Human body posture dynamic intelligent identification system

Abstract

The invention relates to the technical field of medical rehabilitation, and discloses a human body posture dynamic intelligent recognition system, which comprises a video acquisition module comprising a camera and used for acquiring a motion posture and a joint behavior video of a patient; the recognition and posture analysis module is used for analyzing the motion posture and the joint behavior video and recognizing the physical state and action of the patient; and the joint angle evaluation module is used for analyzing the physical state and action of the patient according to standard joint behaviors and determining the joint health degree of the patient. The joint movement angle of the patient can be accurately detected and compared with the standard condition, the joint health degree is judged, diagnosis of a doctor is assisted, rehabilitation of the patient is helped, the diagnosis efficiency of the doctor is improved, and rehabilitation training of the patient is helped.

Description

Technical Field

[0001] The present invention relates to the field of medical rehabilitation technology, and in particular to a human body posture dynamic intelligent recognition system. Background Art

[0002] In the modern medical field, accurate diagnosis of joint range of motion and effective management of patient sports rehabilitation are crucial. However, traditional diagnosis and rehabilitation methods often rely on doctors' subjective judgment and experience, lacking objective, accurate and quantitative evaluation methods.

[0003] With the continuous advancement of science and technology, significant developments have been made in the fields of medical imaging, sensor technology, and computer algorithms. Although medical imaging such as X-rays, CT, and MRI can provide structural information about joints, they have limitations in evaluating dynamic joint range of motion.

[0004] The emergence of sensor technology has made it possible to monitor joint movements in real time and continuously, but early sensors lacked accuracy and stability, and data processing and analysis capabilities need to be improved.

[0005] The development of computer algorithms has made it possible to quickly and accurately process and analyze large amounts of complex joint movement data, laying the foundation for building smarter and more efficient assistance systems.

[0006] However, the existing related systems still have deficiencies in integration, ease of use, and personalized customization, and are difficult to meet the diverse needs of clinicians and patients. Therefore, there is an urgent need for an intelligent human posture dynamic recognition system to assist clinical doctors in diagnosing joint mobility and patient sports rehabilitation. Summary of the invention

[0007] The purpose of the present invention is to provide a human body posture dynamic intelligent recognition system to solve the problem that the existing related systems still have deficiencies in integration, ease of use, personalized customization, etc., and are difficult to meet the diverse needs of clinicians and patients.

[0008] The present invention provides a human body posture dynamic intelligent recognition system, comprising:

[0009] A video acquisition module, including a camera, for acquiring a video of a patient's movement posture and joint behavior;

[0010] A recognition and posture analysis module, used to analyze the movement posture and joint behavior video to recognize the patient's body state and movement;

[0011] The joint angle assessment module is used to analyze the patient's physical state and movements based on standard joint behaviors to determine the patient's joint health.

[0012] Preferably, the camera comprises:

[0013] High-definition image sensors to capture videos of patients’ movement postures and joint behaviors;

[0014] Optical lens, used to determine the clarity of the video of movement posture and joint behavior;

[0015] A focusing unit, used for automatically adjusting the focus according to the patient's movement status;

[0016] Data transmission interface, used to transmit motion posture and joint behavior video to the recognition and posture analysis module.

[0017] Preferably, the recognition and posture analysis module includes:

[0018] A feature extraction unit, used to extract the patient's body feature information from the motion posture and joint behavior video;

[0019] A recognition algorithm unit, used to analyze and recognize the human body feature information to determine the patient's physical state and movements;

[0020] A data transmission unit, used for transmitting the patient's physical state and movements to a storage unit;

[0021] A storage unit is used to store the patient's physical status and movements for information sharing.

[0022] Preferably, the feature extraction unit, when used to extract the patient's human body feature information from the motion posture and joint behavior video, includes:

[0023] Determine key video frames of the patient's motion in the motion posture and joint behavior video, and preprocess the key video frames, wherein the preprocessing includes grayscale processing, histogram equalization processing, and denoising processing;

[0024] The preprocessed key video frames are input into the human feature recognition model to determine the patient's human feature information.

[0025] Preferably, the grayscale processing is performed according to the following formula:

[0026] g(x, y) = a × f(x, y) + b;

[0027] Wherein, f(x, y) represents the grayscale value of the key video frame, g(x, y) represents the grayscale value of the key video frame after grayscale processing, and a and b represent constants.

[0028] Preferably, the histogram equalization processing step includes:

[0029] Determine a grayscale histogram h(rk) of the key video frame after grayscale processing, where rk represents the grayscale level;

[0030] Calculate the cumulative distribution function sk of the key video frame after grayscale processing, the cumulative distribution function n(j) is the number of pixels with gray level j, and N is the total number of pixels in the key video frame;

[0031] According to the cumulative distribution function sk, a new grayscale value gx(x, y) is obtained, gx(x, y)=INT[(L-1)×sk+0.5], where L represents the grayscale level.

[0032] Preferably, the denoising step includes:

[0033] Mean filtering is used for denoising, specifically:

[0034]

[0035] G(x, y) represents the grayscale value of the key video frame after denoising.

[0036] Preferably, the joint angle assessment module comprises:

[0037] An angle analysis unit, used to analyze the patient's physical condition and movements and determine the patient's joint movement angles;

[0038] An angle comparison unit, used for determining a standard joint motion angle of a standard joint behavior, comparing the patient's joint motion angle with the standard joint motion angle, and determining a joint assessment result of the patient;

[0039] An error calibration unit, used to calibrate the patient's joint assessment results to determine the patient's joint health;

[0040] The assessment output unit is used to display the patient's joint assessment results and joint health status.

[0041] Preferably, the angle comparison unit is used to determine the standard joint motion angle of the standard joint behavior, and when comparing the joint motion angle of the patient with the standard joint motion angle to determine the joint assessment result of the patient, it includes:

[0042] Determine the standard joint motion angle according to the standard joint behavior;

[0043] Comparing the patient's joint activity angle with the standard joint activity angle, if the patient's joint activity angle is greater than or equal to the standard joint activity angle, it is determined that the patient's joint has recovered well;

[0044] If the patient's joint motion angle is less than the standard joint motion angle, it is determined that the patient's joint has not recovered.

[0045] Preferably, the error calibration unit, when used to calibrate the patient's joint assessment result to determine the patient's joint health, includes:

[0046] When it is determined that the patient's joint has recovered well, the patient's joint health level is determined to be a first health level L1;

[0047] When it is determined that the patient's joint has not recovered, determining the angle difference between the patient's joint activity angle and the standard joint activity angle, and determining the patient's joint health degree according to the angle difference;

[0048] Wherein, a first angle difference and a second angle difference are preset, and the first angle difference is smaller than the second angle difference;

[0049] If the angle difference is less than the first angle difference, it is determined that the patient's joint health level is a second health level L2;

[0050] If the angle difference is greater than or equal to the first angle difference, and the angle difference is less than the second angle difference, it is determined that the patient's joint health level is a third health level L3;

[0051] If the angle difference is greater than or equal to the second angle difference, it is determined that the patient's joint health level is a fourth health level L4, and L1>L2>L3>L4.

[0052] Compared with the prior art, the beneficial effect of the present invention is that the present invention can accurately detect the patient's joint activity angle, compare it with the standard situation, judge the joint health, assist doctors in diagnosis, help patients recover, improve doctors' diagnostic efficiency and help patients with rehabilitation training. The present invention improves the accuracy and objectivity of joint activity assessment, reduces the influence of doctors' subjective factors; saves evaluation time, improves medical efficiency, and enables doctors to formulate treatment plans for patients more quickly; provides real-time feedback and guidance for patients' sports rehabilitation training, which helps to improve rehabilitation effects; at the same time, it is convenient for long-term tracking and data analysis of patients' conditions and rehabilitation progress, providing strong support for medical research, and has significant practical value and broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0054] Figure 1 It is a functional block diagram of a human body posture dynamic intelligent recognition system of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0056] like Figure 1 As shown, the present invention provides a human body posture dynamic intelligent recognition system, comprising:

[0057] The video acquisition module includes a camera for acquiring the patient's movement posture and joint behavior video.

[0058] The recognition and posture analysis module is used to analyze the movement posture and joint behavior video to recognize the patient's body state and movement.

[0059] The joint angle assessment module is used to analyze the patient's physical state and movements based on standard joint behaviors to determine the patient's joint health.

[0060] In some embodiments of the present application, the camera includes:

[0061] High-definition image sensor used to acquire video of the patient's movement posture and joint behavior.

[0062] Optical lens, used to determine the clarity of the video of movement posture and joint behavior.

[0063] A focusing unit is used to automatically adjust the focus according to the patient's movement status.

[0064] Data transmission interface, used to transmit motion posture and joint behavior video to the recognition and posture analysis module.

[0065] In this embodiment, the camera includes multiple key components to ensure that it can efficiently capture and analyze the patient's motion posture and joint behavior video. First, the high-definition image sensor is the core component of the camera, which is responsible for capturing the patient's motion posture and joint behavior video, ensuring that the image quality is high enough for detailed analysis and recognition. Secondly, the role of the optical lens is to ensure the clarity of the motion posture and joint behavior video. It ensures that the image details are clear and visible through precise optical design and high-quality lens materials, thereby improving the accuracy of the analysis. In addition, the focus unit is an intelligent autofocus system that can automatically adjust the focus according to the patient's motion state. This means that whether the patient is stationary or moving quickly, the camera can quickly adjust the focus to ensure that the image remains clear at all times. This function is crucial for capturing rapidly changing motion postures and joint behaviors because it can reduce blur and distortion and improve the reliability of the data. Finally, the data transmission interface is a bridge between the camera and the recognition and posture analysis module. It is responsible for efficiently and stably transmitting the captured motion posture and joint behavior video to the analysis module. By using high-speed data transmission technology, the data transmission interface ensures that there will be no packet loss or delay in the transmission of video data, thereby ensuring that the analysis module can receive and process data in real time. This link is crucial to the real-time and accuracy of the entire system, as it ensures a seamless process from data capture to analysis, improving overall work efficiency.

[0066] It is understandable that the camera has anti-shake technology, which effectively reduces image blur caused by external interference, thereby providing stable, clear and accurate image data for joint range of motion diagnosis and patient motor rehabilitation. And the high-definition camera technical indicators can be 1080P, 30 frames to ensure that the acquired video has sufficient clarity and frame rate. When collecting videos, patients need to perform specific joint movements, such as flexion, extension, rotation, etc., according to the doctor's instructions. At the same time, in order to ensure the quality and consistency of video data, the acquisition environment should maintain good lighting conditions and minimize background interference. At the same time, a fixed shooting distance and angle may be required for subsequent analysis and processing.

[0067] In some embodiments of the present application, the recognition and gesture analysis module includes:

[0068] The feature extraction unit is used to extract the patient's body feature information from the motion posture and joint behavior video.

[0069] The recognition algorithm unit is used to analyze and recognize the human body feature information to determine the patient's physical state and movements.

[0070] A data transmission unit is used to transmit the patient's physical state and movements to a storage unit.

[0071] A storage unit is used to store the patient's physical status and movements for information sharing.

[0072] In this embodiment, the recognition and posture analysis module mainly includes the following key components: First, the feature extraction unit, the main task of which is to extract the patient's human body feature information from the motion posture and joint behavior video. These feature information include but are not limited to the patient's limb movements, body postures, etc., which are crucial for subsequent analysis and recognition. Secondly, the recognition algorithm unit is responsible for the task of in-depth analysis and recognition of the extracted human body feature information. By using advanced algorithms, the unit can accurately determine the patient's physical state and movement. For example, it can identify whether the patient is in a normal walking state or whether he is performing a specific rehabilitation training movement. The accuracy and efficiency of the recognition algorithm unit directly affect the performance of the entire system. Next, the data transmission unit is responsible for transmitting the patient's physical state and movement information analyzed by the recognition algorithm unit to the storage unit. This process requires ensuring the efficiency and security of data transmission to avoid data loss or tampering during the transmission process. The data transmission unit usually uses encryption technology to ensure the security of data transmission. Finally, the storage unit is used to store the patient's physical state and movement information for information sharing. The storage unit can be a local server or a cloud database, depending on the deployment environment and requirements of the system. The storage unit not only needs to have sufficient storage capacity, but also needs to have efficient data retrieval and management functions so that medical staff can quickly obtain detailed information about patients when needed.

[0073] In some embodiments of the present application, when the feature extraction unit is used to extract the patient's human feature information from the motion posture and joint behavior video, it includes: determining the key video frames of the patient's movement in the motion posture and joint behavior video, and preprocessing the key video frames, wherein the preprocessing includes grayscale processing, histogram equalization processing, and denoising processing; inputting the preprocessed key video frames into a human feature recognition model to determine the patient's human feature information.

[0074] In this embodiment, when the feature extraction unit performs its function, i.e., extracting the patient's human feature information from the motion posture and joint behavior video, it specifically includes the following steps: first, identify and determine the key video frames of the patient's motion in the video. These key video frames are those frames that can best reflect the patient's motion posture and joint behavior. Next, a series of preprocessing operations are performed on these key video frames to ensure the accuracy and efficiency of subsequent processing. The preprocessing step includes three main parts: first, grayscale processing, that is, converting the color video frame into a grayscale image. This process helps to reduce the computational complexity and can retain enough information for subsequent feature extraction. Secondly, histogram equalization processing is a step aimed at improving the contrast of the image, making the details in the image clearer, thereby helping to more accurately identify and extract human features. Finally, denoising processing is performed, which can further improve the accuracy of feature extraction by removing noise in the image. After completing the preprocessing, the processed key video frames are input into the human feature recognition model. This model is trained to identify and extract the patient's human feature information from the input image. These feature information may include but is not limited to the patient's posture, joint angle, motion trajectory, etc. In this way, the system can effectively extract the patient's anthropometric information from the video, providing important data support for subsequent analysis and diagnosis.

[0075] It is understandable that the human feature recognition model is an existing model, such as inputting the preprocessed key video frame into the pattern recognition algorithm based on yolov5 for human recognition and then using the HRNet human key point recognition algorithm to output the position coordinates of the human joint points. Through the joint point position information of multiple consecutive frames, the changes in human posture can be tracked.

[0076] In some embodiments of the present application, the grayscale processing is performed according to the following formula:

[0077] g(x, y) = a × f(x, y) + b;

[0078] Wherein, f(x, y) represents the grayscale value of the key video frame, g(x, y) represents the grayscale value of the key video frame after grayscale processing, and a and b represent constants.

[0079] In some embodiments of the present application, the histogram equalization processing step includes: determining a grayscale histogram h(rk) of the key video frame after grayscale processing, wherein rk represents a grayscale level.

[0080] Calculate the cumulative distribution function sk of the key video frame after grayscale processing, the cumulative distribution function n(j) is the number of pixels with gray level j, and N is the total number of pixels in the key video frame.

[0081] According to the cumulative distribution function sk, a new grayscale value gx(x, y) is obtained, gx(x, y)=INT[(L-1)×sk+0.5], where L represents the grayscale level.

[0082] In some embodiments of the present application, the denoising step includes:

[0083] Mean filtering is used for denoising, specifically:

[0084]

[0085] G(x, y) represents the grayscale value of the key video frame after denoising.

[0086] It is understandable that after determining the patient's body feature information, the recognition algorithm unit can determine the patient's body state and movement according to the deep learning-based human posture estimation algorithm OpenPose.

[0087] In some embodiments of the present application, the joint angle assessment module includes: an angle analysis unit, which is used to analyze the patient's physical state and movements and determine the patient's joint activity angle.

[0088] The angle comparison unit is used to determine the standard joint activity angle of the standard joint behavior, compare the patient's joint activity angle with the standard joint activity angle, and determine the patient's joint assessment result.

[0089] The error calibration unit is used to calibrate the patient's joint assessment results and determine the patient's joint health status.

[0090] The assessment output unit is used to display the patient's joint assessment results and joint health status.

[0091] In this embodiment, the joint angle evaluation module includes multiple key components, which are intended to comprehensively analyze and evaluate the joint health of the patient. First, the angle analysis unit plays a vital role. It can accurately determine the patient's joint activity angle by analyzing the patient's physical state and movements. Next, the angle comparison unit is responsible for comparing the patient's joint activity angle with the standard joint activity angle of the standard joint behavior. Through this comparison, it is possible to more clearly identify whether the patient's joint activity is abnormal or restricted. This link is of great significance for early detection of joint problems and formulation of corresponding treatment plans. In order to further improve the accuracy of the evaluation, the error calibration unit plays an indispensable role. It ensures the reliability and accuracy of the evaluation results by calibrating the patient's joint evaluation results. This process may involve a variety of calibration techniques and algorithms to ensure that the final result can truly reflect the patient's joint health. Finally, the evaluation output unit is responsible for presenting the patient's joint evaluation results and joint health to doctors and patients in an intuitive way. This includes not only numerical data display, but also various forms such as charts and images, so that doctors and patients can better understand and grasp the health of the joints. In this way, patients can understand their physical condition more intuitively, so as to take corresponding rehabilitation measures or preventive measures.

[0092] In some embodiments of the present application, the angle comparison unit is used to determine the standard joint movement angle of the standard joint behavior, and compares the joint movement angle of the patient with the standard joint movement angle to determine the joint assessment result of the patient, including: determining the standard joint movement angle according to the standard joint behavior; comparing the joint movement angle of the patient with the standard joint movement angle, if the joint movement angle of the patient is greater than or equal to the standard joint movement angle, then determining that the patient's joint has recovered well; if the joint movement angle of the patient is less than the standard joint movement angle, then determining that the patient's joint has not recovered.

[0093] In this embodiment, during the joint assessment, the angle comparison unit is used to determine the standard joint motion angle of the standard joint behavior. Specifically, it is first necessary to determine the standard joint motion angle based on the standard joint behavior. This process involves detailed measurement and recording of the range of motion of healthy joints to establish a reference standard. Next, the patient's joint motion angle is compared with this standard joint motion angle. This comparison step is to evaluate the patient's joint recovery. During the comparison process, if it is found that the patient's joint motion angle is greater than or equal to the standard joint motion angle, it can be determined that the patient's joint recovery is good. This means that the patient's joint function has reached a relatively normal level and the range of motion is not significantly restricted. However, if the patient's joint motion angle is less than the standard joint motion angle, it is necessary to determine that the patient's joint has not recovered. This situation usually indicates that the patient's joint function still has certain limitations and the range of motion may be affected. In this case, further treatment or rehabilitation training may be required to help the patient recover joint function.

[0094] In some embodiments of the present application, the error calibration unit is used to calibrate the patient's joint assessment results and determine the patient's joint health level, including: when it is determined that the patient's joints have recovered well, determining the patient's joint health level to be a first health level L1; when it is determined that the patient's joints have not recovered, determining the angle difference between the patient's joint activity angle and the standard joint activity angle, and determining the patient's joint health level based on the angle difference; wherein, the first angle difference and the second angle difference are pre-set, and the first angle difference is less than the second angle difference; if the angle difference is less than the first angle difference, the patient's joint health level is determined to be a second health level L2; if the angle difference is greater than or equal to the first angle difference, and the angle difference is less than the second angle difference, the patient's joint health level is determined to be a third health level L3; if the angle difference is greater than or equal to the second angle difference, the patient's joint health level is determined to be a fourth health level L4, and L1>L2>L3>L4.

[0095] In this embodiment, when the error calibration unit is used to calibrate the patient's joint assessment results to determine the patient's joint health, it specifically includes the following steps: First, the patient's joints are assessed to determine their recovery status. If it is determined that the patient's joints have recovered well, then the patient's joint health can be determined to be a first health level L1. However, if it is determined that the patient's joints have not recovered, it is necessary to further analyze the angle difference between the patient's joint activity angle and the standard joint activity angle. According to this angle difference, the patient's joint health can be determined more accurately. In order to more carefully assess the patient's joint health, two key angle differences are pre-set: a first angle difference and a second angle difference. Among them, the first angle difference is less than the second angle difference. Next, the patient's joint health is determined according to the specific value of the angle difference. Specifically, if the angle difference is less than the first angle difference, then the patient's joint health can be determined to be a second health level L2. If the angle difference is greater than or equal to the first angle difference, but less than the second angle difference, then the patient's joint health can be determined to be a third health level L3. Finally, if the angle difference is greater than or equal to the second angle difference, then the patient's joint health can be determined to be a fourth health level L4. It should be noted that the relationship between these four health levels is: L1>L2>L3>L4. Through this meticulous assessment method, the patient's joint health status can be more accurately understood and a more scientific basis can be provided for subsequent treatment and rehabilitation.

[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

[0097] The system provided in the above embodiment is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be decomposed or combined. For example, the modules in the above embodiment can be combined into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the modules or steps, and are not regarded as improper limitations of the present invention.

[0098] Those skilled in the art should be able to appreciate that the modules and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software or a combination of the two, and the programs corresponding to the software modules and method steps can be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in the technical field. In order to clearly illustrate the interchangeability of electronic hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these 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 to exceed the scope of the present invention.

Claims

1. A human body posture dynamic intelligent recognition system, characterized in that: include: A video acquisition module, including a camera, for acquiring a video of the patient's movement posture and joint behavior; A recognition and posture analysis module is used to analyze the movement posture and joint behavior video to identify the patient's body state and movement; The joint angle assessment module is used to analyze the patient's body state and movements based on standard joint behavior to determine the patient's joint health.

2. The human body posture dynamic intelligent recognition system according to claim 1, characterized in that: The camera includes: High-definition image sensor for capturing video of the patient's movement posture and joint behavior; Optical lens, used to determine the clarity of movement posture and joint behavior video; A focusing unit, used to automatically adjust the focus according to the patient's movement status; Data transmission interface, used to transmit motion posture and joint behavior videos to the recognition and posture analysis module.

3. The human body posture dynamic intelligent recognition system according to claim 1, characterized in that: The recognition and posture analysis module includes: A feature extraction unit, configured to extract the patient's body feature information from the motion posture and joint behavior video; an identification algorithm unit, configured to analyze and identify the human body characteristic information to determine the patient's physical condition and movements; a data transmission unit, configured to transmit the patient's physical state and movements to a storage unit; A storage unit is used to store the patient's physical status and movements for information sharing.

4. The human body posture dynamic intelligent recognition system according to claim 3, characterized in that: The feature extraction unit, when used to extract the patient's body feature information from the motion posture and joint behavior video, includes: Determining key video frames of the patient's motion in the motion posture and joint behavior video, and preprocessing the key video frames, wherein the preprocessing includes grayscale processing, histogram equalization processing, and denoising processing; The preprocessed key video frames are input into the human feature recognition model to determine the patient's human feature information.

5. The human body posture dynamic intelligent recognition system according to claim 4, characterized in that: The grayscale processing is performed according to the following formula: g(x, y) = a × f(x, y) + b; Wherein, f(x, y) represents the grayscale value of the key video frame, g(x, y) represents the grayscale value of the key video frame after grayscale processing, and a and b represent constants.

6. The human body posture dynamic intelligent recognition system according to claim 5, characterized in that: The histogram equalization processing step includes: Determine the grayscale histogram h(rk) of the key video frame after grayscale processing, where rk represents the grayscale level; Calculate the cumulative distribution function sk of the key video frame after grayscale processing, the cumulative distribution function n(j) is the number of pixels with gray level j, and N is the total number of pixels in the key video frame; A new grayscale value gx(x, y) is obtained according to the cumulative distribution function sk, gx(x, y)=INT[(L-1)×sk+0.5], where L represents the grayscale level.

7. The human body posture dynamic intelligent recognition system according to claim 6, characterized in that: The denoising step includes: Mean filtering is used for denoising, specifically: G(x, y) represents the grayscale value of the key video frame after denoising.

8. The human body posture dynamic intelligent recognition system according to claim 1, characterized in that: The joint angle assessment module includes: Angle analysis unit, used to analyze the patient's physical condition and movements and determine the patient's joint movement angle; an angle comparison unit, configured to determine a standard joint motion angle of a standard joint behavior, compare the patient's joint motion angle with the standard joint motion angle, and determine a joint assessment result of the patient; an error calibration unit, used to calibrate the patient's joint assessment results to determine the patient's joint health; The assessment output unit is used to display the patient's joint assessment results and joint health status.

9. The human body posture dynamic intelligent recognition system according to claim 8, characterized in that: The angle comparison unit is used to determine a standard joint motion angle of a standard joint behavior, and compares the joint motion angle of the patient with the standard joint motion angle to determine a joint assessment result of the patient, including: Determine the standard joint motion angle based on the standard joint behavior; Comparing the patient's joint motion angle with the standard joint motion angle, and if the patient's joint motion angle is greater than or equal to the standard joint motion angle, determining that the patient's joint has recovered well; If the patient's joint motion angle is less than the standard joint motion angle, it is determined that the patient's joint has not recovered.

10. The human body posture dynamic intelligent recognition system according to claim 9, characterized in that: The error calibration unit, when used to calibrate the patient's joint assessment results to determine the patient's joint health, includes: When it is determined that the patient's joint has recovered well, the patient's joint health level is determined to be a first health level L1; When it is determined that the patient's joint has not recovered, determining the angle difference between the patient's joint motion angle and the standard joint motion angle, and determining the patient's joint health level based on the angle difference; Wherein, a first angle difference and a second angle difference are preset, and the first angle difference is smaller than the second angle difference; If the angle difference is less than the first angle difference, determining that the patient's joint health level is a second health level L2; If the angle difference is greater than or equal to the first angle difference, and the angle difference is less than the second angle difference, then the patient's joint health level is determined to be a third health level L3; If the angle difference is greater than or equal to the second angle difference, it is determined that the patient's joint health level is a fourth health level L4, and L1>L2>L3>L4.

Images