Posture detection method and device, storage medium and terminal equipment

Abstract

The invention relates to the technical field of medical equipment, and discloses a posture detection method, a storage medium, a device and terminal equipment, and the method comprises the steps: obtaining posture data of a to-be-detected user, the posture data being acquired by a plurality of camera devices arranged around the to-be-detected user and a plurality of limb movement acquisition sensors; inputting the posture data into a trained neural network posture detection model to obtain posture detection information; scoring the posture detection information through pre-stored posture standard information to obtain a posture detection result of the to-be-detected user; and sending the posture detection result and the corresponding posture correction measure information to a display device for display. A plurality of limb movement acquisition devices respectively worn on each limb and trunk of the user to be detected can accurately acquire limb and joint data of the user to be detected, and can accurately reflect smooth values, symmetric values and the like of limb movements of the user. And the neural network posture detection model obtains accurate posture detection information.

Description

Technical Field

[0001] The present invention relates to the technical field of medical equipment, and in particular to a posture detection method, storage medium, device and terminal equipment. Background Art

[0002] Posture testing is a method of evaluating human posture and form, aiming to identify potential body imbalances, muscle tension or bone alignment problems. Poor posture can lead to muscle imbalances and increased joint pressure, which can cause pain and chronic injuries. For example, long-term rounded shoulders and hunched back posture can cause tension in the neck and back muscles, making cervical spondylosis and low back pain more likely. Correct posture makes people look more upright and confident, which helps to improve one's image and temperament.

[0003] In the medical field, doctors can detect potential musculoskeletal problems through posture testing, providing a reference for the diagnosis and treatment of diseases. For example, for patients with scoliosis, early posture testing can help develop appropriate treatment plans. However, in the medical field, medical staff usually observe the standing, walking, sitting and other postures of the tested person with the naked eye to preliminarily determine whether there are common posture problems such as forward head tilt, round shoulders, hunchback, forward or backward pelvic tilt. For example, forward head tilt means the head is in front of the shoulders, and round shoulders mean the shoulders are bent forward. Then use tools such as tape measure and angle measuring instrument to measure various parts of the body. For example, measure the height difference between the two shoulders to determine whether there are high and low shoulders; measure the curvature angle of the spine to assess whether there is scoliosis. Finally, let the tested person perform specific movements, such as squatting, bending over, turning around, etc., to observe the smoothness and symmetry of the movements. The patient is tested through the above process. This detection method relies heavily on the personal experience of doctors. Some inexperienced doctors are prone to misjudgment and misdiagnosis, and cannot provide accurate medical measures to patients.

[0004] Therefore, the above-mentioned technical defects need to be changed urgently. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a posture detection method, storage medium, device and terminal device, which are intended to accurately collect the limb and joint data of the user to be detected, and accurately reflect the smoothness value and symmetry value of the user's limb movements. The neural network posture detection model can obtain accurate posture detection information to avoid misjudgment and misdiagnosis by the posture detection device.

[0006] In order to solve the above technical problems, the first aspect of the embodiment of the present application provides a posture detection method, the method comprising:

[0007] Acquire body posture data of the user to be detected, wherein the body posture data is collected by a plurality of camera devices arranged around the user to be detected and a plurality of body motion collection sensors, and the plurality of body motion collection sensors are respectively worn on the limbs and torso of the user to be detected;

[0008] Inputting the posture data into a trained neural network posture detection model to obtain posture detection information, wherein the neural network posture detection model is trained by multiple posture training data sets, each posture training data set including: training posture data and marking information for marking the posture detection information;

[0009] Scoring the posture detection information by using the pre-stored posture standard information to obtain the posture detection result of the user to be detected;

[0010] The posture detection results and the corresponding posture correction measures information are sent to the display device for display.

[0011] The posture detection method, wherein the step of obtaining the posture data of the user to be detected comprises:

[0012] Using several video cameras to obtain video information of the user to be detected in various postures and movements;

[0013] Acquire limb and joint data of the user to be detected through a number of limb motion collection sensors;

[0014] The limb and joint data are fitted to the video information to obtain the posture data.

[0015] The posture detection method, wherein the step of acquiring limb and joint data of a user to be detected by using a plurality of limb motion acquisition sensors, comprises:

[0016] Initialize several body motion collection sensors;

[0017] Acquire dynamic data from a number of body motion collection sensors, where the dynamic data is collected based on the user to be detected performing standard movements;

[0018] Based on the three-dimensional human body model, the dynamic data is analyzed and calibrated to obtain the limb and joint data of the user to be detected.

[0019] The posture detection method includes a step of analyzing and calibrating dynamic data based on a three-dimensional human body model to obtain limb and joint data of the user to be detected, wherein the limb and joint data include limb reference points, spacing between limbs, connection angles between limbs, and positional relationships between limbs.

[0020] The posture detection method, wherein the step of inputting posture data into a trained neural network posture detection model to obtain posture detection information, comprises:

[0021] Splitting the body data into video frames to obtain a number of video frames, each of which contains corresponding limb and joint data;

[0022] Based on the limb and joint data, several limb reference points of the user to be detected are obtained;

[0023] Compare several limb reference points through a neural network posture detection model and score the video frames;

[0024] The video frame with the highest score is selected as the body detection information.

[0025] The posture detection method, wherein the posture detection information is scored by using pre-stored posture standard information to obtain the posture detection result of the user to be detected, the posture detection result includes:

[0026] Any one or more of the round shoulder detection results, hunchback detection results, head protrusion detection results, knee hyperextension detection results, lumbar excessive flexion detection results, pelvic anterior tilt detection results, pelvic posterior tilt detection results, uneven shoulders detection results, O-shaped legs detection results, X-shaped legs detection results, scoliosis detection results, and pelvic lateral tilt detection results.

[0027] The posture detection method, wherein after the step of sending the posture detection result and the corresponding posture correction measure information to the display device for display, comprises:

[0028] The posture detection result, the corresponding posture correction measure information and the identity information of the user to be detected are sent to the printing device.

[0029] A second aspect of an embodiment of the present application provides a computer-readable storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the steps in any of the above posture detection methods.

[0030] A third aspect of the present application provides a posture detection device, which includes:

[0031] An acquisition module is used to acquire body posture data of the user to be detected, where the body posture data is acquired by a plurality of camera devices arranged around the user to be detected and a plurality of body motion acquisition sensors, where the plurality of body motion acquisition sensors are respectively worn on the limbs and torso of the user to be detected;

[0032] A posture detection module is used to input posture data into a trained neural network posture detection model to obtain posture detection information. The neural network posture detection model is trained by multiple posture training data sets, and each posture training data set includes: training posture data and marking information for marking posture detection information;

[0033] A posture scoring module is used to score the posture detection information according to the pre-stored posture standard information to obtain the posture detection result of the user to be detected;

[0034] The output module is used to send the posture detection results and the corresponding posture correction measures information to the display device for display.

[0035] A fourth aspect of the embodiments of the present application provides a terminal device, comprising: a processor, a memory, and a communication bus; the memory stores a computer-readable program that can be executed by the processor;

[0036] The communication bus realizes the connection and communication between the processor and the memory;

[0037] When the processor executes the computer-readable program, the steps in any of the above-mentioned posture detection methods are implemented.

[0038] Beneficial effects: Compared with the prior art, the present invention provides a posture detection method, storage medium, device and terminal device, wherein the method comprises: obtaining posture data of a user to be detected, the posture data is collected by a plurality of camera devices and a plurality of limb motion acquisition sensors arranged around the user to be detected, the plurality of limb motion acquisition sensors are respectively worn on the limbs and trunk of the user to be detected; inputting the posture data into a trained neural network posture detection model to obtain posture detection information, the neural network posture detection model is trained by a plurality of posture training data sets, each posture training data set comprises: training posture data and marking information for marking posture detection information; scoring the posture detection information by pre-stored posture standard information to obtain the posture detection result of the user to be detected; sending the posture detection result and the corresponding posture correction measure information to a display device for display. The plurality of limb motion acquisition devices respectively worn on the limbs and trunk of the user can accurately collect the limb and joint data of the user to be detected, and can accurately reflect the smoothness value, symmetry value, etc. of the user's limb motion. The neural network posture detection model obtains accurate posture detection information. Fitting limb and joint data to the video information collected by the camera can produce a more intuitive display effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A flow chart of a posture detection method provided by the present invention;

[0040] Figure 2 for Figure 1 Schematic diagram of the process of step S10;

[0041] Figure 3 for Figure 1 Schematic diagram of the process of step S12;

[0042] Figure 4 for Figure 3 Schematic diagram of the process of step S20

[0043] Figure 5 A schematic diagram of the structure of a posture detection device provided by the present invention;

[0044] Figure 6 This is a structural principle diagram of the terminal device provided by the present invention. DETAILED DESCRIPTION

[0045] The present invention provides a posture detection method, storage medium, device and terminal equipment. In order to make the purpose, technical solution and effect of the present invention clearer and more specific, the present invention is further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0046] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when the present application scheme states that an element is "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may also be an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0047] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as herein.

[0048] The invention is further described below by describing embodiments in conjunction with the accompanying drawings.

[0049] First of all, it should be noted that in the medical field, medical staff usually observe the standing, walking, sitting and other postures of the person being tested with the naked eye to preliminarily determine whether there are common posture problems such as forward head tilt, round shoulders, hunchback, forward or backward pelvic tilt. For example, forward head tilt means the head is in front of the shoulders, and round shoulders mean the shoulders are bent forward. Then use tools such as tape measure and angle measuring instrument to measure various parts of the body. For example, measure the height difference between the shoulders on both sides to determine whether there are high and low shoulders; measure the curvature angle of the spine to assess whether there is scoliosis. Finally, let the person being tested perform specific movements, such as squatting, bending over, turning around, etc., to observe the smoothness and symmetry of the movements. The patient is tested through the above process. This detection method relies heavily on the personal experience of doctors. Some inexperienced doctors are prone to misjudgment and misdiagnosis, and cannot provide accurate medical measures to patients.

[0050] To this end, this solution provides a posture detection method, storage medium, device and terminal equipment. The present invention is a posture detection system that integrates intelligence, dataization and informationization. In the medical field, it can automatically perform posture detection on patients and give corresponding posture correction treatment advice. Effectively improve the efficiency and quality of posture detection. In addition, the posture detection system is not only for the medical field. It can also play a role in some family scenes and sports training scenes.

[0051] The device of the present application mainly includes several somatosensory cameras, several limb motion acquisition sensors, computers (hosts), interactive display controllers, and audio equipment. Among them, several somatosensory cameras are set around the user to be detected, and can capture the patient's body posture video images at different angles, which is convenient for establishing a 3D human body model. In addition, several limb motion acquisition sensors are worn on the limbs and torso of the user to be detected, and the data of the limbs and joints of the user to be detected can be captured through the limb motion acquisition sensors. Through these data, parameters such as angles, lengths, and heights of multiple parts of the human body can be measured to comprehensively evaluate the posture. For example, the curvature angle of the spine, the inclination angle of the shoulder and neck, the forward and backward tilt angle of the pelvis, etc. can be measured. The interactive display controller and the audio can establish a channel for human-computer interaction. Interactive information can be displayed through the interactive display controller. Including guiding the user to be detected to perform motion detection. It is convenient for the system to capture the posture of the user to be detected, and to capture the limb and joint data of the user to be detected. The somatosensory camera can also perform image functions such as face recognition and motion recognition on the user to be detected, and establish a personal profile.

[0052] Among them, the execution subject of the posture detection method provided by this scheme can be a server side deployed with a posture detection module, wherein the posture detection module is used to perform posture detection and scoring on the user to be detected and obtain corresponding guidance opinions. It can be understood that the execution subject of this embodiment can be an intelligent terminal connected with a posture detection module such as a smart phone, a smart bracelet or a server host. For example, the host obtains the posture data of the user to be detected, and the posture data is collected by several cameras set around the user to be detected and several limb motion acquisition sensors, and the several limb motion acquisition sensors are respectively worn on the limbs and torso of the user to be detected; the host inputs the posture data into the trained neural network posture detection model to obtain posture detection information, and the neural network posture detection model is trained by multiple posture training data sets, and each posture training data set includes: training posture data and marking information for marking posture detection information; the host scores the posture detection information through the pre-stored posture standard information to obtain the posture detection result of the user to be detected; the host sends the posture detection result and the corresponding posture correction measure information to the display device for display.

[0053] It should be noted that the above application scenarios are only shown to facilitate understanding of the present invention, and the embodiments of the present invention are not limited in this respect. On the contrary, the embodiments of the present invention can be applied to any applicable scenarios.

[0054] In order to further illustrate the content of the invention, embodiments are described in detail below with reference to the accompanying drawings.

[0055] The posture detection method provided in this embodiment is as follows: Figure 1 As shown, the method includes:

[0056] Step S10, obtaining body posture data of the user to be detected, wherein the body posture data is collected by a plurality of camera devices arranged around the user to be detected and a plurality of body motion collection sensors, and the plurality of body motion collection sensors are respectively worn on the limbs and torso of the user to be detected;

[0057] Specifically, several somatosensory cameras are set up around the user to be tested, which can capture the patient's body video images at different angles to facilitate the establishment of a 3D human body model. In addition, several limb motion acquisition sensors are worn on the limbs and torso of the user to be tested. The limb motion acquisition sensors can capture the data of the limbs and joints of the user to be tested. These data can measure the angles, lengths, heights and other parameters of multiple parts of the human body to comprehensively evaluate the body shape. For example, the curvature angle of the spine, the inclination angle of the shoulder and neck, the forward and backward tilt angle of the pelvis, etc. can be measured.

[0058] In order to more quickly initialize the limb motion acquisition sensor and quickly establish the 3D body model of the user to be detected, it is necessary to guide the user to be detected to perform specific movements and postures. By allowing the user to be detected to perform specific movements and postures, the body posture of the user to be detected can be more comprehensively displayed, exposing any problems with the user's body posture.

[0059] To this end, the present application scheme is also provided with interactive devices such as an interactive display controller and audio. These interactive devices can establish a channel for human-computer interaction. Interactive information can be displayed through the interactive display controller. This includes displaying on the display an image of actions and postures for guiding the user to be detected, issuing guiding voices through voice, etc. This guides the user to be detected to perform action detection. It is convenient for the system to capture the posture of the user to be detected, as well as to capture the limb and joint data of the user to be detected. Furthermore, the somatosensory camera can also perform image functions such as face recognition and action recognition on the user to be detected, and establish a personal profile.

[0060] Step S20: inputting the posture data into a trained neural network posture detection model to obtain posture detection information, wherein the neural network posture detection model is trained by a plurality of posture training data sets, each posture training data set including: training posture data and marking information for marking the posture detection information;

[0061] Specifically, the present application scheme uses artificial neural network technology to detect human posture. Neural networks can automatically learn the characteristics of human posture and can process complex image and video data. Compared with traditional posture detection methods, neural network posture detection models have higher accuracy. Since neural networks can process data quickly, real-time posture detection can be achieved. This is very important for application scenarios that require real-time monitoring of human posture (such as fitness training, rehabilitation treatment, etc.). Neural networks can adapt to different lighting conditions, background environments, and changes in human posture. This makes it have good performance in various practical application scenarios. The performance and detection capabilities of neural networks can be continuously improved by increasing training data and adjusting the network structure. At the same time, it can also be combined with other technologies (such as sensor fusion, cloud computing, etc.) to achieve a wider range of applications.

[0062] This solution collects image or video data of the user to be detected through the body sensing camera and the body motion acquisition sensor in step S10. These data may include different postures, movements and performances of the human body in different scenes. Then, the collected data is processed using computer vision technology to extract features related to body posture. These features may include the position, angle, and body contour of the human joints.

[0063] It should be noted that before using the neural network posture detection model, it is necessary to first use a large amount of posture training data sets to train the neural network. Specifically: the features extracted from the posture training data set are used as input, and the human body posture categories (such as normal, hunchback, round shoulders, etc.) are used as output to train the neural network. During the training process, the neural network continuously adjusts its own weights and biases to make the output results as close to the actual posture category as possible.

[0064] When performing actual posture detection, the newly collected human body data is input into the trained neural network, and the network will output the corresponding posture category, thereby realizing posture detection.

[0065] Step S30: Score the posture detection information using pre-stored posture standard information to obtain a posture detection result of the user to be detected;

[0066] Specifically, in step S20, the posture data is input into the trained neural network posture detection model to obtain posture detection information. The neural network posture detection model is trained by multiple posture training data sets, and each posture training data set includes: training posture data and marking information for marking posture detection information. The posture detection information at this time is the user's real posture information. The subsequent steps also need to score the posture detection information to obtain the final detection result. Therefore, in this embodiment, the posture detection information is scored by pre-stored posture standard information to obtain the posture detection result of the user to be detected.

[0067] Step S40: Send the posture detection result and the corresponding posture correction measure information to a display device for display.

[0068] Specifically, the system is preset with a posture correction measure information library, which stores posture correction measure data corresponding to different posture problems. When a patient's posture problem is detected, the corresponding posture correction measure data in the posture correction measure information library is obtained accordingly, making it convenient for doctors and users to master posture correction measures.

[0069] Further, such as Figure 2 As shown, the step of obtaining the body posture data of the user to be detected includes:

[0070] Step S11, obtaining video information of the user to be detected in various postures and movements through a plurality of camera devices;

[0071] Specifically, several cameras shoot the user to be detected in different positions, and during the shooting, the user is guided to perform action and posture demonstration through the interactive device, so that the camera can fully obtain the body posture video data of the patient to be detected.

[0072] Step S12, acquiring limb and joint data of the user to be detected through a plurality of limb motion collection sensors;

[0073] Specifically, in step S11, while collecting video information, the limb motion collection sensors worn on the limbs of the user to be detected synchronously collect the limb and joint data of the user to be detected.

[0074] Step S13: Fit the limb and joint data to the video information to obtain body posture data.

[0075] Specifically, fitting limb and joint data into video information can more intuitively display the posture problems of the user to be detected.

[0076] Further, such as Figure 3 As shown, the steps of acquiring limb and joint data of the user to be detected through a plurality of limb motion acquisition sensors include:

[0077] Step S121, initializing a plurality of body motion acquisition sensors;

[0078] Specifically, in some embodiments, the size and shape of several limb motion acquisition sensors are configured to be consistent. In particular, the size and shape of several limb motion acquisition sensors worn on the patient's limbs are configured to be consistent, which is more versatile. There is no need to develop them specifically. When worn and used, it only needs to be fixed to the approximate parts of the limbs and body of the user to be detected through a fixing. A complete and accurate posture model can be obtained through calculation and interpolation compensation later.

[0079] Therefore, before each posture detection, several limb motion acquisition sensors need to be initialized. Calibration is performed based on the relative positions between the limb motion acquisition sensors. The system is allowed to recalibrate the corresponding position, angle, etc. of each limb motion acquisition sensor.

[0080] Step S122, acquiring dynamic data of a plurality of body motion collection sensors, the dynamic data being collected based on the user to be detected under standard motions;

[0081] Specifically, dynamic data can allow the system to more accurately calculate the position and angle of each limb movement collection sensor. These dynamic data require the user to perform standard movements so that the system can clearly understand what the user is doing and perform targeted calculations.

[0082] Step S123: Analyze and calibrate the dynamic data based on the three-dimensional human body model to obtain limb and joint data of the user to be detected.

[0083] Furthermore, based on the three-dimensional human body model, the dynamic data is analyzed and calibrated to obtain the limb and joint data of the user to be detected. The limb and joint data include limb reference points, distances between limbs, connection angles between limbs, and positional relationships between limbs.

[0084] Further, such as Figure 4 As shown, the steps of inputting posture data into a trained neural network posture detection model to obtain posture detection information include:

[0085] Step S21, splitting the body data into video frames to obtain a plurality of video frames, each of which contains corresponding limb and joint data;

[0086] Specifically, in order to accurately obtain the real body posture of the user to be detected and reflect the problem of body posture, it is necessary to obtain a standard video frame that can be used to calculate and compare the standard body posture in the body posture data. Therefore, it is necessary to split the body posture data into video frames.

[0087] Step S22: acquiring a plurality of limb reference points of the user to be detected based on the limb and joint data;

[0088] Specifically, when screening and acquiring standards that can be used to calculate and compare video frames with standard postures, it is necessary to first obtain several limb reference points of the user to be detected on the video frame, and determine whether the current video frame matches the video frame with the standard posture through the several limb reference points.

[0089] Step S23, comparing several limb reference points through a neural network posture detection model, and scoring the video frames;

[0090] Step S24: Select the video frame with the highest score as body posture detection information.

[0091] Furthermore, in the step of scoring the posture detection information by using the pre-stored posture standard information to obtain the posture detection result of the user to be detected, the posture detection result includes:

[0092] Any one or more of the round shoulder detection results, hunchback detection results, head protrusion detection results, knee hyperextension detection results, lumbar excessive flexion detection results, pelvic anterior tilt detection results, pelvic posterior tilt detection results, uneven shoulders detection results, O-shaped legs detection results, X-shaped legs detection results, scoliosis detection results, and pelvic lateral tilt detection results.

[0093] Furthermore, after the step of sending the posture detection result and the corresponding posture correction measure information to the display device for display, the method further includes:

[0094] The posture detection result, the corresponding posture correction measure information and the identity information of the user to be detected are sent to the printing device.

[0095] In summary, if Figure 1 As shown, this embodiment provides a posture detection method, wherein the method includes: obtaining posture data of a user to be detected, the posture data is collected by a plurality of camera devices and a plurality of limb motion acquisition sensors arranged around the user to be detected, and the plurality of limb motion acquisition sensors are respectively worn on the limbs and trunk of the user to be detected; inputting the posture data into a trained neural network posture detection model to obtain posture detection information, the neural network posture detection model is trained by a plurality of posture training data sets, and each posture training data set includes: training posture data and marking information for marking posture detection information; scoring the posture detection information by pre-stored posture standard information to obtain the posture detection result of the user to be detected; sending the posture detection result and the corresponding posture correction measure information to a display device for display. The plurality of limb motion acquisition devices respectively worn on the limbs and trunk of the user can accurately collect the limb and joint data of the user to be detected, and can accurately reflect the smoothness value, symmetry value, etc. of the user's limb movements. The neural network posture detection model obtains accurate posture detection information. Fitting limb and joint data to the video information collected by the camera can produce a more intuitive display effect.

[0096] In order to better implement the above method, the embodiment of the present application also provides a posture detection device 100, which can be integrated in an electronic device, and the electronic device can be a terminal, a server, a personal computer, etc. For example, in this embodiment, the device may include: 101 acquisition module, 102 posture detection module, 103 posture scoring module and 104 output module, which are as follows (for example Figure 5 ):

[0097] 101 an acquisition module, used to acquire body posture data of the user to be detected, the body posture data is acquired by a plurality of camera devices arranged around the user to be detected and a plurality of body motion acquisition sensors, the plurality of body motion acquisition sensors are respectively worn on the limbs and torso of the user to be detected;

[0098] 102 posture detection module, used for inputting posture data into a trained neural network posture detection model to obtain posture detection information, the neural network posture detection model is trained by a plurality of posture training data sets, each posture training data set includes: training posture data and marking information for marking posture detection information;

[0099] 103 a posture scoring module, used to score the posture detection information according to the pre-stored posture standard information, and obtain the posture detection result of the user to be detected;

[0100] 104 output module, used to send the posture detection result and the corresponding posture correction measure information to the display device for display.

[0101] In some embodiments, a posture detection device 100 includes an acquisition module 101, a posture detection module 102, a posture scoring module 103 and an output module 104, wherein the acquisition module 101 acquires the posture data of the user to be detected, the posture data is collected by a number of cameras arranged around the user to be detected and a number of limb motion acquisition sensors, and the several limb motion acquisition sensors are respectively worn on each limb and torso of the user to be detected; the posture detection module 102 inputs the posture data into a trained neural network posture detection model to obtain posture detection information, the neural network posture detection model is trained by multiple posture training data sets, and each posture training data set includes: training posture data and marking information for marking the posture detection information; the posture scoring module 103 scores the posture detection information by using pre-stored posture standard information to obtain the posture detection result of the user to be detected; the output module 104 sends the posture detection result and the corresponding posture corrective measure information to a display device for display.

[0102] In specific implementation, the above units can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above units can refer to the previous method embodiments, which will not be repeated here.

[0103] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0104] Based on the above posture detection method, this embodiment provides a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the posture detection method of the above embodiment. Figure 1 Steps S10 to S40 of the method, Figure 2 Steps S11 to S13 of the method, Figure 3 Steps S121 to S123 of the method, Figure 4 The method steps S21 to S24 in the embodiment are as follows:

[0105] Acquire body posture data of the user to be detected, where the body posture data is collected by a number of cameras arranged around the user to be detected and a number of body motion collection sensors, where the several body motion collection sensors are respectively worn on the limbs and torso of the user to be detected;

[0106] Inputting the posture data into a trained neural network posture detection model to obtain posture detection information, wherein the neural network posture detection model is trained by multiple posture training data sets, each posture training data set including: training posture data and marking information for marking the posture detection information;

[0107] Scoring the posture detection information by using the pre-stored posture standard information to obtain the posture detection result of the user to be detected;

[0108] The posture detection results and the corresponding posture correction measures information are sent to the display device for display.

[0109] In some embodiments, the step of obtaining the body posture data of the user to be detected includes:

[0110] Using several video cameras to obtain video information of the user to be detected in various postures and movements;

[0111] Acquire limb and joint data of the user to be detected through a number of limb motion collection sensors;

[0112] The limb and joint data are fitted to the video information to obtain the posture data.

[0113] In some embodiments, the step of acquiring limb and joint data of the user to be detected through a plurality of limb motion collection sensors includes:

[0114] Initialize several body motion collection sensors;

[0115] Acquire dynamic data from a number of body motion collection sensors, where the dynamic data is collected based on the user to be detected performing standard movements;

[0116] Based on the three-dimensional human body model, the dynamic data is analyzed and calibrated to obtain the limb and joint data of the user to be detected.

[0117] In some embodiments, the step of identifying posture data by a neural network posture detection model to obtain posture detection information includes:

[0118] Splitting the body data into video frames to obtain a number of video frames, each of which contains corresponding limb and joint data;

[0119] Based on the limb and joint data, several limb reference points of the user to be detected are obtained;

[0120] Compare several limb reference points through a neural network posture detection model and score the video frames;

[0121] The video frame with the highest score is selected as the body detection information.

[0122] Based on the above posture detection method, the present invention also provides a terminal device, such as Figure 6 As shown, it includes at least one processor (processor) 20; display screen 21; and memory (memory) 22, and may also include a communications interface (Communications Interface) 23 and a bus 24. Among them, the processor 20, the display screen 21, the memory 22 and the communication interface 23 can communicate with each other through the bus 24. The display screen 21 is configured to display a preset user guide interface in the initial setting mode. The communication interface 23 can transmit information. The processor 20 can call the logic instructions in the memory 22 to execute the method in the above embodiment.

[0123] In addition, the logic instructions in the memory 22 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product.

[0124] The memory 22 is a computer-readable storage medium that can be configured to store software programs, computer executable programs, such as program instructions or modules corresponding to the methods in the embodiments of the present disclosure. The processor 20 executes functional applications and data processing by running the software programs, instructions or modules stored in the memory 22, that is, implementing the methods in the above embodiments.

[0125] The memory 22 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the terminal device, etc. In addition, the memory 22 may include a high-speed random access memory and may also include a non-volatile memory. For example, a variety of media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, may also be a transient storage medium.

[0126] In addition, the specific process of loading and executing the plurality of instructions in the storage medium and the processor in the mobile terminal has been described in detail in the method above, and will not be described here one by one.

[0127] In summary, compared with the prior art, the present invention has the following beneficial effects: a posture detection method, storage medium, device and terminal device, wherein the method comprises: obtaining posture data of a user to be detected, the posture data is collected by a plurality of camera devices and a plurality of limb motion acquisition sensors arranged around the user to be detected, and the plurality of limb motion acquisition sensors are respectively worn on the limbs and trunk of the user to be detected; inputting the posture data into a trained neural network posture detection model to obtain posture detection information, the neural network posture detection model is trained by a plurality of posture training data sets, and each posture training data set comprises: training posture data and marking information for marking posture detection information; scoring the posture detection information by pre-stored posture standard information to obtain the posture detection result of the user to be detected; sending the posture detection result and the corresponding posture correction measure information to a display device for display. The plurality of limb motion acquisition devices respectively worn on the limbs and trunk of the user can accurately collect the limb and joint data of the user to be detected, and can accurately reflect the smoothness value, symmetry value, etc. of the user's limb motion. The neural network posture detection model obtains accurate posture detection information. Fitting limb and joint data to the video information collected by the camera can produce a more intuitive display effect.

[0128] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A posture detection method, characterized in that: The method comprises: Acquire body posture data of the user to be detected, wherein the body posture data is collected by a plurality of camera devices arranged around the user to be detected and a plurality of body motion collection sensors, wherein the plurality of body motion collection sensors are respectively worn on the limbs and torso of the user to be detected; Inputting the posture data into a trained neural network posture detection model to obtain posture detection information, wherein the neural network posture detection model is trained by a plurality of posture training data sets, each of which includes: training posture data and marking information for marking the posture detection information; Scoring the posture detection information by using pre-stored posture standard information to obtain a posture detection result of the user to be detected; The posture detection result and the corresponding posture correction measure information are sent to a display device for display.

2. The posture detection method according to claim 1, characterized in that: The step of obtaining the body posture data of the user to be detected includes: Acquire video information of the user to be detected in various postures and movements through a plurality of the above-mentioned camera devices; Acquire limb and joint data of the user to be detected through a plurality of limb motion acquisition sensors; The limb and joint data are fitted into the video information to obtain the body posture data.

3. The posture detection method according to claim 2, characterized in that: The step of acquiring limb and joint data of the user to be detected by using a plurality of limb motion acquisition sensors comprises: Initializing a plurality of the body motion acquisition sensors; Acquire dynamic data of a plurality of the body motion acquisition sensors, wherein the dynamic data is acquired based on the acquisition of the user to be detected under standard motions; Based on the three-dimensional human body model, the dynamic data is analyzed and calibrated to obtain the limb and joint data of the user to be detected.

4. The posture detection method according to claim 3, characterized in that: In the step of analyzing and calibrating the dynamic data based on the three-dimensional human body model to obtain the limb and joint data of the user to be detected, the limb and joint data include limb reference points, distances between limbs, connection angles between limbs, and positional relationships between limbs.

5. The posture detection method according to claim 4, characterized in that: The step of inputting the posture data into a trained neural network posture detection model to obtain posture detection information comprises: Performing video frame splitting on the body posture data to obtain a plurality of video frames, each of which contains corresponding limb and joint data; Based on the limb and joint data, obtaining a number of limb reference points of the user to be detected; Comparing a plurality of the limb reference points through a neural network posture detection model and scoring the video frame; The video frame with the highest score is selected as the body posture detection information.

6. The posture detection method according to claim 1, characterized in that: In the step of scoring the posture detection information by using the pre-stored posture standard information to obtain the posture detection result of the user to be detected, the posture detection result includes: Any one or more of the round shoulder detection results, hunchback detection results, head protrusion detection results, knee hyperextension detection results, lumbar excessive flexion detection results, pelvic anterior tilt detection results, pelvic posterior tilt detection results, uneven shoulders detection results, O-shaped legs detection results, X-shaped legs detection results, scoliosis detection results, and pelvic lateral tilt detection results.

7. The posture detection method according to claim 1, characterized in that: After the step of sending the posture detection result and the corresponding posture correction measure information to the display device for display, the method further comprises: The posture detection result, the corresponding posture correction measure information and the identity information of the user to be detected are sent to a printing device.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the posture detection method according to any one of claims 1 to 7.

9. A posture detection device, characterized in that: include: An acquisition module is used to acquire body posture data of the user to be detected, wherein the body posture data is acquired by a plurality of camera devices and a plurality of body motion acquisition sensors arranged around the user to be detected, wherein the plurality of body motion acquisition sensors are respectively worn on the limbs and torso of the user to be detected; A posture detection module, used for inputting the posture data into a trained neural network posture detection model to obtain posture detection information, wherein the neural network posture detection model is trained by a plurality of posture training data sets, each of which includes: training posture data and marking information for marking the posture detection information; A posture scoring module, used to score the posture detection information according to the pre-stored posture standard information, and obtain the posture detection result of the user to be detected; The output module is used to send the posture detection result and the corresponding posture correction measure information to the display device for display.

10. A terminal device, characterized in that: include: Processor, memory and communication bus; The memory stores a computer-readable program executable by the processor; The communication bus realizes the connection and communication between the processor and the memory; When the processor executes the computer-readable program, the steps in the posture detection method according to any one of claims 1 to 7 are implemented.

Images