An image-based tongue color and fur color identification method and system, an intelligent terminal, and a storage medium

Abstract

The application relates to an image-based tongue fur color identification method and system, an intelligent terminal and a storage medium, and relates to the tongue fur identification technical field.The application comprises the following steps: when a preset tongue fur shooting device shoots tongue fur, picture analysis is performed to determine the tongue posture in the picture; when the tongue posture is inconsistent with a preset standard posture, an adjustment prompt of the tongue posture is output, and a tongue fur image under the standard posture is shot; image region segmentation is performed based on the tongue fur image, and refined tongue fur features in the segmented image are extracted; the refined tongue fur features are input into a preset tongue fur color identification model for analysis, so that the fur color information of the tongue fur is identified and output.The application has the effect of improving the diagnosis accuracy of the tongue fur color when different persons detect the tongue posture.

Description

TECHNICAL FIELD

[0001] The present application relates to the tongue fur identification technical field, especially to a tongue fur color identification method and system based on images, an intelligent terminal and a storage medium. BACKGROUND

[0002] Discriminating the color of tongue fur is one of the important diagnostic steps in tongue diagnosis, so it is of great significance to improve the convenience and reliability of tongue fur color discrimination.

[0003] In the related art, with the development of intelligent medical technology, a visual image shooting device is used to shoot images of tongue fur when diagnosing, and a computer is used to identify the features of the tongue fur images and obtain a diagnosis result meeting the features.

[0004] In the above related technology, when shooting images of tongue fur, the overall posture of the tongue will vary with different people, which makes the computer less accurate in positioning the position of the tongue fur on the tongue, and is not conducive to improving the accuracy of diagnosis. SUMMARY

[0005] In order to improve the diagnostic accuracy of tongue fur color when the posture of the tongue of different people varies, the present application provides a tongue fur color identification method and system based on images, an intelligent terminal and a storage medium.

[0006] In the first aspect, the present application provides a tongue fur color identification method based on images, which adopts the following technical solution:

[0007] A tongue fur color identification method based on images, comprising:

[0008] When the tongue is shot by a preset tongue fur shooting device, picture analysis is performed to determine the posture of the tongue in the picture;

[0009] When the posture of the tongue is inconsistent with the preset standard posture, an adjustment prompt of the posture of the tongue is output, and a tongue fur image under the standard posture is shot;

[0010] Based on the tongue fur image, image region segmentation is performed, and refined tongue fur features in the segmented image are extracted;

[0011] The refined tongue fur features are input into a preset tongue fur color identification model for analysis to identify and output the color information of the tongue fur.

[0012] By adopting the above technical solution, the posture of the tongue when shooting tongue fur is prompted accordingly, so that the person can adjust the posture accordingly, and the tongue fur image shot under the standard posture can clearly show the actual situation, which helps to improve the accuracy of tongue fur color detection by different people.

[0013] Optionally, the step of adjusting the prompt according to the tongue posture comprises:

[0014] inputting the tongue posture and the standard posture into a preset fitting model to output posture adjustment parameters for fitting the tongue posture with the standard posture;

[0015] generating a prompt animation for adjusting the tongue posture based on the posture adjustment parameters;

[0016] collecting a tongue adjustment posture during the prompt animation, and issuing an adjustment completion prompt when the tongue adjustment posture and the standard posture are fitted;

[0017] if the tongue adjustment posture and the standard posture are not fitted, assisting the tongue in posture adjustment by a preset auxiliary adjustment method until the tongue posture and the standard posture are fitted.

[0018] By using the above technical solution, the tongue posture is adjusted in time during the adjustment of the tongue with a non-standard posture, the tongue adjustment posture is fitted with the required standard posture, and when the person cannot adjust the tongue to the standard posture by himself / herself, the tongue posture is adjusted by the auxiliary adjustment, so that the person can easily adjust the tongue to the standard posture, which helps to obtain a clear tongue coating image.

[0019] Optionally, the auxiliary adjustment method comprises:

[0020] comparing and analyzing the local abnormal tongue body of the tongue adjustment posture and the standard posture;

[0021] performing feature analysis on the local abnormal tongue body to obtain a tongue body adjustment position and a tongue body adjustment mode, the tongue body adjustment mode comprising local tongue body support and local depression;

[0022] indicating a preset tongue body auxiliary device to adjust the local abnormal tongue body based on the tongue body adjustment position and the tongue body adjustment mode;

[0023] issuing an image collection prompt when the local abnormal tongue body is adjusted.

[0024] By using the above technical solution, when the tongue has a local part that is difficult to adjust during the adjustment, the adjustment position and the adjustment mode of the abnormal local tongue body are analyzed, and the appropriate adjustment mode is selected to assist the adjustment of the tongue body, so that the tongue body can be fully expanded to clearly display the tongue coating, which helps to improve the recognition accuracy.

[0025] Optionally, when the tongue coating image in the standard posture is shot, it comprises:

[0026] when the main visual camera performs wide-angle shooting on the tongue body, the auxiliary visual camera performs side-symmetrical shooting on the tongue body to obtain a tongue coating center image and a tongue body side image.

[0027] Identify the coincident feature point positions of the tongue fur center image and the tongue body side image;

[0028] Integrate the tongue fur center image and the tongue body side image according to the coincident feature points to obtain an optimized image;

[0029] Divide the tongue fur region of the optimized image according to the tongue fur features, and extract a tongue fur image after the tongue fur region is reduced.

[0030] By adopting the above technical solutions, the main vision camera captures the center image of the tongue fur to collect the main features of the tongue fur, and the auxiliary vision camera captures the side of the tongue body to further collect the detailed features at the edge, and an optimized image with rich image features is generated by integration, and the required tongue fur image is further divided, which can obtain complete tongue fur features while reducing the size of the image, thereby helping to reduce memory occupation.

[0031] Optionally, the step of obtaining the tongue fur center image of the tongue body further comprises:

[0032] Analyze and identify the abnormal tongue fur color of the tongue fur center image;

[0033] When the abnormal tongue fur color is consistent with the preset food attachment color, extract the attachment area of the attachment color;

[0034] Perform regional side tongue fur feature analysis on the attachment area to determine the associated tongue fur color;

[0035] Replace the abnormal tongue fur color of the attachment area based on the associated tongue fur color.

[0036] By adopting the above technical solutions, the abnormal color of the tongue fur center image is adjusted, so that when the tongue fur is uncomfortable due to food color attachment, the corresponding color correction is performed, so that the subsequent tongue fur analysis is not prone to analysis errors, which helps to improve the analysis accuracy of the tongue fur color.

[0037] Optionally, analyzing and identifying the abnormal tongue fur color of the tongue fur center image further comprises:

[0038] Analyze and identify the normal tongue fur color of the tongue fur center image;

[0039] Match the normal tongue fur color to determine the priority lighting parameters for observing the tongue fur color;

[0040] Indicate the preset lighting lamp to adjust the lighting brightness and lighting color according to the priority lighting parameters, and obtain a secondary tongue fur center image;

[0041] The conventional tongue fur color in the secondary tongue fur center image is subjected to image separation processing to determine an abnormal tongue fur image and a conventional tongue fur image;

[0042] Color analysis is performed on the abnormal tongue fur image to determine the abnormal tongue fur color.

[0043] By adopting the technical solution, the conventional tongue fur color is separated to screen out the conventional tongue fur image and the abnormal tongue fur image, so that the abnormal tongue fur color can be analyzed.

[0044] Optionally, after the fur color information of the tongue fur is identified and output, the following steps are further included:

[0045] The tongue fur image is analyzed to determine the tongue fur thickness;

[0046] The tongue fur thickness is compared and analyzed to determine whether the tongue fur thickness is greater than a preset reference thickness;

[0047] If yes, a preset cleaning parameter is matched, and a preset tongue fur cleaning device is instructed to clean the tongue fur;

[0048] The tongue fur image obtained again is subjected to identification analysis to determine second tongue fur fur color information.

[0049] By adopting the technical solution, after the tongue fur with a large thickness on the tongue body is subjected to initial image shooting, the tongue fur is cleaned accordingly, so that the deep tongue fur fur color can be further observed after the tongue fur thickness is reduced, and the corresponding detection parameter is obtained, thereby helping to further know more fur color information of the tongue fur.

[0050] In a second aspect, the application provides a tongue color and fur color identification system based on images, which adopts the following technical solution:

[0051] A tongue color and fur color identification system based on images includes:

[0052] A posture analysis module analyzes a picture to determine a tongue posture in the picture when the tongue is subjected to tongue fur shooting;

[0053] A posture adjustment module outputs an adjustment prompt of the tongue posture when the tongue posture is inconsistent with a preset standard posture, and shoots a tongue fur image in a standard posture after the tongue posture is adjusted;

[0054] A fur color identification module divides an image area based on the tongue fur image, and extracts a refined tongue fur feature in the divided image;

[0055] The refined tongue fur feature is input into a preset tongue fur color identification model for analysis, so as to identify and output fur color information of the tongue fur;

[0056] A memory for storing a program of any one of the image-based tongue color and fur color identification methods;

[0057] A processor, the program in the memory can be loaded and executed by the processor to implement the image-based tongue color and fur color identification method of any one.

[0058] By adopting the above technical solution, the tongue posture during tongue fur shooting is prompted accordingly, so that the personnel can adjust the posture accordingly, and the tongue fur image shot in the standard posture can clearly show the actual situation, which helps to improve the accuracy of tongue fur color detection by different personnel.

[0059] In a third aspect, the present application provides an intelligent terminal, which adopts the following technical solution:

[0060] An intelligent terminal, comprising a memory and a processor, and the memory stores a computer program capable of being loaded and executed by the processor to implement the image-based tongue color and fur color identification method described above.

[0061] By adopting the above technical solution, through the use of the intelligent terminal,

[0062] In a fourth aspect, the present application provides a computer storage medium, which can store the corresponding program and has the characteristics of improving the accuracy of tongue fur color detection by different tongue postures, and adopts the following technical solution:

[0063] A computer readable storage medium, which stores a computer program capable of being loaded and executed by the processor to implement the image-based tongue color and fur color identification method described above.

[0064] By adopting the above technical solution, the computer program of the image-based tongue color and fur color identification method is stored in the storage medium, the tongue posture during tongue fur shooting is prompted accordingly, so that the personnel can adjust the posture accordingly, and the tongue fur image shot in the standard posture can clearly show the actual situation, which helps to improve the accuracy of tongue fur color detection by different personnel.

[0065] In summary, the present application includes at least one of the following beneficial technical effects:

[0066] 1. The tongue posture during tongue fur shooting is prompted accordingly, so that the personnel can adjust the posture accordingly, and the tongue fur image shot in the standard posture can clearly show the actual situation, which helps to improve the accuracy of tongue fur color detection by different personnel;

[0067] 2. When the tongue appears to be locally difficult to adjust during the adjustment process, the abnormal local tongue is analyzed in terms of adjustment position and adjustment method, and the appropriate adjustment method is selected to assist in adjusting the tongue, so that the tongue can be fully expanded to clearly display the tongue fur, which helps to improve the identification accuracy;

[0068] 3. The tongue fur center image is adjusted for abnormal color, so that when the tongue fur appears to be caused by food color attachment, the corresponding color correction is performed, so that the tongue fur is not prone to analysis errors during subsequent analysis of the corresponding fur color, which helps to improve the analysis accuracy of the tongue fur color. BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 is a method flowchart of steps S100 to S103 in the present application.

[0070] Figure 2 is a method flowchart of steps S200 to S203 in the present application.

[0071] Figure 3 is a method flowchart of steps S300 to S303 in the present application.

[0072] Figure 4 is a method flowchart of steps S400 to S403 in the present application.

[0073] Figure 5 is a method flowchart of steps S500 to S503 in the present application.

[0074] Figure 6 is a method flowchart of steps S600 to S604 in the present application.

[0075] Figure 7 is a method flowchart of steps S700 to S703 in the present application. DETAILED DESCRIPTION

[0076] In order to make the purpose, technical solutions and advantages of the present application clearer, the following will combine the drawings of the specification and the examples to further describe the present application in detail. It should be understood that the specific examples described herein are only used to explain the present application and not to limit the present application. Figures 1-7

[0077] The embodiments of the present application will be further described in detail below in combination with the drawings of the specification.

[0078] ​The embodiment of the application discloses an image-based tongue color and fur color identification method, which analyzes the posture when taking a tongue fur image of a tongue, adjusts the posture when the tongue posture and the standard posture are inconsistent, and makes the tongue posture in a posture that clearly displays the color of the tongue fur, so that the required tongue fur image can be obtained, the tongue fur color information of the tongue fur image is identified according to a trained identification model, the probability of incomplete acquisition of the tongue fur image caused by different postures is reduced, and the detection accuracy and reliability of the tongue fur color are improved.

[0079] Reference Figure 1 The method flow of the image-based tongue color and fur color identification method comprises the following steps:

[0080] Step S100: When a tongue fur shooting device shoots the tongue fur, picture analysis is performed to determine the posture of the tongue in the picture.

[0081] The tongue fur shooting device is an image shooting device provided with a high-definition resolution visual camera, and is provided with a shooting chamber for a person to align the open mouth and shoot the tongue fur image of the tongue. The posture of the tongue is identified by analyzing the picture posture of the image during shooting, and the identified dynamic characteristics of the tongue are defined as the posture of the tongue.

[0082] Step S101: When the posture of the tongue is inconsistent with the preset standard posture, an adjustment prompt of the posture of the tongue is output, and the tongue fur image under the standard posture is shot.

[0083] The standard posture is a posture in which the tongue is stretched forward when the mouth is opened, so that the tongue is fully expanded, and the tongue fur can be fully displayed. The specific posture is collected by the staff and the image characteristics of the standard posture are established. By comparing the identified posture of the tongue with the preset standard posture, whether the posture of the tongue is consistent with the standard posture can be determined.

[0084] If the posture of the tongue is consistent with the standard posture, it indicates that the tongue fur on the surface of the tongue can be fully displayed during the expansion of the tongue, and the effective tongue fur surface image can be obtained, which is beneficial to the subsequent identification and detection of the tongue fur color.

[0085] If the posture of the tongue is inconsistent with the standard posture, it indicates that the current posture of the tongue cannot well display the tongue fur surface during the expansion of the tongue, and the tongue may be bent or partially blocked, so that the posture needs to be adjusted. At this time, a display screen is arranged on the tongue fur shooting device, the posture of the tongue is identified, and a posture adjustment prompt is given, so that the staff can adjust the tongue to the standard posture according to the prompt, and the image is shot, and the obtained image is defined as the tongue fur image.

[0086] Step S102: image region segmentation is performed based on the tongue coating image, and refined tongue coating features in the segmented image are extracted.

[0087] In the tongue coating image, by dividing the tongue body into regions, multiple small region images can be obtained. The tongue coating features of the segmented image are collected and extracted accordingly to obtain the tongue coating features, which facilitates subsequent tongue coating color recognition.

[0088] Step S103: input the refined tongue coating features into a preset tongue coating color recognition model for analysis to recognize and output the coating color information of the tongue coating.

[0089] The tongue coating color recognition model is a pre-constructed image recognition network model. Through deep learning on a large number of different classified tongue coating color images, the tongue coating color recognition model can recognize the tongue coating color of the input tongue coating image and output the corresponding coating color information, so that personnel can know the corresponding tongue coating color situation when performing tongue coating color detection.

[0090] Referring to Figure 2 The step of adjusting and prompting according to the tongue posture comprises:

[0091] Step S200: input the tongue posture and the standard posture into a preset fitting model to output posture adjustment parameters for fitting the tongue posture and the standard posture.

[0092] The fitting model is a pre-established posture analysis network model, which can dynamically adjust and fit two models to output the image dynamic simulation process formed when the two models are overlapped. Therefore, by dynamically analyzing the tongue posture and the standard posture, the tongue movement parameters when the tongue posture is adjusted to the standard posture can be obtained, and the tongue movement parameters are defined as the posture adjustment parameters.

[0093] Step S201: generate a prompt animation when the tongue posture is adjusted based on the posture adjustment parameters.

[0094] After generating the dynamic adjustment parameters, the tongue posture image dynamically simulated by the dynamic adjustment parameters is output to the display screen, so that personnel can know whether the current tongue adjustment is consistent with the personnel's expectation.

[0095] Step S202: collect the tongue adjustment posture during the prompt process of the prompt animation, and issue an adjustment completion prompt when the tongue adjustment posture and the standard posture are fitted.

[0096] During the prompt animation process, the tongue posture is collected in real time, and the tongue adjustment posture during this process is defined as the tongue adjustment posture. When the tongue adjustment posture and the standard posture are fitted, it means that the adjustment of the tongue posture is completed, and an adjustment completion prompt is issued at this time.

[0097] Step S203: If the tongue adjustment posture does not match the standard posture, the preset auxiliary adjustment method is used to assist the tongue in posture adjustment until the tongue posture matches the standard posture.

[0098] If the tongue adjustment posture does not match the standard posture, it means that even if the personnel are prompted in detail, the personnel cannot be fully adjusted, and there is a certain difference between the tongue adjustment posture and the standard posture. At this time, the corresponding external auxiliary method is set to assist the tongue posture adjustment, so that the tongue adjustment posture matches the standard posture, thereby further improving the comprehensive tongue surface image when different personnel shoot the tongue coating image. The specific steps of the auxiliary adjustment method are described in detail later.

[0099] Reference Figure 3 The auxiliary adjustment method includes:

[0100] Step S300: Comparative analysis of local abnormal tongue body of tongue adjustment posture and standard posture.

[0101] By analyzing the features of the tongue adjustment posture and the standard posture, the local features that do not coincide can be obtained, and the tongue body corresponding to the local features that do not coincide can be divided and defined as local abnormal tongue body.

[0102] Step S301: Feature analysis of local abnormal tongue body to obtain tongue body adjustment position and tongue body adjustment mode, including local tongue body support and local depression.

[0103] By analyzing the contour features of the image of the local abnormal tongue body, the tongue body external force action position required for the auxiliary adjustment of the local abnormal tongue body can be obtained, which is defined as the tongue body adjustment position, such as the tongue tip pressing position or the tongue body edge support position. Different adjustment positions have corresponding adjustment modes, such as tongue body support and local depression. By establishing a corresponding adjustment auxiliary parameter database, different tongue body adjustment positions and tongue body adjustment modes are associated and stored in the adjustment auxiliary parameter database. The local abnormal tongue body corresponding features associated with the adjustment mode are also stored. When the local features of the local abnormal tongue body are input, the corresponding tongue body adjustment mode and tongue body adjustment position can be automatically matched and called for adjustment.

[0104] Step S302: Based on the tongue body adjustment position and the tongue body adjustment mode, the preset tongue body auxiliary device is instructed to adjust the local abnormal tongue body.

[0105] The tongue body auxiliary device is a adjustment support pre-set on the tongue coating shooting device, which is connected with a micro transmission mechanical arm to enable the adjustment support to contact the tongue body and achieve accurate position support or depression, thereby achieving the adjustment of the local abnormal tongue body.

[0106] Step S303: An image acquisition prompt is sent when the local abnormal tongue body is adjusted.

[0107] When the adjustment of the local abnormal tongue body is completed, a more comprehensive tongue image can be acquired, and at this time, an image acquisition prompt is sent to take a picture of the tongue image.

[0108] Referring to Figure 4 , when taking a picture of the tongue image in a standard posture, the method comprises:

[0109] Step S400: When the main visual camera takes a wide-angle picture of the tongue body, the auxiliary visual camera takes a side-symmetrical picture of the tongue body to obtain a tongue center image and a tongue side image.

[0110] When taking a picture of the tongue image, the main data camera and the auxiliary visual camera are set to take pictures from different positions to obtain an image at the center position of the tongue and an image along the side of the tongue. The image at the center position is defined as the tongue center image, and the image along the side of the tongue is defined as the tongue side image. The overall image and the detailed image of the side are acquired to obtain a more comprehensive distribution of the tongue.

[0111] Step S401: Identify the coincident feature point position of the tongue center image and the tongue side image.

[0112] The tongue center image and the tongue side image have coincident feature points in terms of features when taking pictures. The coincident feature points are marked.

[0113] Step S402: Integrate the tongue center image and the tongue side image according to the coincident feature points to obtain an optimized image.

[0114] The coincident feature points are marked by an image fusion network training model, and the tongue side image and the tongue center image are fused to form a three-dimensional image. The three-dimensional image can clearly display the front tongue image and the side tongue image. The three-dimensional image is defined as an optimized image for subsequent tongue image analysis.

[0115] Step S403: Divide the tongue area of the optimized image according to the tongue features, and extract a reduced tongue image of the tongue area.

[0116] The optimized image is divided into regions, and the image after division is analyzed for tongue features to screen out image regions with tongue features, thereby achieving the purpose of reducing the image size. Subsequent analysis of the tongue color information can reduce storage occupation and improve the efficiency of image uploading and analysis.

[0117] When steps S400 to S403 are performed, the main visual camera captures the center image of the tongue coating to collect the main features of the tongue coating, and the auxiliary visual camera captures the side of the tongue body to further collect the detailed features at the edge, and the optimized image with rich image features is generated by integration, and the required tongue coating image is further divided, which can obtain complete tongue coating features while reducing the size of the image, thereby reducing the memory occupation.

[0118] Referring to Figure 5 , the steps of acquiring the center image of the tongue coating of the tongue body further include:

[0119] Step S500: analyzing and identifying the abnormal tongue coating color of the center image of the tongue coating.

[0120] The abnormal tongue coating color is the tongue coating color that is different from the conventional tongue coating color, indicating that the tongue coating color is in an abnormal state. The abnormal color may be affected by the diet of the person, so that the tongue coating color changes to an abnormal color. By recording and collecting the influence of different diets and foods on the tongue coating, the abnormal color generated by different diets on the tongue coating can be known, and an abnormal color recognition network model is established. Different abnormal tongue coating colors are stored in the database of the abnormal color recognition network model. When the center image of the tongue coating is input, it is automatically identified and analyzed whether there is an abnormal tongue coating color.

[0121] Step S501: when the abnormal tongue coating color is consistent with the preset food attachment color, the attachment area of the attachment color is extracted.

[0122] If the abnormal tongue coating color is consistent with the preset food attachment color, it means that the current tongue coating color is affected by the food, and the food color will have a certain influence on the tongue coating color. However, a part of the original tongue coating color will remain in the area of the abnormal tongue coating color. Therefore, the original tongue coating color can be presented in the detected tongue coating image by exchanging the color of the area corresponding to the food attachment color, and the personnel does not need to clean the tongue coating, thereby reducing the probability of inaccurate detection results caused by damaging the original tongue coating on the tongue.

[0123] Step S502: analyzing the area side tongue coating features of the attachment area to determine the associated tongue coating color.

[0124] Before replacing the tongue coating color, the features of the tongue coating color of the area around the abnormal tongue coating color are collected and analyzed to determine the tongue coating color that is different from the abnormal tongue coating color. The tongue coating color is used as the associated tongue coating color, and the associated tongue coating color represents the original tongue coating color on the tongue coating.

[0125] Step S503: Replace the abnormal tongue fur color of the attached area based on the associated tongue fur color.

[0126] After learning the associated tongue fur color, replace the abnormal tongue fur color with the associated tongue fur color, so that the normal tongue fur color under the influence of food abnormal color can be learned.

[0127] When steps S500 to S503 are executed, the abnormal color of the tongue fur center image is adjusted, so that when the tongue fur appears uncomfortable due to food color attachment, the corresponding color correction is performed, so that the tongue fur is not easy to cause analysis error when subsequent analysis of the corresponding fur color is performed, which helps to improve the analysis accuracy of the tongue fur color.

[0128] Reference Figure 6 When the abnormal tongue fur color of the tongue fur center image is analyzed and identified, it also includes:

[0129] Step S600: Analyze and identify the normal tongue fur color of the tongue fur center image.

[0130] During the identification of the abnormal tongue fur color, most of the tongue fur color is affected by the food in the diet, so that most of the tongue fur color is different from the original tongue fur color. The original tongue fur color can be made more obvious by changing the light of the lighting lamp, so that the normal tongue fur color corresponding to the original tongue fur color is displayed by color, and the normal tongue fur color in the tongue fur center image is identified for further analysis.

[0131] Step S601: Match the normal tongue fur color to determine the priority lighting parameter for observing the tongue fur color.

[0132] The priority lighting parameter represents the lighting brightness and lighting color of the light. Different lighting colors can further increase the display effect of the corresponding normal tongue fur color of the tongue body. For example, in a 30-watt light brightness under white light color, white tongue fur can be displayed more clearly.

[0133] By establishing a significant lighting parameter database, different priority lighting parameters are stored in the significant lighting parameter database, and the priority lighting parameters corresponding to the significant lighting parameters are also stored. When the normal tongue fur color is input, the corresponding priority lighting parameter can be automatically matched and output.

[0134] Step S602: Direct the preset lighting lamp to adjust the lighting brightness and lighting color according to the priority lighting parameter, and obtain a secondary tongue fur center image.

[0135] After the priority lighting parameter that can significantly increase the original tongue fur color is selected, the lighting lamp on the tongue fur shooting device is instructed to adjust the color of the lighting lamp according to the priority lighting parameter, and the brightness of the lighting lamp is gradually adjusted in an increasing manner. Each time the brightness is adjusted, a tongue fur center image is obtained. The obtained multiple tongue fur center images are compared, and a tongue fur center image with high definition is selected and defined as a secondary tongue fur center image.

[0136] Step S603: Image separation processing is performed on the conventional tongue fur color in the secondary tongue fur center image to determine the abnormal tongue fur image and the conventional tongue fur image.

[0137] During the image separation processing, the secondary tongue fur center image is divided into two parts. One part removes the tongue fur color of the abnormal tongue fur image to obtain the conventional tongue fur image, and the other part removes the conventional tongue fur color to obtain the abnormal tongue fur image.

[0138] Step S604: Color analysis is performed on the abnormal tongue fur image to determine the abnormal tongue fur color.

[0139] After the abnormal tongue fur image is separated, the abnormal tongue fur color of the tongue fur can be directly extracted by comparing and analyzing the tongue fur features of the abnormal tongue fur image.

[0140] During the identification process of the abnormal tongue fur color, the conventional tongue fur image and the abnormal tongue fur image are screened by corresponding separation of the conventional tongue fur color during the execution of steps S600 to S603, so as to facilitate the analysis of the abnormal tongue fur color.

[0141] Referring to Figure 7 After identifying and outputting the fur color information of the tongue fur, the method further includes:

[0142] Step S700: Analyzing the tongue fur image to determine the thickness of the tongue fur.

[0143] During the analysis of the tongue fur color of the tongue fur, only the tongue fur color of the surface layer of the tongue fur can be analyzed, which is not conducive to the analysis of the distribution of the tongue fur with a relatively thin thickness on the tongue body. Therefore, during the analysis of the fur color of the tongue fur, the relatively thick tongue fur needs to be further cleaned. During the analysis, the thickness of the tongue fur is determined by measuring the wavelength of the reflected light source through light source reflection detection of the tongue fur.

[0144] Step S701: Comparative analysis to determine whether the thickness of the tongue fur is greater than a preset reference thickness.

[0145] The reference thickness is a tongue fur thickness value set by the staff according to the collected data. The reference thickness value represents the thickness of the tongue fur under normal circumstances. When the thickness exceeds this value, it indicates that the thickness of the tongue fur is abnormal.

[0146] Step S702: If greater, match the preset cleaning parameter, and instruct the preset tongue cleaning device to clean the tongue coating.

[0147] The tongue cleaning device is a preset cleaning brush. When the thickness of the tongue coating is greater than the reference thickness, it indicates that the specific conditions of the inner layer of the tongue coating cannot be observed at this time. The cleaning parameter is the number of cleaning times and the cleaning strength required when the thickness of the tongue coating is less than or equal to the reference thickness during cleaning of the tongue coating by the tongue cleaning device. The cleaning parameter can be obtained by performing corresponding experiments and recording. By pre-establishing a tongue coating thickness adjustment database, different cleaning parameters are stored in the tongue coating thickness adjustment database, and the tongue coating thickness corresponding to the cleaning parameter is also stored. When the tongue coating thickness is input, the corresponding cleaning parameter is automatically matched. After the corresponding cleaning parameter is matched, the tongue cleaning device is instructed to clean according to the cleaning parameter.

[0148] Step S703: The tongue coating image obtained again is analyzed to determine the second tongue coating color information.

[0149] After the thick tongue coating is cleaned, the tongue coating image is obtained again, which is used as the image analysis source for the tongue coating color. The tongue coating image obtained again is input into the tongue coating color recognition model to identify the tongue coating color information in the deep layer of the tongue coating.

[0150] When steps S700 to S703 are performed, the thick tongue coating on the tongue is initially imaged, and then the corresponding tongue coating is cleaned, so that the thickness of the tongue coating is reduced, the deep tongue coating color can be further observed, and the corresponding detection parameter can be obtained, thereby helping to further obtain more coating color information of the tongue coating.

[0151] Based on the same inventive concept, the embodiment of the present application provides a tongue color and coating color recognition system based on images, comprising:

[0152] The posture analysis module analyzes the picture when the tongue is imaged to determine the posture of the tongue in the picture.

[0153] The posture adjustment module outputs an adjustment prompt for the tongue posture when the tongue posture is inconsistent with the preset standard posture, and images the tongue coating in the standard posture after adjusting the tongue posture.

[0154] The coating color recognition module performs image region segmentation based on the tongue coating image, and extracts the refined tongue coating features in the segmented image.

[0155] The refined tongue coating features are input into the preset tongue coating color recognition model for analysis to identify and output the coating color information of the tongue coating.

[0156] A memory for storing a program of any one of the image-based tongue color and fur color recognition methods.

[0157] A processor, the program in the memory can be loaded and executed by the processor to implement any one of the image-based tongue color and fur color recognition methods.

[0158] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional modules is exemplified, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0159] The embodiment of the application provides a computer readable storage medium, which stores a computer program capable of being loaded and executed by a processor to implement an image-based tongue color and fur color recognition method.

[0160] The computer storage medium includes, for example, a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk and various media that can store program codes.

[0161] Based on the same inventive concept, the embodiment of the application provides an intelligent terminal, which comprises a memory and a processor, and the memory stores a computer program capable of being loaded and executed by the processor to implement an image-based tongue color and fur color recognition method.

[0162] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional modules is exemplified, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0163] The above are preferred embodiments of the application, and are not intended to limit the protection scope of the application, any feature disclosed in the specification (including the abstract and drawings) can be replaced by other equivalent or similar features unless specifically described. That is, unless specifically described, each feature is only an example of a series of equivalent or similar features.

Claims

1. A method for identifying tongue color and tongue coating color based on images, characterized in that, include: The preset tongue coating imaging device performs image analysis when photographing the tongue coating to determine the tongue posture in the image; When the tongue posture is inconsistent with the preset standard posture, output the tongue posture adjustment prompt and take a picture of the tongue coating under the standard posture; When taking images of the tongue coating in a standard pose, the following should be included: When the main vision camera takes a wide-angle shot of the tongue, the auxiliary vision camera takes a symmetrical side shot of the tongue to obtain a central image of the tongue coating and a side image of the tongue. Identify the overlapping feature points between the central image of the tongue coating and the lateral image of the tongue body; integrate the central image of the tongue coating and the lateral image of the tongue body based on the overlapping feature points to obtain an optimized image; divide the optimized image into tongue coating regions based on tongue coating features, and extract the tongue coating image after the tongue coating region is reduced; The steps involved in obtaining a central image of the tongue coating also include: The system analyzes and identifies abnormal tongue coating colors in the central image of the tongue coating. When the abnormal tongue coating color matches the preset food attachment color, the attachment area of ​​the food attachment color is extracted. The system performs peripheral tongue coating feature analysis on the attachment area to determine the associated tongue coating color. The abnormal tongue coating color in the attachment area is replaced based on the associated tongue coating color. Image region segmentation is performed based on the tongue coating image, and refined tongue coating features are extracted from the segmented image. The refined tongue coating features are then input into a preset tongue coating color recognition model for analysis, in order to identify and output the tongue coating color information.

2. The image-based tongue color and coating color recognition method according to claim 1, characterized in that, The steps for adjusting tongue posture include: Input the tongue posture and standard posture into the preset fitting model to output the posture adjustment parameters for fitting the tongue posture and standard posture. Animated prompts for tongue posture adjustment are generated based on posture adjustment parameters. The animation captures the tongue's posture adjustment during the prompting process, and a prompt indicating that adjustment is complete is issued when the tongue's adjusted posture fits the standard posture. If the tongue's adjusted posture does not fit the standard posture, a preset auxiliary adjustment method will be used to assist the tongue in adjusting its posture until the tongue's posture fits the standard posture.

3. The image-based tongue color and coating color recognition method according to claim 2, characterized in that, The auxiliary adjustment method includes: Comparative analysis of localized abnormalities in the tongue during regulated posture and in the standard posture; Characteristic analysis of local abnormalities of the tongue body is used to obtain the tongue body adjustment position and tongue body adjustment mode, which includes local tongue body support and local downward pressure. Based on the tongue adjustment position and tongue adjustment mode, a preset tongue auxiliary device is used to adjust the tongue in local abnormalities. An image acquisition prompt will be issued when the local abnormal tongue body completes adjustment.

4. The image-based tongue color and coating color recognition method according to claim 1, characterized in that, The analysis and identification of abnormal tongue coating colors in the central image of the tongue coating also includes: The analysis identifies the typical tongue coating color in the center image of the tongue coating. Priority lighting parameters for significantly observing tongue coating color were determined by matching the standard tongue coating color. The system instructs the preset lighting to adjust the brightness and color of the lighting according to the priority lighting parameters, and acquires a secondary image of the center of the tongue coating. Image separation processing is performed on the normal tongue coating color in the secondary tongue coating center image to identify abnormal tongue coating images and normal tongue coating images; Color analysis is performed on images of abnormal tongue coating to determine the color of the abnormal tongue coating.

5. The image-based tongue color and coating color recognition method according to claim 1, characterized in that, After identifying and outputting the color information of the tongue coating, the following is also included: Tongue coating image analysis to determine tongue coating thickness; Comparative analysis is used to determine whether the thickness of the tongue coating is greater than the preset baseline thickness; If the value is greater than the preset cleaning parameters, the preset tongue cleaning device will be matched and the tongue will be cleaned. The acquired tongue coating image is then analyzed to determine the color information of the second tongue coating.

6. An image-based tongue color and coating color recognition system, characterized in that, include: The posture analysis module analyzes the image when photographing the tongue coating to determine the tongue's posture in the image. The posture adjustment module outputs a tongue posture adjustment prompt when the tongue posture is inconsistent with the preset standard posture. After adjusting the tongue posture, it takes a picture of the tongue coating under the standard posture. The tongue coating recognition module performs image region segmentation based on the tongue coating image and extracts refined tongue coating features from the segmented image; The refined tongue coating features are input into a preset tongue coating color recognition model for analysis, in order to identify and output the tongue coating color information; A memory for storing a program for an image-based tongue color and tongue coating color recognition method as described in any one of claims 1 to 5; The processor and the program in the memory can be loaded and executed by the processor to implement an image-based tongue color and coating color recognition method as described in any one of claims 1 to 5.

7. A smart terminal, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as any one of the image-based tongue color and coating color recognition methods as claimed in claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer program is stored that can be loaded by a processor and executed as any one of the image-based tongue color and coating color recognition methods as described in claims 1 to 5.

Images