Semi-supervised intelligent acupoint selection method and system for head acupuncture based on visual positioning

Abstract

The present invention discloses a semi-supervised intelligent acupoint selection method and system for head acupuncture based on visual positioning, which belongs to the field of intelligent assisted medical technology. The method includes collecting facial images and extracting standard two-dimensional facial feature points; using the three-dimensional spatial coordinates of the standard two-dimensional facial feature points and the three-dimensional spatial coordinates of the head acupoint feature points to construct a first training data set and train a multi-layer perceptron mapping network; using a second training data set constructed with virtual facial images and the two-dimensional coordinates of the head acupoint feature points to train a Yolo two-dimensional mapping network; inputting unlabeled facial images into the trained multi-layer perceptron mapping network and the trained Yolo network, weightedly fusing the two-dimensional coordinates output by the two networks, and completing intelligent assisted acupoint selection for head acupuncture. The present invention realizes semi-supervised precise recognition and positioning of facial acupoints, thereby improving the accuracy and safety of patients' self-help acupuncture treatment under the premise of relying on a small amount of labeled data.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent auxiliary medical treatment, and particularly relates to a head acupuncture semi-supervised intelligent point selection method and system based on visual positioning. BACKGROUND

[0002] Traditional acupuncture treatment needs to go to the hospital for registration and treatment, and patients need to pay a lot of time cost. Therefore, self-help auxiliary acupuncture is a method with strong clinical application value. However, for head acupuncture, due to the dense facial acupoints and complex surrounding tissue structure, the accuracy and safety of point selection are extremely high, and the acupoints located in the scalp are more difficult to obtain, which brings difficulties to self-help auxiliary point selection of head acupuncture.

[0003] Existing acupuncture point selection auxiliary tools and methods, such as acupuncture atlas and body surface marker positioning method, have certain subjectivity and uncertainty. For example, although the acupuncture atlas provides the approximate position of the acupoint, due to individual differences of patients and changes in facial morphology, actual point selection still needs to rely on the experience and touch of doctors for fine adjustment; the body surface marker positioning method is easily affected by factors such as expression changes and skin relaxation, resulting in inaccurate point selection. On the other hand, existing artificial intelligence diagnosis and treatment technology often relies on a large amount of labeled data, and the human resources of doctors with professional knowledge are limited, and it is difficult to provide a sufficient number of labeled samples in the traditional sense.

[0004] Therefore, it is particularly important to develop a method that can accurately and quickly locate facial acupuncture points with a small amount of labeled data. With the rapid development of computer vision and deep learning technology, these technologies provide a new solution for acupuncture point selection, which can realize semi-supervised accurate recognition and positioning of facial acupoints, thereby improving the accuracy and safety of patient self-acupuncture treatment under the premise of relying on a small amount of labeled data. SUMMARY

[0005] In order to solve the problem of head self-help auxiliary acupuncture point selection positioning, the application proposes a head acupuncture semi-supervised intelligent point selection method and system based on visual positioning, which uses neural network and computer vision technology to intelligently assist point selection through acupuncture point three-dimensional coordinate inference and two-dimensional coordinate mapping, and supports intelligent auxiliary treatment methods such as acupuncture treatment for retinal optic nerve injury.

[0006] To achieve the above purpose, the technical scheme adopted by the application is as follows:

[0007] The head acupuncture semi-supervised intelligent point selection method based on visual positioning comprises the following steps:

[0008] Step S1, in the training stage of the application, first recognize 68 facial standard feature points through the input face image using the Dlib facial standard feature point recognition library . Use 3DMM model to model the three-dimensional space coordinates of standard facial feature points , as shown below:

[0009] ,

[0010] in, is the average position of the standard two-dimensional facial feature points, For the face shape base, is the facial shape feature coefficient, For facial expression base, is the facial expression feature coefficient, m and n represent the total number of facial shape bases and facial expression bases respectively.

[0011] Construct the reprojection error objective function, the objective function is as follows:

[0012] ,

[0013] Where, is the standard projection matrix, is the face pose and rotation matrix relative to the camera, is the face's translation vector relative to the camera's pose. By minimizing the above equation, we can solve 、 、 、 , thereby obtaining the face's position relative to the camera, as well as the 3D coordinates of the standard facial feature points and the characteristic coefficients of the patient's 3DMM model. The obtained 3DMM characteristic coefficients are then used to obtain the 3D point cloud of the face.

[0014] Step S2: A professional doctor selects acupuncture point feature points on the 3D point cloud of the patient's head obtained in step S1. ,in is the number of acupuncture points. Construct the first training data set for the element, and train the multilayer perceptron mapping network with the following objective function ;

[0015] ,

[0016] The trained multilayer perceptron mapping network Input the 3D coordinates of the standard facial feature points and output the 3D coordinates of the acupuncture points.

[0017] In step S3, the 3D point cloud of the patient's head and the 3D coordinates of the acupuncture points obtained in step S1 are input into the Unreal Engine virtual engine. Multiple virtual cameras are evenly distributed around the common acquisition locations of the patient's head point cloud. The virtual engine automatically generates a virtual face image and the corresponding 2D coordinates of the acupuncture points. This expands the second training set by using a small amount of data annotated by professional doctors. The Yolo network is trained with this expanded second training set, so that the input face image outputs the 2D coordinates of the acupuncture points.

[0018] Step S4, in the inference test phase, for general patients who are not labeled. Use a camera to capture the patient's facial image, and similar to step S1, use the Dlib library to identify 68 standard feature points, and use the 3DMM model to obtain the three-dimensional coordinates of the patient's 68 standard feature points and the camera pose. Then, the two-branch inference network is used to infer the two-dimensional coordinates of the patient's acupuncture points. Branch 1: Input the three-dimensional coordinates of the 68 standard feature points into the multi-layer perceptron trained in step S2. Generate the 3D coordinates of acupuncture feature points, and then use the camera pose information to calculate the 2D coordinates of the supervised branch acupuncture points: ; Branch 2: Input the patient image into the trained Yolo structure network branch of step S3, and output the two-dimensional reference coordinates of the acupuncture points of the semi-supervised branch: Finally, the acupuncture point reference coordinates generated by the two branches are weighted according to the following formula to generate the final acupuncture point two-dimensional reasoning coordinates.

[0019] ,

[0020] in, 、 is a hyperparameter, determined through parameter adjustment experiments. The final two-dimensional coordinates of the acupuncture points.

[0021] In another aspect, the present invention provides a semi-supervised intelligent acupoint selection system for head acupuncture based on visual positioning, the system comprising:

[0022] The 3DMM model solving module is used to collect facial images and extract standard two-dimensional facial feature points. It uses the 3DMM model to solve the camera pose of the facial image, the three-dimensional spatial coordinates of the standard two-dimensional facial feature points, and the three-dimensional point cloud of the face.

[0023] a multilayer perceptron mapping network training module for selecting the locations of acupuncture point feature points on the three-dimensional point cloud of the face, constructing a first training data set using the three-dimensional spatial coordinates of the standard two-dimensional facial feature points and the three-dimensional spatial coordinates of the acupuncture point feature points, and training a multilayer perceptron mapping network from the three-dimensional spatial coordinates of the standard two-dimensional facial feature points to the three-dimensional spatial coordinates of the acupuncture point feature points;

[0024] a two-dimensional mapping network training module, configured to input the three-dimensional spatial coordinates of the facial three-dimensional point cloud and the standard two-dimensional facial feature points into a virtual engine, and evenly arrange multiple virtual cameras around the three-dimensional facial point cloud to generate virtual facial images at different distances and postures and the corresponding two-dimensional coordinates of the head acupuncture point feature points, and generate a second training data set constructed with the virtual facial images and the two-dimensional coordinates of the head acupuncture point feature points. The second training data set is used to train a two-dimensional mapping network from facial images to acupuncture point feature points based on a Yolo network;

[0025] The intelligent assisted acupoint selection module is used to input the unlabeled facial image captured by the camera into a trained multi-layer perceptron mapping network to infer the three-dimensional spatial coordinates of the acupuncture point feature points, use the inferred three-dimensional spatial coordinates of the acupuncture point feature points and the camera posture projection to calculate the coordinates of the acupuncture point feature points on the two-dimensional image, input the unlabeled facial image into the trained Yolo network to generate the two-dimensional reference coordinates of the acupuncture point feature points; and weightedly fuse the coordinates on the two-dimensional image and the two-dimensional reference coordinates to complete intelligent assisted acupuncture point selection on the head.

[0026] In a third aspect, the present invention provides an electronic device comprising: one or more processors; a memory for storing one or more programs; wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the aforementioned semi-supervised intelligent acupoint selection method for head acupuncture based on visual positioning.

[0027] In a fourth aspect, the present invention provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, enables the processor to implement the aforementioned semi-supervised intelligent acupoint selection method for head acupuncture based on visual positioning.

[0028] The beneficial effects of the present invention are:

[0029] The present invention can be widely used in the field of autonomous acupuncture treatment, providing patients with reliable self-service acupoint positioning support, improving the convenience of acupuncture treatment, and saving patients the time cost of frequent medical treatment. By utilizing the semi-supervised method of the virtual engine, the workload of doctors in labeling data can be effectively reduced, which is conducive to saving doctors' labor costs and improving the application speed of the system for different diseases. In addition, the method proposed in the present invention can locate acupoints in the visible area of ​​the face as well as invisible acupoints under the scalp, expanding the application range of this acupuncture-assisted positioning method. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a flow chart of the training phase of the semi-supervised intelligent acupoint selection method for head acupuncture based on visual positioning of the present invention;

[0031] Figure 2 This is a flow chart of the reasoning and application phase of the semi-supervised intelligent acupoint selection method for head acupuncture based on visual positioning of the present invention. DETAILED DESCRIPTION

[0032] The present invention will be further described below with reference to the accompanying drawings and examples.

[0033] A semi-supervised intelligent acupoint selection method for head acupuncture based on visual positioning is used to assist patients in self-help acupuncture treatment. The method is divided into a training phase and an inference application phase.

[0034] like Figure 1 As shown, in the training stage, multiple patient face images are collected, and 68 standard feature points are extracted from the two-dimensional images. The 3DMM model is used to solve the camera pose of the patient image, the three-dimensional spatial coordinates of the 68 standard facial feature points and the three-dimensional point cloud of the patient's head; the doctor marks the position of the acupoints on the generated three-dimensional point cloud of the patient's head, and the 68 standard facial feature points and acupuncture points of each patient form the first training data set; the data of all the labeled patients are used to train the first supervised acupoint two-dimensional projection inference branch network: the first branch network is a multilayer perceptron neural network that maps the three-dimensional coordinates of the 68 standard facial feature points to the three-dimensional coordinates of the acupuncture points. In addition, the 3D point cloud of the patient's head is input into the Unreal Engine virtual engine, and multiple virtual camera positions are set around the 3D head model to generate virtual facial images and acupoint 2D image point coordinates under each virtual camera. The virtual facial images and acupoint 2D point coordinates constitute a second extended training data set, and the second extended training data set is used to train the Yolo network, which inputs real or virtual facial images and outputs semi-supervised acupoint 2D mapping.

[0035] like Figure 2 As shown in the figure, in the inference application stage, the facial image of the unlabeled patient is collected by the camera and the 68 standard feature points of the face are detected. The first inference branch uses the 3DMM model to solve the three-dimensional spatial coordinates of the camera pose and the 68 standard feature points of the patient's collected image; the acupoint three-dimensional mapping multi-layer perceptron trained in the training stage is used to map the acupoints. The 3D coordinates of acupuncture points on unlabeled patients are inferred. The supervised 2D projections of the acupuncture points are calculated using the inferred 3D coordinates and the camera pose projection of the patient image. The second inference branch, the semi-supervised 2D acupuncture point mapping branch, takes the real patient image as input and uses the Yolo network trained during the training phase to generate the semi-supervised 2D projections of the acupuncture points. Finally, the predictions from the two branches are weighted and fused to complete intelligent assisted acupuncture point selection on the head and obtain the acupuncture point coordinates.

[0036] The method specifically comprises the following steps:

[0037] Step S1, in the training phase of the application, first identify 68 facial standard feature points by using the Dlib facial standard feature point recognition library through the input patient image . Model the three-dimensional space coordinates of the standard feature points by using the 3DMM model , as shown in the following formula:

[0038] ,

[0039] wherein, is the average position of the facial standard two-dimensional feature points, is the face shape base, is the face shape feature coefficient, is the face expression base, is the face expression feature coefficient, and m and n respectively represent the total number of the face shape base and the face expression base.

[0040] Construct a re-projection error objective function, and the objective function is as follows:

[0041] ,

[0042] wherein, is the standard projection matrix, is the face relative to the camera pose rotation matrix, is the face relative to the camera pose translation vector. By minimizing the above formula, solve , , , , so as to obtain the face relative to the camera pose, and the three-dimensional coordinates of the facial standard feature points and the feature coefficient of the patient 3DMM model. Then, the 3DMM feature coefficient obtained is used to obtain the patient head dense three-dimensional point cloud.

[0043] Step S2, select acupuncture point feature points by a professional doctor on the patient head three-dimensional point cloud obtained in step S1 wherein, is the number of acupuncture point. Take as an element to construct a first training data set, and the following formula is an objective function, and a multilayer perceptron mapping network is trained ;

[0044] ,

[0045] The trained perceptron inputs the three-dimensional coordinates of the facial standard feature points, and outputs the three-dimensional coordinates of the acupuncture point.

[0046] Step S3, inputting the three-dimensional point cloud of the patient's head obtained in step S1 and the three-dimensional point coordinates of the acupuncture points into the UnrealEngine virtual engine, setting a plurality of virtual cameras uniformly in the common collection position area around the patient's head point cloud, and automatically generating a virtual human face image and corresponding two-dimensional point coordinates of acupuncture points by the virtual engine, so as to generate an expanded second training set by using a small amount of professional doctor annotation data. The Yolo structure network is trained with the expanded second training set, so that the input human face image outputs two-dimensional acupuncture point coordinates.

[0047] Step S4, in the reasoning test phase, for a general patient without annotation, using a camera to collect a patient's face image, using the Dlib library to identify 68 standard feature points similar to step S1, and using a 3DMM model to obtain three-dimensional coordinates of the 68 standard feature points of the patient and the camera pose. Then, the two-branch reasoning network is used to infer the two-dimensional point coordinates of the acupuncture points of the patient respectively. Branch 1: inputting the three-dimensional coordinates of the 68 standard feature points into the multilayer perceptron trained in step S2 generate three-dimensional coordinates of acupuncture feature points, and then calculate the two-dimensional coordinates of acupuncture points of the supervised branch using the camera pose information: ; Branch 2: inputting the patient image into the Yolo structure network branch trained in step S3, and outputting the two-dimensional point reference coordinates of the semi-supervised branch acupuncture points: . Finally, the acupuncture point reference coordinates generated by the two branches are weighted to generate the final two-dimensional reasoning coordinates of the acupuncture points according to the following formula.

[0048] ,

[0049] wherein, 、 is a hyperparameter, which is determined by a parameter tuning experiment, is the final determined two-dimensional coordinate of the acupuncture point.

[0050] On the other hand, the application provides a head acupuncture semi-supervised intelligent point selection system based on visual positioning, which comprises various modules capable of realizing the steps of the foregoing method. Specifically, the system comprises:

[0051] a 3DMM model solving module for collecting a human face image and extracting standard two-dimensional feature points of a face, using a 3DMM model to solve the camera pose of the human face image, the three-dimensional space coordinates of the standard two-dimensional feature points of the face, and the three-dimensional point cloud of the face;

[0052] The multi-layer perceptron mapping network training module is configured to select acupuncture point feature point positions on the face three-dimensional point cloud, use three-dimensional space coordinates of the face standard two-dimensional feature points and three-dimensional space coordinates of the acupuncture point feature points to construct a first training data set, and train a multi-layer perceptron mapping network from the three-dimensional space coordinates of the face standard two-dimensional feature points to the three-dimensional space coordinates of the acupuncture point feature points.

[0053] The two-dimensional mapping network training module is configured to input the face three-dimensional point cloud and the three-dimensional space coordinates of the face standard two-dimensional feature points into a virtual engine, set a plurality of virtual cameras around the face three-dimensional point cloud, generate virtual face images and corresponding head acupuncture point feature point two-dimensional coordinates in different distance postures, generate a second training data set constructed from the virtual face images and the head acupuncture point feature point two-dimensional coordinates, and use the second training data set to train a two-dimensional mapping network from the face image to the acupuncture point feature points based on a Yolo network as a basic structure.

[0054] The intelligent auxiliary acupuncture point taking module is configured to input an unannotated face image collected by a camera into the trained multi-layer perceptron mapping network to infer three-dimensional space coordinates of acupuncture point feature points, use the inferred three-dimensional space coordinates of the acupuncture point feature points and camera pose projection to calculate coordinates of the acupuncture point feature points on a two-dimensional image, input the unannotated face image into the trained Yolo network to generate two-dimensional reference coordinates of the acupuncture point feature points, and weight and fuse the coordinates on the two-dimensional image and the two-dimensional reference coordinates to complete head acupuncture intelligent auxiliary acupuncture point taking.

[0055] In a third aspect, the present application provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the foregoing head acupuncture semi-supervised intelligent acupuncture point taking method based on visual positioning.

[0056] In a fourth aspect, the present application provides a computer readable storage medium having stored executable instructions, which, when executed by a processor, enable the processor to implement the foregoing head acupuncture semi-supervised intelligent acupuncture point taking method based on visual positioning.

[0057] Those skilled in the art should clearly understand that the modules and method steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware, computer software or a combination of both. The programs corresponding to the software modules and method steps can be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. In order to clearly illustrate the interchangeability of electronic hardware and software, the components and steps of each example have been described in the above description in general terms. Whether the functions are performed in electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0058] The terms "first", "second", and the like are used to distinguish similar objects, rather than to describe or indicate a particular order or sequence.

[0059] The term "comprising" or any other similar term is intended to encompass non-exclusive inclusion, so that a process, method, article or device / apparatus including a series of elements includes not only those elements, but also other elements not explicitly listed, or inherent to the process, method, article or device / apparatus.

[0060] The above specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only for specific embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A semi-supervised intelligent acupoint selection method for head acupuncture based on visual positioning, characterized in that: The method comprises the following steps: Step S1: collect a face image and extract standard two-dimensional facial feature points, and use the 3DMM model to solve the camera pose of the face image, the three-dimensional spatial coordinates of the standard two-dimensional facial feature points, and the three-dimensional point cloud of the face; Step S2, selecting the locations of acupuncture point feature points on the three-dimensional point cloud of the face, constructing a first training data set using the three-dimensional spatial coordinates of the standard two-dimensional facial feature points and the three-dimensional spatial coordinates of the acupuncture point feature points, and training a multilayer perceptron mapping network from the three-dimensional spatial coordinates of the standard two-dimensional facial feature points to the three-dimensional spatial coordinates of the acupuncture point feature points; Step S3: inputting the three-dimensional spatial coordinates of the facial three-dimensional point cloud and the standard two-dimensional facial feature points into a virtual engine, and evenly setting multiple virtual cameras around the three-dimensional facial point cloud to generate virtual facial images at different distances and postures and the corresponding two-dimensional coordinates of the head acupuncture point feature points, and generating a second training data set constructed with the virtual facial images and the two-dimensional coordinates of the head acupuncture point feature points. The second training data set is used to train a two-dimensional mapping network from facial images to acupuncture point feature points based on the Yolo network. Step S4: Input the unlabeled face image captured by the camera into the trained multi-layer perceptron mapping network to infer the three-dimensional spatial coordinates of the acupuncture point feature points, use the inferred three-dimensional spatial coordinates of the acupuncture point feature points and the camera posture projection to calculate the coordinates of the acupuncture point feature points on the two-dimensional image, input the unlabeled face image into the trained Yolo network to generate the two-dimensional reference coordinates of the acupuncture point feature points; weighted fusion of the coordinates on the two-dimensional image and the two-dimensional reference coordinates to complete the intelligent assisted acupuncture point selection on the head.

2. A semi-supervised intelligent acupoint selection method for head acupuncture based on visual positioning according to claim 1, characterized in that: The step S1 includes identifying 68 standard two-dimensional facial feature points on the collected face image. , using the 3DMM model to solve the 3D spatial coordinates of 68 standard 2D facial feature points : ， in, is the average position of the standard two-dimensional facial feature points, For the face shape base, is the facial shape feature coefficient, For facial expression base, is the facial expression feature coefficient, m and n represent the total number of facial shape bases and facial expression bases respectively; Construct the reprojection error objective function: ， Where, is the standard projection matrix, is the face pose and rotation matrix relative to the camera, is the translation vector of the face relative to the camera pose; Minimize the reprojection error objective function and solve 、 、 、 , obtain the face pose relative to the camera, the three-dimensional spatial coordinates of the standard two-dimensional facial feature points, and the 3DMM model feature coefficients; The obtained 3DMM feature coefficients are used to obtain the three-dimensional point cloud of the face.

3. The semi-supervised intelligent acupoint selection method for head acupuncture based on visual positioning according to claim 1 is characterized in that: The step S2 includes selecting acupuncture point feature points on the three-dimensional point cloud of the face. ,in is the number of selected acupuncture point feature points, Construct the first training dataset for the elements, For the objective function, train the multilayer perceptron mapping network ; 。 4. The semi-supervised intelligent acupoint selection method for head acupuncture based on visual positioning according to claim 1 is characterized in that: In step S4, the coordinates of the acupuncture point feature points on the two-dimensional image are calculated using the inferred three-dimensional spatial coordinates and the camera posture projection: ; The two-dimensional reference coordinates of the acupuncture point feature points are: The weighted fusion of the coordinates on the two-dimensional image and the two-dimensional reference coordinates to generate the final two-dimensional reasoning coordinates of the acupuncture points includes: ， in, 、 is a hyperparameter, determined through parameter adjustment experiments. The final two-dimensional coordinates of the acupuncture points.

5. A semi-supervised intelligent acupoint selection system for head acupuncture based on visual positioning, characterized in that: The system comprises: The 3DMM model solving module is used to collect facial images and extract standard two-dimensional facial feature points. It uses the 3DMM model to solve the camera pose of the facial image, the three-dimensional spatial coordinates of the standard two-dimensional facial feature points, and the three-dimensional point cloud of the face. a multilayer perceptron mapping network training module for selecting the locations of acupuncture point feature points on the three-dimensional point cloud of the face, constructing a first training data set using the three-dimensional spatial coordinates of the standard two-dimensional facial feature points and the three-dimensional spatial coordinates of the acupuncture point feature points, and training a multilayer perceptron mapping network from the three-dimensional spatial coordinates of the standard two-dimensional facial feature points to the three-dimensional spatial coordinates of the acupuncture point feature points; a two-dimensional mapping network training module, configured to input the three-dimensional spatial coordinates of the facial three-dimensional point cloud and the standard two-dimensional facial feature points into a virtual engine, and evenly arrange multiple virtual cameras around the three-dimensional facial point cloud to generate virtual facial images at different distances and postures and the corresponding two-dimensional coordinates of the head acupuncture point feature points, and generate a second training data set constructed with the virtual facial images and the two-dimensional coordinates of the head acupuncture point feature points. The second training data set is used to train a two-dimensional mapping network from facial images to acupuncture point feature points based on a Yolo network; The intelligent assisted acupoint selection module is used to input the unlabeled facial image captured by the camera into a trained multi-layer perceptron mapping network to infer the three-dimensional spatial coordinates of the acupuncture point feature points, use the inferred three-dimensional spatial coordinates of the acupuncture point feature points and the camera posture projection to calculate the coordinates of the acupuncture point feature points on the two-dimensional image, input the unlabeled facial image into the trained Yolo network to generate the two-dimensional reference coordinates of the acupuncture point feature points; and weightedly fuse the coordinates on the two-dimensional image and the two-dimensional reference coordinates to complete intelligent assisted acupuncture point selection on the head.

6. An electronic device, characterized in that: include: one or more processors; a memory for storing one or more programs; Wherein, when one or more programs are executed by the one or more processors, the one or more processors implement the semi-supervised intelligent acupoint selection method for head acupuncture based on visual positioning as described in any one of claims 1-4.

7. A computer-readable storage medium, characterized in that Executable instructions are stored thereon, which, when executed by a processor, enable the processor to implement a semi-supervised intelligent acupoint selection method for head acupuncture based on visual positioning as described in any one of claims 1-4.

Images