Search

CN-116897012-B - Method and device for identifying physique of traditional Chinese medicine, electronic equipment, storage medium and program

CN116897012BCN 116897012 BCN116897012 BCN 116897012BCN-116897012-B

Abstract

The present disclosure provides a method, an apparatus, an electronic device, a storage medium, and a program for identifying physique of traditional Chinese medicine, which belong to the technical field of computers. The method comprises the steps of obtaining an infrared human body image of a target user, inputting the infrared human body image into an infrared key point identification model to obtain traditional Chinese medicine human body key points in the infrared human body image, determining the temperature type of the traditional Chinese medicine human body key points in the infrared human body image according to the temperature distribution condition in the infrared human body image, and identifying the traditional Chinese medicine physique of the target user according to the temperature type.

Inventors

  • WANG SIFAN

Assignees

  • 京东方科技集团股份有限公司

Dates

Publication Date
20260512
Application Date
20210726

Claims (13)

  1. 1. A method for identifying physique of traditional Chinese medicine, which is characterized by comprising the following steps: acquiring an infrared human body image of a target user; inputting the infrared human body image into an infrared key point identification model to obtain traditional Chinese medicine human body key points in the infrared human body image; Determining the temperature type of the key points of the traditional Chinese medicine human body in the infrared human body image according to the temperature distribution condition in the infrared human body image; Identifying the traditional Chinese medicine physique of the target user according to the temperature type; The infrared key point identification model is obtained through the following steps: acquiring a natural light key point identification model and a sample infrared human body image; marking key points of traditional Chinese medicine human bodies in the sample infrared human body image; and performing transfer learning on the natural light key point identification model by using the marked sample infrared human body image to obtain an infrared key point identification model.
  2. 2. The method according to claim 1, wherein the performing the migration learning on the natural light keypoint identification model by using the noted sample infrared human body image to obtain an infrared keypoint identification model includes: Adjusting all-connection-layer nodes of the natural light key point identification model based on the traditional Chinese medicine human body key points; And under the condition of keeping the model parameters of the feature extraction layer unchanged, training the natural light key point identification model after node adjustment by using the noted sample infrared human body image to obtain an infrared key point identification model.
  3. 3. The method according to claim 2, wherein the original human body keypoints of the natural light keypoint identification model comprise the traditional Chinese medicine human body keypoints, and the number of the original human body keypoints is greater than the number of the traditional Chinese medicine human body keypoints; the adjusting the full-connection layer node of the natural light key point identification model based on the traditional Chinese medicine human body key points comprises the following steps: and deleting human body key points corresponding to node labels from all-connection-layer nodes of the natural light key point identification model, wherein the human body key points are not all-connection-layer nodes of the traditional Chinese medicine human body key points.
  4. 4. The method according to claim 1, wherein the performing the migration learning on the natural light keypoint identification model by using the noted sample infrared human body image to obtain an infrared keypoint identification model includes: Adjusting all-connection-layer nodes of the natural light key point identification model based on the traditional Chinese medicine human body key points; And training the natural light key point recognition model after node adjustment by using the noted sample infrared human body image to obtain an infrared key point recognition model.
  5. 5. The method of claim 1, wherein the feature extraction network in the natural light keypoint identification model is a lightweight feature extraction network having a smaller amount of parameter data than an original feature extraction network of the natural light keypoint identification model.
  6. 6. The method of claim 1, wherein the Chinese medical body keypoints comprise at least one of a head keypoint, a neck keypoint, a shoulder keypoint, an elbow keypoint, a hand keypoint, an abdomen keypoint, a crotch keypoint, a leg elbow keypoint, a foot keypoint.
  7. 7. The method of claim 1, wherein the temperature type comprises at least one of a high temperature type and a low temperature type; the determining the temperature type of the key points of the traditional Chinese medicine human body in the infrared human body image according to the temperature distribution condition in the infrared human body image comprises the following steps: Determining a first number of temperature maximum positions and/or a second number of temperature minimum positions in the infrared human body image; And determining the temperature type of the key points of the traditional Chinese medicine human body corresponding to the image area where the temperature maximum value position is located as high Wen Leixing, and/or determining the temperature type of the key points of the traditional Chinese medicine human body corresponding to the image area where the temperature minimum value position is located as low-temperature type.
  8. 8. The method of claim 7, wherein the determining a first number of temperature maxima locations and/or a second number of temperature minima locations in the infrared human image comprises: acquiring image positions of a first number of temperature maxima in the infrared human body image through a maximum filter, wherein the image positions are used as temperature maxima; Taking the average value of the maximum value and the minimum value of the temperature in the infrared human body image as a turnover plane, and reversing the temperature value in the infrared human body image; And acquiring positions of the second number of temperature maxima in the inverted infrared human body image through the maximum filter to serve as temperature minima positions.
  9. 9. The method according to any one of claims 1-8, wherein said identifying the subject user's chinese medical constitution according to the temperature type comprises: inquiring the traditional Chinese medicine physique which is matched with the temperature type of each traditional Chinese medicine human body key point in the traditional Chinese medicine physique mapping relation.
  10. 10. A device for identifying physical constitution of traditional Chinese medicine, characterized in that the device comprises: The receiving module is configured to acquire an infrared human body image of a target user; the model prediction module is configured to input the infrared human body image into an infrared key point identification model to obtain traditional Chinese medicine human body key points in the infrared human body image; The identification module is configured to determine the temperature type of the key points of the traditional Chinese medicine human body in the infrared human body image according to the temperature distribution condition in the infrared human body image; And identifying the traditional Chinese medicine physique of the target user according to the temperature type.
  11. 11. A computing processing device, comprising: a memory having computer readable code stored therein; One or more processors, the computing processing device performing the method of recognition of a chinese medical physique according to any one of claims 1-8 when the computer readable code is executed by the one or more processors.
  12. 12. A computer program comprising computer readable code which, when run on a computing processing device, causes the computing processing device to perform the method of recognition of a chinese medical physique according to any one of claims 1 to 9.
  13. 13. A computer-readable medium in which a computer program of the method for identifying physical constitution of chinese medical science according to any one of claims 1 to 9 is stored.

Description

Method and device for identifying physique of traditional Chinese medicine, electronic equipment, storage medium and program Technical Field The disclosure belongs to the technical field of computers, and in particular relates to a method and a device for identifying physique of traditional Chinese medicine, electronic equipment, a storage medium and a program. Background Infrared imaging is used as an extension of the hope in the looking and smelling inquiry of traditional Chinese medicine, traditional Chinese medicine infrared image analysis is mostly that traditional Chinese medicine doctor processes infrared human body images manually by using color chromatography after training, and then analyzes cold and heat symptoms of body parts by watching the processed infrared human body images, and traditional Chinese medicine physique classification such as cold depression, heat stasis and the like is obtained according to comprehensive analysis of cold and heat symptoms of a plurality of key parts such as hands, feet, heads, necks, abdomen, stomach and the like, so that traditional Chinese medicine doctors who are not trained or have shallow experience cannot accurately judge the symptoms of a patient. Disclosure of Invention The present disclosure provides a method, an apparatus, an electronic device, a storage medium, and a program for identifying physique of traditional Chinese medicine. Some embodiments of the present disclosure provide a method for identifying physique of traditional Chinese medicine, the method comprising: acquiring an infrared human body image of a target user; inputting the infrared human body image into an infrared key point identification model to obtain traditional Chinese medicine human body key points in the infrared human body image; Determining the temperature type of the key points of the traditional Chinese medicine human body in the infrared human body image according to the temperature distribution condition in the infrared human body image; And identifying the traditional Chinese medicine physique of the target user according to the temperature type. Optionally, the infrared keypoint identification model is obtained by: acquiring a natural light key point identification model and a sample infrared human body image; marking key points of traditional Chinese medicine human bodies in the sample infrared human body image; and performing transfer learning on the natural light key point identification model by using the marked sample infrared human body image to obtain an infrared key point identification model. Optionally, the performing transfer learning on the natural light key point recognition model by using the noted sample infrared human body image to obtain an infrared key point recognition model includes: Adjusting all-connection-layer nodes of the natural light key point identification model based on the traditional Chinese medicine human body key points; And under the condition of keeping the model parameters of the feature extraction layer unchanged, training the natural light key point identification model after node adjustment by using the noted sample infrared human body image to obtain an infrared key point identification model. Optionally, the original human body key points of the natural light key point identification model comprise the traditional Chinese medicine human body key points, and the number of the original human body key points is larger than that of the traditional Chinese medicine human body key points; the adjusting the full-connection layer node of the natural light key point identification model based on the traditional Chinese medicine human body key points comprises the following steps: and deleting human body key points corresponding to node labels from all-connection-layer nodes of the natural light key point identification model, wherein the human body key points are not all-connection-layer nodes of the traditional Chinese medicine human body key points. Optionally, the performing transfer learning on the natural light key point recognition model by using the noted sample infrared human body image to obtain an infrared key point recognition model includes: Adjusting all-connection-layer nodes of the natural light key point identification model based on the traditional Chinese medicine human body key points; And training the natural light key point recognition model after node adjustment by using the noted sample infrared human body image to obtain an infrared key point recognition model. Optionally, the feature extraction network in the natural light key point recognition model is a lightweight feature extraction network, and the parameter data size of the lightweight feature extraction network is smaller than the parameter data size of the original feature extraction network of the natural light key point recognition model. Optionally, the Chinese medicine human body key points at least comprise at least one of head key points, neck key