CN-120392525-B - Method for stably identifying back acupoints of human body and intelligent terminal

CN120392525BCN 120392525 BCN120392525 BCN 120392525BCN-120392525-B

Abstract

The invention discloses a method and an intelligent terminal for stably identifying back acupoints of a human body, wherein the method for stably identifying the back acupoints of the human body comprises the steps of obtaining a back image of a current user, obtaining positioning coordinates of reference acupoints, calculating human body proportions according to the reference acupoints, dividing the reference acupoints into a plurality of groups, determining a reference straight line according to representative points in each group, obtaining a spine trend identification line, and calculating specific positions of other acupoints according to the human body proportions and the spine trend identification line. According to the invention, the back image of the user is acquired, the reference acupoint coordinates are marked, the representative points of a plurality of groups divided by the reference acupoints are calculated in a grouping manner to determine the reference straight lines of the plurality of groups, so that the spine trend identification line is generated, and the positions of other acupoints are determined by combining the human body proportion and the spine trend identification line. The invention comprehensively considers the trend of the spine and the proportion of the human body, realizes accurate acupoint marking, improves the recognition efficiency, and ensures the accuracy of the positioning result and the physiological rationality.

Inventors

Mo Jixue
ZHANG QI
LI XIAOHUI

Assignees

广东具身风暴机器人有限公司

Dates

Publication Date: 20260508
Application Date: 20250427

Claims (6)

1. A method for stably identifying points on the back of a human body, the method comprising: acquiring a back image of a current user, acquiring positioning coordinates of a reference acupoint according to the back image, and calculating the human body proportion of the current user according to the positioning coordinates of the reference acupoint; Dividing the reference acupoints into a plurality of groups comprising the reference acupoints which are bilaterally symmetrical, and calculating representative points on the left side and the right side in each group according to the positioning coordinates; Acquiring a spine trend identification line of the current user according to the reference straight line; calculating the specific positions of other acupuncture points according to the human body proportion and the spine trend identification line; the calculating representative points on the left and right sides in each group according to the positioning coordinates comprises: calculating the representative points in a plurality of groups respectively And ; The representative point Is the mean point of a plurality of acupoints on the left side in the group; The representative point Is the mean point of a plurality of acupoints on the right side in the group; The determining a reference straight line according to the representative point includes: According to the representative points in a plurality of the groups And Calculating reference straight lines of corresponding groups Slope of (2) And intercept of According to the slope And intercept of Determining the reference straight line ; The slope of The calculation formula of (2) is as follows: ; The intercept of The calculation formula of (2) is as follows: ; the step of obtaining the spine trend identification line of the current user according to the reference straight line comprises the following steps: According to a plurality of said reference lines Calculating the slope of the spinal trend marker line According to the slope At the reference straight line Generating the spine trend identification line; slope of the spine trend marker line The calculation formula of (2) is as follows: ; the calculating the specific positions of other acupuncture points according to the human body proportion and the spine trend identification line comprises the following steps: calculating basic dimensions According to the basic scale Determining specific positions of other acupoints ; The basic dimensions Is the highest point in the group And the lowest point The distance between them; The basic dimensions The calculation formula of (2) is as follows: ; The specific positions of other acupoints The calculation formula of (2) is as follows: ; ; t is a scaling factor determined according to the traditional bone mineral scale theory.
2. The method of claim 1, wherein the acquiring the back image of the current user and acquiring the positioning coordinates of the reference acupoint according to the back image comprises: processing the back image of the current user by using a training model; The method for processing the back image of the current user by using the training model further comprises the following steps: And constructing a training model, wherein the training model comprises a human back acupoint data set, and the human back acupoint data set comprises a plurality of human back images with coordinate data of the reference acupoints.
3. The method of claim 2, wherein the human back acupoint dataset further comprises enhancement data; The enhancement data is an image obtained by randomly rotating, scaling and brightness processing a plurality of pairs of human back images.
4. The method of claim 1, wherein the dividing the reference acupoints into groups including the reference acupoints that are bilaterally symmetrical comprises: grouping according to the areas of the reference acupoints in the back image; The group includes a shoulder region group, a back region group, and a waist region group; The shoulder region group comprises 8 acupoints of Jianzhu, bingfeng and Tianzong which are bilaterally symmetrical; the back middle region group comprises 6 acupoints of Geshu, mianmen and Sanjiaoshu which are bilaterally symmetrical; The waist region group comprises 12 acupoints of Guanyuanshu, navigator, bai Huan, bian Ji, huiyang and Jinggan left-right symmetry.
5. The method of claim 4, wherein the dividing the reference acupoints into groups including the reference acupoints that are bilaterally symmetrical further comprises: or, grouping according to the meridian connection relation of the reference acupoints in the back image; or, grouping according to the functional attribute of the reference acupoint in the back image; or, grouping according to the recognition confidence of the reference acupuncture points in the back image.
6. An intelligent terminal, characterized in that the intelligent terminal comprises a memory, a processor and a program which is stored in the memory and can run on the processor and is used for stably identifying the back acupuncture points of a human body, and the method for stably identifying the back acupuncture points of the human body is realized when the processor executes the program for stably identifying the back acupuncture points of the human body according to any one of the claims 1-5; The intelligent terminal further comprises a display, wherein the display can simultaneously display at least two display pages, one display page is used for displaying the back image of the current user, and the other display page is used for displaying the back image of the current user after the reference acupuncture points and other acupuncture points are marked.

Description

Method for stably identifying back acupoints of human body and intelligent terminal Technical Field The invention relates to the technical field of robot vision recognition, in particular to a method for stably recognizing back acupoints of a human body and an intelligent terminal. Background Along with the continuous development of the traditional Chinese medicine diagnosis and treatment technology, the accuracy and the efficiency of acupoint positioning are increasingly valued by the medical community. Particularly in the traditional treatment methods such as acupuncture, massage, moxibustion and the like, the accurate positioning of back acupoints directly relates to the treatment effect and the patient experience. The back is an important therapeutic area of the human body, contains up to 95 key points, and the points are distributed on important meridians such as the governor vessel and the bladder channel, are closely related to viscera functions and play an important role in disease prevention and treatment. Accurate acupoint positioning can not only improve treatment effect, but also reduce medical risks such as false needle and false moxibustion, and has important significance for traditional Chinese medicine teaching, clinical research and standardized construction. In the prior art, the positioning of back acupoints mainly depends on doctors to measure through an osseous size division method, the method has the problems of strong subjectivity, low operation efficiency, lack of standardization and the like, when a plurality of acupoints are required to be positioned, the positioning results among different doctors are different, although the prior art also has a method for identifying the acupoints through machine vision, such as a meridian acupoint massage method based on deep learning disclosed in CN115429683A, the prior mechanical identification cannot position the acupoints in a targeted manner according to the physical conditions of different users, and the problem of inaccurate acupoint positioning exists. Accordingly, the prior art is still in need of improvement and development. Disclosure of Invention In order to solve the problems that in the prior art, when a plurality of acupuncture points on the back of a user are positioned, the prior art relies on subjective measurement of doctors, the efficiency is low, standardization is lacking, and the difference of positioning results among different doctors is large, but the acupuncture points can not be positioned pertinently according to the figure conditions of different users through mechanical vision identification, and the problem that the positioning of the acupuncture points is inaccurate is solved, the invention provides a method for stably identifying the acupuncture points on the back of a human body and an intelligent terminal. The invention is realized by the following technical scheme: A method for stably identifying points on the back of a human body, wherein the method for stably identifying points on the back of the human body comprises the following steps: acquiring a back image of a current user, acquiring positioning coordinates of a reference acupoint according to the back image, and calculating the human body proportion of the current user according to the positioning coordinates of the reference acupoint; Dividing the reference acupoints into a plurality of groups comprising the reference acupoints which are bilaterally symmetrical, and calculating representative points on the left side and the right side in each group according to the positioning coordinates; Acquiring a spine trend identification line of the current user according to the reference straight line; and calculating the specific positions of other acupuncture points according to the human body proportion and the spine trend identification line. The method for stably identifying the acupoints on the back of the human body, wherein the calculating the representative points on the left and right sides in each group according to the positioning coordinates comprises: calculating the representative points P l(xl,yl) and P r(xr,yr) in several of the groups, respectively; The representative point P l(xl,yl) is the average point of a plurality of acupoints on the left side in the group; the representative point P r(xr,yr) is the mean point of a plurality of acupoints on the right side of the group. The method for stably identifying the back acupoints of the human body, wherein the determining the reference straight line according to the representative points comprises the following steps: Calculating the slope k and the intercept b of a reference straight line L 1 of a corresponding group according to the representative points P l(xl,yl) and P r(xr,yr in a plurality of groups, and determining the reference straight line L 1 according to the slope k and the intercept b; The calculation formula of the slope k is k= (y r-yl)/(xr-xl); The calculation formula of the int