object set corresponding to the user's tongue feature information, and mark the elements in the obtained historical reference label object set that belong to the first reference label object set.

[0090] S5, generating a recommended sequence of diagnostic auxiliary reference information corresponding to the user's tongue surface feature information based on the historical reference label object set corresponding to the user's tongue surface feature information and the disease diagnosis information corresponding to each label data in the historical data;

[0091] The S5 includes:

[0092] Obtain the disease diagnosis information corresponding to the face label corresponding to the kth element in the historical reference label object set corresponding to the user's tongue surface feature information and the tongue label bound to it in the historical data; summarize the obtained disease diagnosis information, and sort the disease diagnosis information in descending order according to the risk assessment value corresponding to each disease diagnosis information, and generate a recommended sequence of diagnosis auxiliary reference information corresponding to the user's tongue surface feature information; mark the corresponding sequence elements in the recommended sequence of diagnosis auxiliary reference information corresponding to the user's tongue surface feature information for the disease diagnosis information corresponding to the marked element in the historical reference label object set corresponding to the user's tongue surface feature information;

[0093] In the process of obtaining the risk assessment value corresponding to the disease diagnosis information, the risk assessment value corresponding to each disease diagnosis information is equal to the maximum value of the facial image feature fitness corresponding to each face label corresponding to the corresponding disease diagnosis information in the historical reference label object set corresponding to the user's tongue surface feature information and the average value of the facial image feature fitness corresponding to the tongue label bound to it; each disease diagnosis information has one or more face labels corresponding to it in the historical reference label object set corresponding to the user's tongue surface feature information.

[0094] In this embodiment, if the historical reference label object set corresponding to the user's tongue surface feature information contains four elements, namely the first surface label, the second surface label, the third surface label and the fourth surface label,

[0095] If the first label and the second label correspond to the first disease diagnosis information (first type of disease);

[0096] If the third label corresponds to the second disease diagnosis information (second type of disease);

[0097] If the fourth label corresponds to the third disease diagnosis information (third type of disease);

[0098] If the average of the facial image feature fitness corresponding to the first face label and its bound tongue label and the facial image feature fitness is recorded as g1;

[0099] If the average value of the facial image feature fitness corresponding to the second face label and its bound tongue label and the facial image feature fitness is recorded as g2;

[0100] If the average value of the facial image feature fitness corresponding to the third face label and its bound tongue label and the facial image feature fitness is recorded as g3;

[0101] If the average of the facial image feature fitness corresponding to the fourth face label and its bound tongue label and the facial image feature fitness is recorded as g4;

[0102] If g1>g2,

[0103] Then the risk assessment value corresponding to the first disease diagnosis information is g1; the risk assessment value corresponding to the second disease diagnosis information is g3; and the risk assessment value corresponding to the third disease diagnosis information is g4.

[0104] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0105] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent image processing method for tongue surface diagnosis, characterized in that: The method comprises the following steps: S1. Collecting the user's facial image and tongue image through a tongue diagnosis instrument, and extracting features from the collected images based on image recognition technology to generate user tongue feature information, wherein the user tongue feature information includes multi-dimensional facial feature information, multi-dimensional tongue feature information, and device collection status information; S2, comparing the multi-dimensional facial feature information in the user's tongue and face feature information with the historical label data preset in the database, generating the facial feature adaptation degree of the user based on the corresponding historical label data and the facial feature adaptation adjustment coefficient based on the corresponding historical label data; S3, comparing the tongue image feature multi-dimensional information in the user's tongue surface feature information with the historical label data preset in the database, generating the user's tongue image feature adaptation degree based on the corresponding historical label data and the tongue image feature adaptation adjustment coefficient based on the corresponding historical label data; S4, obtaining the binding relationship between the face label and the tongue label in the historical label data, and based on the analysis results in step S2 and step S3, locking the historical reference label object set corresponding to the user's tongue face feature information; S5, generating a recommended sequence of diagnostic auxiliary reference information corresponding to the user's tongue surface feature information based on the historical reference label object set corresponding to the user's tongue surface feature information and the disease diagnosis information corresponding to each label data in the historical data; The multi-dimensional facial feature information includes the color level of each viscera partition of the face, the color level of the viscera partition is equal to the color level corresponding to the average pixel value of the corresponding pixel point in the facial image area where the corresponding viscera partition is located, and the color level corresponding to the average pixel value of the pixel point is a constant preset in the database; the number of viscera partitions included in the multi-dimensional facial feature information is a preset constant; The multi-dimensional information of tongue image features includes the color grade of the tongue, the tongue coating area, the tongue coating thickness, and the position distribution of tongue cracks; the color grade of the tongue is equal to the color grade corresponding to the average value of the corresponding pixel points in the tongue image area where the corresponding tongue is located; The device acquisition status information includes the usage time of the tongue and facial diagnosis instrument and the interval between the last maintenance time of the tongue and facial diagnosis instrument and the current time; The calculation formula for generating the user's facial feature adaptability based on the corresponding historical label data in S2 is as follows: ， Among them, MS i represents the facial feature adaptation degree of the user based on the i-th face label in the corresponding historical label data; the face label represents the multi-dimensional facial feature information of the user's tongue feature information in the historical data; C （i，j） represents the absolute value of the difference between the jth parameter in the i-th face label in the corresponding historical label data and the jth parameter in the multidimensional information of the user's facial image feature; j1 represents the number of parameters in the i-th face label in the corresponding historical label data; MG represents the tolerance for the deviation fluctuation of the internal organs color level preset in the database; DBM{} represents the facial image feature adaptation comparison function, when C （i，j） ≥MG, then judge Otherwise, it is judged ; The calculation formula for generating the face feature adaptation adjustment coefficient based on the corresponding historical label data is as follows: ， Among them, PM i represents the facial feature adaptation adjustment coefficient of the i-th face label in the corresponding historical label data; TW represents the usage time of the tongue-face diagnosis instrument in the device collection status information; TW i Indicates the usage time of the tongue-face diagnostic instrument corresponding to the i-th face label in the corresponding historical label data; Indicates that the usage time of the same tag object model interval in the historical data is |TW-TW i | respectively corresponds to the average value of the color level difference of the facial viscera partition in the multi-dimensional information of the monitored facial features; TG represents the interval between the last maintenance time of the tongue and face diagnostic instrument and the current time in the equipment collection status information; Indicates that the interval between the last maintenance time of the tongue and facial diagnosis instrument for the same labeled object model in the historical data is |TG-TG i | respectively corresponds to the average value of the color level difference of the facial viscera partitions in the multi-dimensional information of the monitored facial features; the same labeled object model represents a facial model preset in the database and the color of each facial viscera partition in the preset facial model is fixed; max{} represents the operation of finding the maximum value; The calculation formula for generating the tongue image feature adaptability of the user based on the corresponding historical label data in S3 is as follows: ， Among them, SS i represents the adaptability of the user's tongue image feature based on the i-th tongue label in the corresponding historical label data; the tongue label represents the multi-dimensional information of the tongue image feature of the user's tongue surface feature information in the historical data; CS i represents the absolute value of the difference between the color level of the tongue in the multidimensional information of the tongue image feature and the color level of the tongue corresponding to the i-th tongue tag in the corresponding historical tag data; MG1 represents the tolerance of the tongue color level deviation fluctuation preset in the database; DBS{} represents the tongue image feature adaptation comparison function, when CS i ≥MG1, then judge Otherwise, it is judged SR i Represents the intersection area of ​​the tongue coating area in the multidimensional information of tongue image features and the tongue coating area corresponding to the i-th tongue label in the corresponding historical label data; SRB i represents the union area of ​​the tongue coating area in the multidimensional information of tongue image features and the tongue coating area corresponding to the i-th tongue label in the corresponding historical label data; SDi represents the absolute value of the difference between the tongue coating thickness in the multidimensional information of tongue image features and the tongue coating thickness corresponding to the i-th tongue label in the corresponding historical label data; SHi max Represents the maximum value between the tongue coating thickness in the multidimensional information of tongue image features and the tongue coating thickness corresponding to the i-th tongue label in the corresponding historical label data; SL i It indicates the ratio of the maximum overlap area between the position distribution of the tongue crack corresponding to the corresponding tongue label and the position distribution of the tongue crack in the multidimensional information of tongue image features to the area of ​​the tongue crack in the multidimensional information of tongue image features after adjusting the angle of the image belonging to the i-th tongue label in the corresponding historical label data; if there is no tongue crack in the multidimensional information of tongue image features, then SL i =0; The calculation formula for generating the tongue image feature adaptation adjustment coefficient based on the corresponding historical label data is as follows: ， Among them, PS i represents the tongue image feature adaptation adjustment coefficient of the i-th face label in the corresponding historical label data; μ represents the preset tongue face image feature conversion factor.

2. The method for intelligent image processing for tongue surface diagnosis according to claim 1, characterized in that: The S4 includes: S41, obtaining the binding relationship between the face label and the tongue label in the historical label data, and binding the face label and the tongue label that have the same collection time and the corresponding collection device are the same tongue and face diagnosis instrument; S42, obtaining the user's face feature adaptation degree MS based on the i-th face label in the corresponding historical label data i , and the user's tongue image feature adaptation degree SS based on the i-th tongue label in the corresponding historical label data i ; S43, obtaining a set of all face labels whose face image feature adaptation degree between the face image feature multi-dimensional information in the user's tongue face feature information is greater than or equal to the face image adaptation threshold and whose tongue image feature adaptation degree between the bound tongue label and the tongue image feature multi-dimensional information in the user's tongue face feature information is greater than or equal to the tongue image adaptation threshold as a first reference label object set; the face image adaptation threshold and the tongue image adaptation threshold are both constants preset in the database; S44, obtaining a facial feature adaptation adjustment coefficient based on the corresponding historical label data, and a tongue feature adaptation adjustment coefficient based on the corresponding historical label data; and recording a set of all face labels whose facial feature adaptation degree between the facial feature multidimensional information in the user's tongue face feature information is greater than or equal to the facial feature adaptation threshold based on the mapping value of the corresponding face label, and whose tongue feature adaptation degree between the bound tongue label and the tongue feature multidimensional information in the user's tongue face feature information is greater than or equal to the tongue image adaptation threshold based on the mapping value of the corresponding tongue label as a second reference label object set; The reference face labels are different, and the corresponding face image adaptation thresholds are different based on the mapping values ​​of the corresponding face labels; the mapping values ​​of the face image adaptation thresholds based on the corresponding face labels are equal to the quotient obtained by dividing the corresponding face image adaptation threshold by the sum of the face image feature adaptation adjustment coefficient of the corresponding face label and 1; the reference tongue labels are different, and the corresponding tongue image adaptation thresholds are different based on the mapping values ​​of the corresponding tongue labels; the mapping values ​​of the tongue image adaptation thresholds based on the corresponding tongue labels are equal to the quotient obtained by dividing the corresponding tongue image adaptation threshold by the sum of the tongue image feature adaptation adjustment coefficient of the corresponding tongue label and 1; S45. Record the union of the first reference label object set and the second reference label object set as the historical reference label object set corresponding to the user's tongue feature information, and mark the elements in the obtained historical reference label object set that belong to the first reference label object set.

3. The method for intelligent image processing for tongue surface diagnosis according to claim 2, characterized in that: The S5 includes: Obtain the disease diagnosis information corresponding to the face label corresponding to the kth element in the historical reference label object set corresponding to the user's tongue surface feature information and the tongue label bound to it in the historical data; summarize the obtained disease diagnosis information, and sort the disease diagnosis information in descending order according to the risk assessment value corresponding to each disease diagnosis information, and generate a recommended sequence of diagnosis auxiliary reference information corresponding to the user's tongue surface feature information; mark the corresponding sequence elements in the recommended sequence of diagnosis auxiliary reference information corresponding to the user's tongue surface feature information for the disease diagnosis information corresponding to the marked element in the historical reference label object set corresponding to the user's tongue surface feature information; In the process of obtaining the risk assessment value corresponding to the disease diagnosis information, the risk assessment value corresponding to each disease diagnosis information is equal to the maximum value of the facial image feature fitness corresponding to each face label corresponding to the corresponding disease diagnosis information in the historical reference label object set corresponding to the user's tongue surface feature information and the average value of the facial image feature fitness corresponding to the tongue label bound to it; each disease diagnosis information has one or more face labels corresponding to it in the historical reference label object set corresponding to the user's tongue surface feature information.

4. An image intelligent processing system for tongue surface diagnosis, the system applying an image intelligent processing method for tongue surface diagnosis according to any one of claims 1 to 3, characterized in that: The system includes the following modules: A tongue surface feature acquisition module, which acquires the user's facial image and tongue image through a tongue surface diagnostic instrument, and extracts features from the acquired images based on image recognition technology to generate user tongue surface feature information, wherein the user tongue surface feature information includes multi-dimensional facial feature information, multi-dimensional tongue feature information, and device acquisition status information; A facial feature adaptation analysis module, which compares the facial feature multi-dimensional information in the user's tongue and face feature information with the historical label data preset in the database, and generates the facial feature adaptation degree of the user based on the corresponding historical label data and the facial feature adaptation adjustment coefficient based on the corresponding historical label data; A tongue image feature adaptation analysis module, which compares the tongue image feature multi-dimensional information in the user's tongue surface feature information with the historical label data preset in the database, and generates the user's tongue image feature adaptation degree based on the corresponding historical label data and the tongue image feature adaptation adjustment coefficient based on the corresponding historical label data; A label object analysis module, which obtains the binding relationship between the face label and the tongue label in the historical label data, and locks the historical reference label object set corresponding to the user's tongue and face feature information based on the analysis results in the face feature adaptation analysis module and the tongue feature adaptation analysis module; An image auxiliary information management module generates a recommended sequence of diagnostic auxiliary reference information corresponding to the user's tongue surface feature information based on a historical reference label object set corresponding to the user's tongue surface feature information and disease diagnosis information corresponding to each label data in the historical data.

5. The image intelligent processing system for tongue surface diagnosis according to claim 4, characterized in that: The facial feature adaptation analysis module includes a facial feature adaptation degree analysis unit and a facial feature adaptation adjustment coefficient analysis unit. The facial feature adaptability analysis unit compares the facial feature multi-dimensional information in the user's tongue and face feature information with the historical label data preset in the database to generate the facial feature adaptability of the user based on the corresponding historical label data; The facial feature adaptation adjustment coefficient analysis unit compares the facial feature multi-dimensional information in the user's tongue feature information with the historical label data preset in the database to generate the facial feature adaptation adjustment coefficient of the user based on the corresponding historical label data.

Images