Method for monitoring posture development of user based on AI digital human hospital guide system

Abstract

The invention discloses a method for monitoring posture development of a user based on an AI digital human hospital guide system, and the method comprises the following steps: a virtual character model library is pre-stored in intelligent hospital guide equipment, and the virtual character model library comprises a virtual character model and a virtual character joint control point P; the camera device collects image information of a user; a joint point calculation module analyzes and processes the image information in the step A based on a computer vision technology, and then extracts a joint feature point Q corresponding to a virtual character joint control point P on the body of the user; the intelligent hospital guide equipment establishes a joint feature point three-dimensional model according to the joint feature points Q on the body of the user; and the model analysis module analyzes the distance proportion and the symmetry degree of each joint feature point according to the joint feature point three-dimensional model. When the distance proportion in the step S5 and the preset value exceed the preset threshold value, or the symmetry difference exceeds the preset threshold value, the intelligent hospital guide equipment sends prompt suggestion information, posture development is monitored, and postnatal limb deformation is prevented.

Description

Technical Field

[0001] The present invention specifically relates to a method for monitoring the body development of a user based on an AI digital human diagnosis guidance system. Background Art

[0002] With the growing demand for medical services and the rapid advancement of technology, hospitals are facing the important challenge of how to improve service quality and optimize patients' medical experience. The traditional medical process usually involves multiple links, including registration, waiting for the call, medical treatment, examination and medication. This series of processes is not only time-consuming, but also often requires patients to run back and forth between different departments of the hospital, which brings great inconvenience to patients, especially elderly patients and those with limited mobility. In order to solve the above problems, many large hospitals have introduced intelligent electronic diagnosis and treatment systems and guidance equipment to improve patients' medical experience. For example, the "Hospital Diagnosis and Treatment Management System and Method" with patent number CN201910745328.9 simplifies the registration and medical treatment process by setting up a self-service terminal, thereby improving efficiency. In addition, the "A Hospital Intelligent Medical Guidance Device and Method" with patent number CN201810093728.1 also provides an intelligent guidance device that can guide patients to quickly find the required departments and services.

[0003] However, although these intelligent systems and devices have achieved remarkable results in process optimization, they lack the function of monitoring the user's body posture. In real life, even clinicians find it difficult to detect some limb problems caused by acquired habits due to lack of observation conditions. For example, high and low shoulders usually refer to the height inconsistency between the two shoulders. This may be caused by muscle tension or imbalance due to long-term maintenance of bad posture (such as a single-shoulder backpack, long-term limb movement to one side, etc.). O-shaped legs (knee valgus): refers to standing with the feet together, but the knees cannot be close together, forming an "O" shape. Children may have a slight O-shaped leg for a period of time during their growth and development. This is a normal phenomenon, but most of them will naturally improve with age. On the contrary, older children and adults may have factors such as long-term wearing of inappropriate shoes and inaccurate walking posture. In addition, there are many limb problems such as hunchback, flat feet, X-shaped legs and knee valgus, most of which are caused by acquired habits. Most of the above problems are due to the inability to obtain effective clinical observation and suggestions.

[0004] The present invention is just produced based on the above-mentioned shortcomings. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method for managing user biometric information in a medical guidance system based on artificial intelligence, which can automatically provide reminders and further improve the patient's medical experience and safety.

[0006] The present invention is achieved through the following technical solutions:

[0007] The present invention provides a method for monitoring the body development of a user based on an AI digital human medical guidance system, comprising an intelligent medical guidance device, wherein the intelligent medical guidance device is provided with a display module for displaying AI virtual characters and text and picture information, and a camera device for collecting user image information;

[0008] The method for monitoring the user's body development based on the AI ​​digital human diagnosis guidance system comprises the following steps:

[0009] S1. The intelligent medical guidance device has a virtual character model library pre-stored therein, and the virtual character model library includes a virtual character model and a virtual character joint control point P;

[0010] S2. When the user uses the intelligent medical guidance device, the camera device collects the user's image information;

[0011] S3. The intelligent medical guidance device is further provided with a node calculation module. The joint point calculation module analyzes and processes the image information of step A based on computer vision technology, and then extracts joint feature points Q corresponding to the virtual character joint control points P on the user;

[0012] S4, the intelligent diagnosis guidance device establishes a three-dimensional model of joint feature points according to the joint feature points Q on the user;

[0013] S5. The intelligent medical guidance device is also provided with a model analysis module, which analyzes the distance ratio and symmetry of each joint feature point according to the three-dimensional model of the joint feature points.

[0014] S6. When the distance ratio in step S5 exceeds the preset threshold value with respect to the preset value, or when the difference in the symmetry in step S5 exceeds the preset threshold value, the intelligent diagnosis guidance device issues a prompt suggestion message.

[0015] The method for monitoring the user's body development based on the AI ​​digital human diagnosis guidance system as described above,

[0016] In step S1, the intelligent diagnosis guidance device also pre-stores a joint control point motion trajectory library, and the joint control point motion trajectory library at least includes a walking motion trajectory;

[0017] In step S5, the intelligent medical guidance device is further provided with an action comparison module, wherein the joint point calculation module calculates the motion path of the user's joint feature point Q according to the dynamic change of the position of the joint feature point Q in the dynamic image information, generates the user's joint dynamic path and sends it to the action comparison module, the action comparison module retrieves the motion trajectory of the virtual character's joint control point in the control point motion trajectory library, and the action comparison module calculates and compares the degree of overlap between the motion path of the user's joint feature point Q and the motion path of the virtual character's joint control point P;

[0018] In step S6, a preset threshold value of the degree of overlap is preset in the action comparison module. When the degree of overlap calculated by the action comparison module in step S5 exceeds the preset threshold value of the degree of overlap, it is deemed that the intelligent medical guidance device issues a prompt suggestion information.

[0019] The method for monitoring the user's body development based on the AI ​​digital human diagnosis guidance system as described above,

[0020] The virtual character joint control points P include the upper center point of the torso, the lower center point of the torso, the center point of the head, the center point of the left shoulder, the center point of the right shoulder, the center point of the left elbow, the center point of the right elbow, the center point of the left wrist, the center point of the right wrist, the center point of the left hip, the center point of the right hip, the center point of the left knee, the center point of the right knee, the center point of the left ankle, and the center point of the right ankle;

[0021] The joint feature points Q include upper trunk feature points, lower trunk feature points, head feature points, left shoulder feature points, right shoulder feature points, left elbow feature points, right elbow feature points, left wrist feature points, right wrist feature points, left hip feature points, right hip feature points, left knee feature points, right knee feature points, left ankle feature points, and right ankle feature points, which correspond one to one with each joint control point P.

[0022] In the above-mentioned method for monitoring the user's body development based on the AI ​​digital human guidance system, the virtual character joint control points P include the center point of the nose, the center point of the chin, the center point of the left eye, and the center point of the left eye; the joint feature points Q include the feature point of the nose, the feature point of the chin, the feature point of the left eye, and the feature point of the left eye;

[0023] In step S1, the joint control point motion trajectory library also includes a head motion trajectory;

[0024] In step S5, the intelligent medical guidance device is further provided with an action comparison module, wherein the joint point calculation module calculates the motion path of the user's joint feature point Q according to the dynamic change of the position of the joint feature point Q in the dynamic image information, generates a motion path data packet and sends it to the action comparison module, the action comparison module retrieves the motion trajectory of the virtual character joint control point in the control point motion trajectory library, and the action comparison module calculates and compares the degree of overlap between the motion path of the user's joint feature point Q and the motion path of the virtual character joint control point P;

[0025] In step S6, a preset threshold value of the degree of overlap is pre-set in the action comparison module. When the degree of overlap calculated by the action comparison module in step S5 exceeds the preset threshold value of the degree of overlap, the intelligent diagnosis guidance device issues a prompt suggestion message.

[0026] The method for monitoring the user's body development based on the AI ​​digital human diagnosis guidance system as described above,

[0027] In step S3, the joint point calculation module performs frame processing on the image information to obtain time sequence coding and corresponding image frames, and extracts each joint feature point Q using the Alphapose algorithm or the Openpose tool;

[0028] In step S4, the three-dimensional model of joint feature points and the dynamic path of the user's joints are generated by Transformer-based structure modeling.

[0029] The method for monitoring the user's body development based on the AI ​​digital human guidance system as described above, according to the gymnastics movement demonstration system of an intelligent device with a virtual character according to claim 1, the camera device is a binocular camera, and the model analysis module calculates the position of the feature point Q of each joint of the human body according to the parallax principle.

[0030] Compared with the prior art, the present invention has the following advantages:

[0031] The present invention provides a method for monitoring the user's body development based on an AI digital human guidance system. When the user's limb proportions or limb movements are greatly different from a preset virtual character model, corrections or medical reminders are promptly made, thereby promptly correcting the user's bad habits, improving the quality of health monitoring and reminder services for the user, further enhancing the patient's medical experience and benefiting the user's health. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a structural schematic diagram of the intelligent diagnosis guidance device of the present invention;

[0033] Figure 2 is a schematic diagram of a virtual character model and joint control points of the present invention;

[0034] Figure 3 is a schematic diagram of a three-dimensional model of user joint feature points of the present invention;

[0035] Figure 4 It is a schematic diagram of the principle of the intelligent medical guidance device of the present invention;

[0036] Figure 5 This is a schematic diagram of the principle of a method for monitoring the user's body development based on an AI digital human guidance system. DETAILED DESCRIPTION

[0037] The invention will be further described below in conjunction with the accompanying drawings:

[0038] The directions described in the specification of the invention, such as "up", "down", "left", "right", "front", "back", etc., are based on the directions of the drawings and are intended to facilitate the description of the relationship between the various components. They do not indicate the unique or absolute positional relationship between the various components. They are only one of the implementation methods of the invention and are not a limitation on its implementation method.

[0039] The present invention provides a method for monitoring the physical development of users based on an AI digital human guidance system, including an intelligent guidance device 1, which is provided with a display module 11 for displaying AI virtual characters and text and picture information, and a camera device 12 for collecting user image information; in addition, the intelligent guidance device 1 is provided with an intelligent response system for conducting voice dialogues with users, and the response system is an intelligent response model based on a recurrent neural network. The intelligent response system can also refer to the patent number CN202410722371.4, and the name is a technical solution of a medical service platform guidance analysis method and system based on intelligent response. The intelligent guidance device 1 is used for collecting user voice information. The sound receiving device 101 and the sound generating device 102 for sound output.

[0040] The method for monitoring the user's body development based on the AI ​​digital human diagnosis guidance system comprises the following steps:

[0041] S1. The intelligent medical guidance device 1 pre-stores a virtual character model library 13, which includes a virtual character model and virtual character joint control points P;

[0042] S2. When the user uses the intelligent medical guidance device 1, the camera device 12 collects the user's image information;

[0043] S3. The intelligent medical guidance device 1 is further provided with a node calculation module 14. The joint point calculation module 14 analyzes and processes the image information of step A based on computer vision technology, and then extracts joint feature points Q corresponding to the virtual character joint control points P on the user;

[0044] S4, the intelligent diagnosis guidance device 1 establishes a three-dimensional model of joint feature points according to the joint feature points Q on the user;

[0045] S5. The intelligent medical guidance device 1 is further provided with a model analysis module 15. The model analysis module 15 analyzes the distance ratio and symmetry of each joint feature point according to the three-dimensional model of the joint feature points.

[0046] S6. When the distance ratio in step S5 exceeds the preset threshold value with respect to the preset value, or when the difference in the symmetry in step S5 exceeds the preset threshold value, the intelligent medical guidance device 1 issues a prompt suggestion message.

[0047] In step S1, a virtual character model created in advance is imported into the intelligent medical guidance device 1. The virtual character model includes a topological structure, a contour, and joint control points P of the model. In detail, Figure 2 As shown, the virtual character joint control points P include the center point P on the torso 01 , the lower center point of the trunk P 02 , head center point P 03 , left shoulder center point P 11 , right shoulder center point P 12 , left elbow center point P 111 , right elbow center point P 121 , left wrist center point P 1111 , right wrist center point P 1211 , left hip center point P 21 , right hip center point P 22 , left knee center point P 211 , right knee center point P 221 , left ankle center point P 2111 , right ankle center point P 2211 Moreover, in the pre-imported virtual character model library 13, a virtual character skeleton three-dimensional model is established according to the above joint control points P, with the center point P on the trunk 01 , the lower center point of the trunk P 02 , head center point P 03 The connecting line is the central axis skeleton, with the left shoulder center point P 11 , right shoulder center point P 12 The connecting line is the chest skeleton.

[0048] In step S4, the intelligent diagnosis guidance device 1 establishes a three-dimensional model of joint feature points according to the joint feature points Q on the user, such as Figure 3 As shown, each joint feature point of the three-dimensional model of joint feature points corresponds to each node of the three-dimensional model of the virtual character skeleton;

[0049] Among them, in step S5, the distance ratio of each joint feature point analyzed by the three-dimensional model means that the distance between two feature points in the three-dimensional model of joint feature points is first used as a reference, and the distance between other feature points and the reference is calculated and analyzed, as well as the ratio between the distance between the other two feature points and the reference. For example, the distance between the center point P on the trunk is calculated. 01 , the lower center point of the trunk P 02 The connection line L 01-02 As a benchmark, the left shoulder feature point Q 11 Connect with L 01-02 The distance between them is D1, and the left elbow feature point Q 111 Connect with L 01-02 The distance between them is D2, and the left shoulder feature point Q 11 and the left elbow feature point Q 111The line L between 11-111 , connect L 11-111 Connect with L 01-02 The distance D3 between them also has corresponding distances D11, D21, and D31 in the 3D model of the virtual character skeleton. The calculation is performed accordingly, that is, the difference D10 between D1 and D11, the difference D20 between D2 and D21, and the difference D30 between D3 and D31. Whether D10, D20, and D30 exceed the threshold preset by the system. The symmetry of each joint feature point analyzed by the 3D model refers to the center point P on the trunk. 01 , the lower center point of the trunk P 02 The connection line L 01-02 As a benchmark, judge the symmetry of the center points on the left and right sides of the benchmark relative to the benchmark.

[0050] Among them, in step S5, the intelligent medical guidance device 1 issues a prompt suggestion information, which may be: "The user's left and right shoulders are less symmetrical, it is recommended to see a doctor", "The user's gait is incorrect, it is recommended to correct it in time or see a doctor in time"

[0051] As a preferred implementation:

[0052] In step S1, the intelligent medical guidance device 1 also pre-stores a joint control point motion trajectory library 16, which at least includes walking motion trajectories. The motion trajectory of the virtual character model is pre-imported into the intelligent medical guidance device 1. In this example, the walking motion trajectory of the virtual character model is imported;

[0053] In step S5, the intelligent medical guidance device 1 is further provided with an action comparison module 17. The joint point calculation module 14 calculates the motion path of the user's joint feature point Q according to the dynamic change of the position of the joint feature point Q in the dynamic image information, generates the user's joint dynamic path and sends it to the action comparison module 17. The action comparison module 17 retrieves the virtual character joint control point motion trajectory in the control point motion trajectory library. The action comparison module 17 calculates and compares the degree of overlap between the motion path of the user's joint feature point Q and the motion path of the virtual character joint control point P. In this step, the camera device 12 collects the user's dynamic image information in real time, the joint point calculation module 14 calculates the three-dimensional position change of the joint feature point Q according to the dynamic image information, and the joint point calculation module 14 calculates the motion path of the user's joint feature point Q according to the dynamic change of the position of the joint feature point Q in the image information. In this embodiment, the degree of overlap between the motion path of the user's joint feature point Q and the motion path of the virtual character joint control point P includes: the change in the three-dimensional angle of the line between two adjacent joint feature points Q. For example, in a gymnastics movement, the left shoulder feature point Q 11 and the left elbow feature point Q 111 The line L between 11-111, with the left shoulder feature point Q 11 As the reference point, connect the line L 11-111 The three-dimensional angle of rotation around the reference point is γ, which is then compared with the three-dimensional angle γ of the joint control point P corresponding to the virtual character. For more details, please refer to the paper DOI: 10.26914 / c.cnkihy.2021.065155, titled "Action Sequence Similarity Calculation Algorithm Based on OpenPose".

[0054] In step S6, a preset threshold value of the overlap degree is preset in the action comparison module 17. When the overlap degree calculated by the action comparison module 17 in step S5 exceeds the preset threshold value of the overlap degree, the intelligent medical guidance device 1 issues a prompt suggestion message.

[0055] like Figure 2 As shown, the virtual character joint control points P include the upper center point of the torso, the lower center point of the torso, the center point of the head, the center point of the left shoulder, the center point of the right shoulder, the center point of the left elbow, the center point of the right elbow, the center point of the left wrist, the center point of the right wrist, the center point of the left hip, the center point of the right hip, the center point of the left knee, the center point of the right knee, the center point of the left ankle, and the center point of the right ankle;

[0056] like Figure 3 As shown, the joint feature points Q include upper trunk feature points, lower trunk feature points, head feature points, left shoulder feature points, right shoulder feature points, left elbow feature points, right elbow feature points, left wrist feature points, right wrist feature points, left hip feature points, right hip feature points, left knee feature points, right knee feature points, left ankle feature points, and right ankle feature points, which correspond one by one to each joint control point P.

[0057] As a preferred implementation:

[0058] The virtual character joint control points P include the center point of the nose, the center point of the chin, the center point of the left eye, and the center point of the left eye; the joint feature points Q include the feature point of the nose, the feature point of the chin, the feature point of the left eye, and the feature point of the left eye;

[0059] In step S1, the joint control point motion trajectory library 16 also includes a head motion trajectory;

[0060] In step S5, the intelligent medical guidance device 1 is also provided with a motion comparison module 17. The joint point calculation module 14 calculates the motion path of the user's joint feature point Q according to the dynamic changes of the position of the joint feature point Q in the dynamic image information, generates a motion path data packet and sends it to the motion comparison module 17. The motion comparison module 17 retrieves the motion trajectory of the virtual character joint control point in the control point motion trajectory library. The motion comparison module 17 calculates and compares the degree of overlap between the motion path of the user's joint feature point Q and the motion path of the virtual character joint control point P. The principle is the same as step S5 above.

[0061] In step S6, a preset threshold value of the overlap degree is preset in the action comparison module 17. When the overlap degree calculated by the action comparison module 17 in step S5 exceeds the preset threshold value of the overlap degree, the intelligent medical guidance device 1 issues a prompt suggestion message.

[0062] In detail, in step S3, the joint point calculation module 14 performs frame processing on the image information to obtain time sequence coding and corresponding image frames, and extracts each joint feature point Q using the Alphapose algorithm or the Openpose tool;

[0063] In step S4, the three-dimensional model of joint feature points and the dynamic path of the user's joints are generated by Transformer-based structure modeling.

[0064] The joint point calculation module 14 calculates the position of each joint feature point Q of the human body according to the parallax principle, wherein the derivation formula of the distance parallax principle D1= / calculates the distance D1 between the camera and the user, wherein f=Focallength, i.e., the focal length of the camera of the binocular camera 13, b=Baseline, i.e., b is the baseline distance between the two cameras of the binocular camera 13, d=Disparityvalue, i.e., d is the parallax, and the parallax refers to the difference in the pixel position between the two camera images. Assuming that the left image in the stereo vision camera has a pixel at position (1,30), and the same pixel exists at position (4,30) in the right image, the parallax value or difference is (4-1)=3, ps=Pixel size, i.e., ps is the size of a single pixel in the image sensor. Since the binocular camera uses two cameras, the pixel size of the image sensor must be the same. The angle calculation module 15 includes an inertial measurement unit IMU composed of a gyroscope and an accelerometer, and the use of a binocular camera can save computing power.

[0065] The above is only an embodiment of the present invention, and the common knowledge such as the known specific structure and characteristics in the scheme is not described in detail here. It should be pointed out that for those skilled in the art, several deformations and improvements can be made without departing from the structure of the present invention, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A method for monitoring the user's body development based on an AI digital human diagnosis guidance system, characterized in that: The intelligent medical guidance device (1) is provided with a display module (11) for displaying an AI virtual character and text and picture information, and a camera device (12) for collecting user image information; The method for monitoring the user's body development based on the AI ​​digital human diagnosis guidance system comprises the following steps: S1. The intelligent medical guidance device (1) pre-stores a virtual character model library (13), wherein the virtual character model library (13) includes a virtual character model and virtual character joint control points P; S2. When the user uses the intelligent medical guidance device (1), the camera device (12) collects image information of the user; S3. The intelligent medical guidance device (1) is further provided with a node calculation module (14). The joint point calculation module (14) analyzes and processes the image information of step A based on computer vision technology, and then extracts joint feature points Q corresponding to the joint control points P of the virtual character on the user; S4, the intelligent diagnosis guidance device (1) establishes a three-dimensional model of joint feature points according to the joint feature points Q of the user; S5. The intelligent medical guidance device (1) is further provided with a model analysis module (15). The model analysis module (15) analyzes the distance ratio and symmetry of each joint feature point according to the three-dimensional model of the joint feature points. S6. When the distance ratio in step S5 exceeds the preset threshold value, or the difference in the symmetry in step S5 exceeds the preset threshold value, the intelligent medical guidance device (1) issues a prompt suggestion message.

2. The method for monitoring the user's body development based on the AI ​​digital human diagnosis guidance system according to claim 1 is characterized in that: In step S1, the intelligent medical guidance device (1) also pre-stores a joint control point motion trajectory library (16), and the joint control point motion trajectory library (16) at least includes a walking motion trajectory; In step S5, the intelligent medical guidance device (1) is further provided with an action comparison module (17), wherein the joint point calculation module (14) calculates the motion path of the user's joint feature point Q according to the dynamic change of the position of the joint feature point Q in the dynamic image information, generates the user's joint dynamic path and sends it to the action comparison module (17), the action comparison module (17) retrieves the motion trajectory of the virtual character's joint control point in the control point motion trajectory library, and the action comparison module (17) calculates and compares the degree of overlap between the motion path of the user's joint feature point Q and the motion path of the virtual character's joint control point P; In step S6, a preset threshold value of the degree of overlap is pre-set in the action comparison module (17). When the degree of overlap calculated by the action comparison module (17) in step S5 exceeds the preset threshold value of the degree of overlap, it is deemed that the intelligent medical guidance device (1) issues a prompt suggestion message.

3. The method for monitoring the user's body development based on the AI ​​digital human diagnosis guidance system according to claim 1 is characterized in that: The virtual character joint control points P include the center point P on the torso 01 , the lower center point of the trunk P 02 , head center point P 03 , left shoulder center point P 11 , right shoulder center point P 12 , left elbow center point P 111 , right elbow center point P 121 , left wrist center point P 1111 , right wrist center point P 1211 , left hip center point P 21 , right hip center point P 22 , left knee center point P 211 , right knee center point P 221 , left ankle center point P 2111 , right ankle center point P 2211 ; The joint feature points Q include the trunk feature points Q corresponding to each joint control point P. 01 , the lower feature point Q of the torso 02 , head feature point Q 03 , left shoulder feature point Q 11 , right shoulder feature point Q 12 , left elbow feature point Q 111 , right elbow feature point Q 121 , left wrist feature point Q 1111 , right wrist feature point Q 1211 , left hip feature point Q 21 , right hip feature point Q 22 , left knee feature point Q 211 , right knee feature point Q 221 , left ankle feature point Q 2111 , right ankle feature point Q 2211 .

4. The method for monitoring the user's body development based on the AI ​​digital human diagnosis guidance system according to claim 1 is characterized in that: The virtual character joint control points P include the nose center point P 31 , Chin center point P 32 , left eye center point P 41 , left eye center point P 42 ; Joint feature points Q include, nose feature points Q 31 , chin feature point Q 32 , left eye feature point Q 41 , left eye feature point Q 42 ; In step S1, the joint control point motion trajectory library (16) also includes a head motion trajectory; In step S5, the intelligent medical guidance device (1) is further provided with a motion comparison module (17), wherein the joint point calculation module (14) calculates the motion path of the user's joint feature point Q according to the dynamic change of the position of the joint feature point Q in the dynamic image information, generates a motion path data packet and sends it to the motion comparison module (17), the motion comparison module (17) retrieves the motion trajectory of the virtual character's joint control point in the control point motion trajectory library, and the motion comparison module (17) calculates and compares the degree of overlap between the motion path of the user's joint feature point Q and the motion path of the virtual character's joint control point P; In step S6, a preset threshold value of the degree of overlap is pre-set in the action comparison module (17). When the degree of overlap calculated by the action comparison module (17) in step S5 exceeds the preset threshold value of the degree of overlap, the intelligent medical guidance device (1) issues a prompt suggestion message.

5. The method for monitoring the user's body development based on the AI ​​digital human diagnosis guidance system according to claim 2 is characterized in that: In step S3, the joint point calculation module (14) performs frame processing on the image information to obtain time sequence coding and corresponding image frames, and extracts each joint feature point Q using the Alphapose algorithm or the Openpose tool; In step S4, the three-dimensional model of joint feature points and the dynamic path of the user's joints are generated by Transformer-based structure modeling.

6. The method for monitoring the user's body development based on the AI ​​digital human diagnosis guidance system according to claim 1 is characterized in that: According to the gymnastics action demonstration system for an intelligent device with a virtual character as claimed in claim 1, the camera device (12) is a binocular camera, and the model analysis module (15) calculates the position of the characteristic point Q of each joint of the human body according to the parallax principle.

Images