An identification method based on a mathematical model for regional division of the tongue surface contour ring in tongue diagnosis

Abstract

The present invention discloses an identification method based on a mathematical model for dividing the tongue surface contour ring region for tongue diagnosis. S1. Tongue image acquisition: During the process of obtaining the tongue image, the user logs in to the corresponding tongue diagnosis identification software, and takes a picture of their own tongue through the image taking function in the software. The present invention relates to the technical field of tongue surface recognition. This identification method based on the mathematical model for dividing the tongue surface contour ring region for tongue diagnosis enables clearer subsequent feature extraction and recognition through automated processing of the image during the acquisition operation. At the same time, after dividing the position of the tongue image and determining and classifying the corresponding regions, through this digital partitioning calculation method, the tongue is divided into several corresponding regions, thereby obtaining the tongue color, petechiae, cracks, texture, fatness and thinness, etc. values of the corresponding regions, and analyzing the health status of the corresponding body parts, making computerized tongue diagnosis a reality.

Description

Technical Field

[0001] The present invention relates to the technical field of tongue surface recognition, and in particular to a recognition method based on a mathematical model of tongue surface contour ring area division for tongue diagnosis. Background Art

[0002] The tongue is the sprout of the heart and the external sign of the spleen. The coating is produced by the stomach qi. The internal organs are connected to the tongue through the meridians, which is an important external system for Chinese medicine to diagnose diseases. The collaterals of the hand Shaoyin are at the root of the tongue, the foot Shaoyin meridian holds the root of the tongue, the foot Jueyin meridian is at the root of the tongue, and the foot Taiyin meridian is connected to the root of the tongue and spreads under the tongue. Therefore, organ diseases can be reflected in the tongue quality and tongue coating. Tongue diagnosis mainly examines the shape, color, moisture and dryness of the tongue quality and tongue coating to judge the nature of the disease, the severity of the disease, the prosperity and decline of qi and blood, the surplus and deficiency of body fluids, and the emptiness and fullness of internal organs. However, in the current tongue diagnosis process, on the one hand, human judgment is relatively subjective and objective, and there are thousands of faces for thousands of people. On the other hand, digitization cannot achieve precise zoning, so that digitization of tongue diagnosis cannot be truly accurate and intelligent.

[0003] In the current life, people often cannot understand their own situation and do not know how to observe the condition of their tongue surface. Therefore, the present invention provides a recognition method based on a mathematical model of tongue surface contour ring area division for tongue diagnosis. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides an identification method based on a mathematical model of tongue surface contour ring area division for tongue diagnosis, which solves the existing problem that people often cannot understand their own conditions in the course of life and do not know how to observe the conditions of their tongue surfaces.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a recognition method based on a mathematical model of tongue surface contour ring area division for tongue diagnosis, specifically comprising the following operation steps:

[0006] S1. Tongue image acquisition: During the process of acquiring tongue images, users log in to the corresponding tongue diagnosis and recognition software, take pictures of their own tongues through the image in the software, and the taken images are used after preprocessing;

[0007] S2. Region division: A model is built for the processed image. A rectangle is made by taking the length and width of the outermost area of ​​the tongue in the image, and the tongue area inside the rectangle is divided into 10 closed ring areas. The 10 ring areas are divided by the diagonal lines of the rectangle.

[0008] S3. Region recognition: For the specific situations where corresponding regions correspond to different organ positions, through the established digital model, during the recognition process, the feature vectors extracted from the image are used as a query, calculate the distances between it and each ring, and classify the query vector into the tongue image category represented by the ring closest to it according to the nearest neighbor principle.

[0009] Preferably, in S1, the preprocessing is to perform denoising and color correction on these images after obtaining the tongue images. The method of using a two-way filtering algorithm for denoising and referring to a standard color card for color correction is adopted. Subsequently, the tongue images need to be segmented to extract the tongue part in the images for subsequent feature extraction and recognition and other processes.

[0010] Preferably, in S2, taking the contour shape of the tongue as a ring, it is divided into 10 rings. Among them, all of the 1st, 2nd, 3rd, and 4th rings are taken in the middle of the tongue, the lower slices of the 7th, 8th, 9th, and 10th rings are taken at the tip of the tongue, the left and right slices of the 8th and 9th rings are taken on the sides of the tongue, and the upper slices of the 1st, 2nd, 3rd, 4th, 5th, 6th, 7th, and 8th rings are taken at the root of the tongue. And the selected regions correspond to the health conditions of various body parts of the human body.

[0011] Preferably, in the nearest neighbor principle of S3, the query vector is composed of features extracted from the segmented tongue image and is a point in the feature space. By calculating the defined distance from the point to the ring region, the corresponding tongue image category at that place is identified.

[0012] Preferably, the distance measurement formula involved in the nearest neighbor principle of S3 is: , where: is the point to be classified, and the ring region M = {C1, C2,..., Ci}.

[0013] Preferably, the final tongue image category in the distance measurement formula is: .

[0014] Beneficial effects

[0015] The present invention provides a recognition method based on a mathematical model for dividing the tongue surface contour ring region for tongue diagnosis. Compared with the prior art, it has the following beneficial effects:

[0016] The recognition method based on the mathematical model for dividing the tongue surface contour ring region in tongue diagnosis. During the process of S1, tongue image acquisition: when obtaining the tongue image, the user logs in to the corresponding tongue diagnosis recognition software, and takes pictures of their own tongue through the image taking function in the software. At the same time, the taken image is used after preprocessing. In S2, region division: a model is established for the processed picture. The length and width of the outermost region of the tongue part in the picture are used to form a rectangle, and the tongue region inside the rectangle is divided into 10 closed circular regions. And the 10-ring region is divided by the diagonal of the rectangle. In S3, region recognition: corresponding to the specific situation of different organ positions in the corresponding regions, through the established digital model, during the recognition process, the feature vector extracted from the image is used as a query, and the distance between it and each circular ring is calculated. According to the nearest neighbor principle, the query vector is classified into the tongue image category represented by the circular ring closest to it. By setting up the mathematical model for dividing the tongue surface contour ring region, through the automatic processing of the image during the acquisition operation process, the subsequent feature extraction and recognition are made clearer. At the same time, after dividing the position of the tongue image and determining and classifying the corresponding regions, through this digital partitioning calculation method, the tongue is divided into corresponding regions, thereby obtaining the tongue color, petechiae, cracks, texture, fatness and thinness, etc. values of the corresponding regions, and analyzing the health status of the corresponding body parts, making tongue diagnosis computerized a reality. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is the tongue image contour division diagram of the present invention;

[0018] Figure 2 is the operation flow chart of the recognition method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0020] Please refer to Figure 1-2 , the present invention provides a technical solution: a recognition method based on the mathematical model for dividing the tongue surface contour ring region in tongue diagnosis, specifically including the following operation steps:

[0021] S1. Tongue image acquisition: During the process of obtaining the tongue image, the user logs in to the corresponding tongue diagnosis recognition software, and takes pictures of their own tongue through the image taking function in the software. At the same time, the taken image is used after preprocessing;

[0022] S2, Region Division: Establish a model for the processed image. Take the length and width of the outermost region of the tongue part in the image to form a rectangle, and divide the tongue region inside the rectangle into 10 closed circular regions. Then divide the 10-ring region through the diagonal of the rectangle;

[0023] S3, Region Recognition: For the specific situations corresponding to different organ positions in the corresponding regions, through the established digital model, during the recognition process, the feature vector extracted from the image is used as a query, calculate the distance between it and each ring, and classify the query vector into the tongue image category represented between the rings closest to it according to the nearest neighbor principle.

[0024] In the embodiment of the present invention, in S1, the preprocessing is to perform denoising and color correction on these images after obtaining the tongue images. The method of using a two-way filtering algorithm for denoising and referring to a standard color card for color correction is adopted. Subsequently, the tongue images need to be segmented to extract the tongue part in the images for subsequent feature extraction and recognition and other processes;

[0025] In the embodiment of the present invention, in S2, taking the contour shape of the tongue as a ring, it is divided into 10 rings. Among them, all of the 1st, 2nd, 3rd, and 4th rings are taken from the middle of the tongue, the lower part of the 7th, 8th, 9th, and 10th rings is taken from the tip of the tongue, the left and right parts of the 8th and 9th rings are taken from the sides of the tongue, and the upper part of the 1st, 2nd, 3rd, 4th, 5th, 6th, 7th, and 8th rings is taken from the root of the tongue. And the selected regions correspond to the health conditions of various parts of the human body.

[0026] In the embodiment of the present invention, in S3, in the nearest neighbor principle, the query vector is composed of features extracted from the segmented tongue image and is a point in the feature space. By calculating the distance from the defined point to the circular region, the tongue image category corresponding to that place is identified.

[0027] In the embodiment of the present invention, the distance measurement formula involved in the nearest neighbor principle in S3 is: , where: is the point to be classified, and the circular region M = {C 1 , C 2 , …, C i}.

[0028] In the embodiment of the present invention, the final tongue image category in the distance measurement formula is: .

[0029] In summary, by setting up a mathematical model for the division of the tongue surface contour ring area, through the automatic processing of images during the acquisition operation, the subsequent feature extraction and recognition become clearer. At the same time, after dividing the position of the tongue image and determining and classifying the corresponding areas, through this digital zoning calculation method, the tongue is divided into several corresponding areas, so as to obtain the tongue color, petechiae, cracks, texture, thickness, etc. values of the corresponding areas, and analyze the health status of the corresponding body parts, making tongue diagnosis computerized a reality.

[0030] At the same time, the content not described in detail in this specification belongs to the prior art well known to those skilled in the art.

[0031] It should be noted that in this text, relational terms such as "first" and "second" 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 variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0032] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An identification method based on a mathematical model for regional division of the tongue surface contour ring in tongue diagnosis, characterized in that: Specifically, it includes the following operation steps: S1. Tongue image acquisition: During the process of obtaining the tongue image, the user logs in to the corresponding tongue diagnosis recognition software, and takes a picture of their own tongue through the image taking function in the software. At the same time, the captured image is used after preprocessing; S2. Regional division: Establish a model for the processed picture. Take the length and width of the outermost region of the tongue part in the picture to form a rectangle, and divide the tongue region inside the rectangle into 10 closed circular regions. And divide the 10-ring region through the diagonal of the rectangle; S3. Regional identification: For the specific situation where the corresponding region corresponds to different organ positions, through the established digital model, during the identification process, the feature vector extracted from the image is used as a query, calculate its distance from each circular ring, and classify the query vector into the tongue image category represented by the circular ring closest to it according to the nearest neighbor principle; In S2, taking the contour shape of the tongue as a ring, it is divided into 10 rings. Among them, all of the 1st, 2nd, 3rd, and 4th rings are taken in the middle of the tongue, the lower part of the 7th, 8th, 9th, and 10th rings are taken at the tip of the tongue, the left and right parts of the 8th and 9th rings are taken on the sides of the tongue, and the upper part of the 1st, 2nd, 3rd, 4th, 5th, 6th, 7th, and 8th rings are taken at the root of the tongue. And the selected regions correspond to the health conditions of various parts of the human body.

2. The identification method based on a mathematical model for regional division of the tongue surface contour ring in tongue diagnosis according to claim 1, characterized in that: In S1, the preprocessing is to perform denoising and color correction on these images after obtaining the tongue images. The method of using a two-way filtering algorithm for denoising and referring to a standard color card for color correction is adopted. Subsequently, the tongue images need to be segmented to extract the tongue part in the images for subsequent feature extraction and identification and other processes.

3. The identification method based on a mathematical model for regional division of the tongue surface contour ring in tongue diagnosis according to claim 1, characterized in that: In S3, in the nearest neighbor principle, the query vector is composed of features extracted from the segmented tongue image, which is a point in the feature space. By calculating the defined distance from the point to the circular ring region, the corresponding tongue image category is identified.

4. The identification method based on a mathematical model for regional division of the tongue surface contour ring in tongue diagnosis according to claim 1, characterized in that: The distance measurement formula involved in the nearest neighbor principle in S3 is as follows: , where: is the point to be classified, and the circular region M = {C 1 , C 2 , …, C i}.

5. The identification method based on a mathematical model for regional division of the tongue surface contour ring in tongue diagnosis according to claim 4, characterized in that: The final tongue image category in the distance measurement formula is: .

Images