A method and device for capturing human posture balance based on intelligent wearable device

Abstract

The present invention discloses a method and device for capturing human posture balance based on a smart wearable device, wherein the method comprises: obtaining a first training posture of a first target user; obtaining a first initial training action; collecting skeletal posture data for the first initial training action to generate a first skeletal posture collection; constructing a human posture balance correction database; inputting the first initial training action into the human posture balance correction database for training screening to obtain first preset skeletal posture information; judging whether the first skeletal posture collection meets the first preset skeletal posture information; if not, generating a first compensation instruction, and performing action compensation on the first initial training action. The method solves the technical problem in the prior art that it is impossible to collect all-round dynamic data of the user's balance rehabilitation training action, so that the user's rehabilitation action cannot meet the expected standard, thereby affecting the user's training efficiency.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to a method and device for capturing human posture balance based on an intelligent wearable device. Background Art

[0002] As an important physiological function of the human body, balance ability plays an irreplaceable role in maintaining normal posture and activities of the human body. The evaluation of balance ability is an important basis for diagnosis, treatment and rehabilitation training of patients with balance disorders.

[0003] However, the inventors of this application found that the above technology has at least the following technical problems in the process of implementing the technical solution of the invention in the embodiment of this application:

[0004] There is a technical problem in the prior art that it is impossible to collect all-round dynamic data of the user's balance rehabilitation training movements, so that the user's rehabilitation movements cannot reach the expected standards, thereby affecting the user's training efficiency. Summary of the invention

[0005] In view of the defects in the prior art, the purpose of the embodiments of the present application is to provide a method and device for capturing human posture balance based on a smart wearable device, thereby solving the technical problem in the prior art that it is impossible to collect all-round dynamic data of the user's balance rehabilitation training movements, so that the user's rehabilitation movements cannot reach the expected standard, thereby affecting the user's training efficiency. Through the smart wearable device, the training posture of the target user is traversed and data is collected. By building a human posture balance correction database, the skeletal posture collection corresponding to the actual training movement of the target user is compared with the skeletal posture collection of the expected standard, and then the substandard training movements are compensated, so that the individual training movements of the target user are all in a scientific and standard state, and the target user's human posture balance is fully monitored and captured, and real-time compensation reminders are given for substandard training movements, thereby improving the technical effect of improving the efficiency of the target user's human posture balance training.

[0006] On the one hand, an embodiment of the present application provides a human posture balance capture method based on a smart wearable device, wherein the method is applied to a human posture balance capture system, and the method includes: obtaining a first training posture of a first target user, wherein the first training posture includes a training action collection, and the single training actions in the training action collection are in a coherent state; according to the training action collection, obtaining a first initial training action of the first target user, wherein the training action collection all meets the first demeanor requirement; based on the smart wearable device, performing skeletal posture data collection on the first initial training action of the first target user, and generating a first skeletal posture collection of the first initial training action; constructing a human posture balance correction database, and the human posture balance correction database is included in the human posture balance capture system; inputting the first initial training action into the human posture balance correction database for training screening, and obtaining a first preset skeletal posture information corresponding to the first initial training action; judging whether the first skeletal posture collection meets the first preset skeletal posture information; if the first skeletal posture collection does not meet the first preset skeletal posture information, the human posture balance capture system generates a first compensation instruction to perform action compensation on the first initial training action.

[0007] On the other hand, the present application also provides a human posture balance capture system based on a smart wearable device, wherein the system includes: a first acquisition unit: the first acquisition unit is used to obtain a first training posture of a first target user, wherein the first training posture includes a collection of training movements, and the individual training movements in the collection of training movements are in a coherent state; a second acquisition unit: the second acquisition unit is used to obtain a first initial training movement of the first target user based on the collection of training movements, wherein the collection of training movements all meet the first posture requirement; a first acquisition unit: the first acquisition unit is used to perform skeletal posture data collection on the first initial training movement of the first target user based on the smart wearable device, and generate a first skeletal posture collection of the first initial training movement. A set; a first construction unit: the first construction unit is used to construct a human body posture balance correction database, and the human body posture balance correction database is included in a human body posture balance capture system; a first input unit: the first input unit is used to input the first initial training action into the human body posture balance correction database for training screening, and obtain the first preset bone posture information corresponding to the first initial training action; a first judgment unit: the first judgment unit is used to judge whether the first bone posture collection meets the first preset bone posture information; a first generation unit: the first generation unit is used for if the first bone posture collection does not meet the first preset bone posture information, the human body posture balance capture system generates a first compensation instruction to perform action compensation on the first initial training action.

[0008] In the third aspect, an embodiment of the present application provides a human body posture balance capture device based on a smart wearable device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the methods described in the first aspect when executing the program.

[0009] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0010] By obtaining a first training posture of a first target user, wherein the first training posture includes a collection of training movements, and the individual training movements in the collection of training movements are in a coherent state; according to the collection of training movements, obtaining a first initial training movement of the first target user, wherein the collection of training movements all meet the first posture requirement; based on a smart wearable device, collecting skeletal posture data of the first initial training movement of the first target user, and generating a first skeletal posture collection of the first initial training movement; constructing a human body posture balance correction database, and the human body posture balance correction database is included in the human body posture balance capture system; inputting the first initial training movement into the human body posture balance correction database for training screening, and obtaining first preset skeletal posture information corresponding to the first initial training movement; judging whether the first skeletal posture collection meets the first preset skeletal posture information; if the first skeletal posture collection does not meet the first preset skeletal posture information, the human body posture balance capture system generates a first compensation instruction to perform movement compensation on the first initial training movement. Through the intelligent wearable device, the training posture data of the target user is traversed and collected. By building a human posture balance correction database, the skeletal posture collection corresponding to the actual training movements of the target user is compared with the expected standard skeletal posture collection, and then the substandard training movements are compensated, so that the individual training movements of the target user are all in a scientific and standard state, achieving the technical effect of all-round monitoring and capturing of the human posture balance of the target user, and providing real-time motion compensation reminders for substandard training movements, thereby improving the efficiency of the human posture balance training of the target user.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0013] Figure 1 A schematic diagram of a flow chart of a method for capturing a human posture balance based on a smart wearable device according to an embodiment of the present application;

[0014] Figure 2 A schematic diagram of a process of performing motion recovery on the first initial training motion in a method for capturing human posture balance based on a smart wearable device according to an embodiment of the present application;

[0015] Figure 3 A schematic diagram of a flow chart of building a human posture balance correction database in a human posture balance capture method based on a smart wearable device according to an embodiment of the present application;

[0016] Figure 4 This is a structural schematic diagram of a human posture balance capture system based on a smart wearable device according to an embodiment of the present application;

[0017] Figure 5 It is a schematic diagram of the structure of an exemplary electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0018] The embodiment of the present application provides a method and device for capturing human posture balance based on a smart wearable device, which solves the technical problem in the prior art that it is impossible to collect all-round dynamic data of the user's balance rehabilitation training movements, so that the user's rehabilitation movements cannot reach the expected standard, thereby affecting the user's training efficiency. Through the smart wearable device, the training posture of the target user is traversed and data is collected. By building a human posture balance correction database, the skeletal posture collection corresponding to the actual training movement of the target user is compared with the skeletal posture collection of the expected standard, and then the training movements that do not meet the standard are compensated, so that the individual training movements of the target user are all in a scientific and standard state, and the target user's human posture balance is fully monitored and captured, and real-time compensation reminders are given for training movements that do not meet the standard, thereby improving the technical effect of the human posture balance training efficiency of the target user.

[0019] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0020] Application Overview

[0021] Since the prior art cannot collect all-round dynamic data of the user's balance rehabilitation training movements, the user's rehabilitation movements cannot meet the expected standards. Under the premise of not affecting the user's rehabilitation training efficiency, the present invention collects data on the training posture of the target user, and provides real-time movement compensation reminders for training movements that do not meet the standards, thereby improving the target user's human posture balance training efficiency.

[0022] In response to the above technical problems, the overall idea of ​​the technical solution provided by this application is as follows:

[0023] The embodiment of the present application provides a human posture balance capture method based on a smart wearable device, characterized in that the method is applied to a human posture balance capture system, and the method includes: obtaining a first training posture of a first target user, wherein the first training posture includes a training action collection, and the single training actions in the training action collection are in a coherent state; obtaining a first initial training action of the first target user according to the training action collection, wherein the training action collection all meets the first demeanor requirement; based on the smart wearable device, performing skeletal posture data collection on the first initial training action of the first target user to generate a first skeletal posture collection of the first initial training action; constructing a human posture balance correction database, and the human posture balance correction database is included in the human posture balance capture system; inputting the first initial training action into the human posture balance correction database for training screening to obtain first preset skeletal posture information corresponding to the first initial training action; judging whether the first skeletal posture collection meets the first preset skeletal posture information; if the first skeletal posture collection does not meet the first preset skeletal posture information, the human posture balance capture system generates a first compensation instruction to perform action compensation on the first initial training action.

[0024] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0025] Embodiment 1

[0026] like Figure 1 As shown, an embodiment of the present application provides a method for capturing a human posture balance based on a smart wearable device, wherein the method is applied to a human posture balance capturing system, and the method includes:

[0027] Step S100: obtaining a first training posture of a first target user, wherein the first training posture includes a training action collection, and individual training actions in the training action collection are in a coherent state;

[0028] Step S200: obtaining a first initial training action of the first target user according to the training action collection, wherein the training action collection all meets the first expression requirement;

[0029] Specifically, balance ability, as an important physiological function of the human body, plays an irreplaceable role in maintaining the normal posture and activities of the human body. The evaluation of balance ability is an important basis for the diagnosis, treatment and rehabilitation training of patients with balance disorders. Therefore, in order to reasonably evaluate the patient's balance ability, this application proposes a human posture balance capture method based on a smart wearable device, that is, the data of the patient's rehabilitation training posture is collected through the smart wearable device, which provides a scientific data reference for the evaluation of the patient's balance ability.

[0030] Among them, the first target user is a user who needs to perform balance rehabilitation training, the first training posture is the rehabilitation training posture that the user needs to complete, including coordinated movements of the upper limbs, torso and lower limbs, so as to evaluate the user's rehabilitation status, and the training action collection includes various balance rehabilitation training action sets under scientific guidance, and each training action is connected, that is, the first training posture can be understood as a complete and coherent set of rehabilitation training actions.

[0031] Furthermore, the first initial training action is the starting action in the above-mentioned complete and coherent rehabilitation training action. For example, the first initial training action can be understood as: the user is in an upright standing posture, and is required to put his hands together, raise them upright above his head, and bend his left leg naturally at 90°, and maintain this posture for about one minute. In addition, when completing the collection of training actions, the target user needs to meet the first expression requirement, wherein the first expression requirement can be understood as the user's eyes are closed. Generally speaking, in order to make the user's balance rehabilitation training more stable, the user can be required to complete the training action with his eyes closed. Since visual input is the basis for the body's positioning in the environment, by blocking the user's visual input, the user's rehabilitation balance training can be more stable.

[0032] Step S300: Based on the smart wearable device, collecting skeletal posture data of the first initial training action of the first target user to generate a first skeletal posture collection of the first initial training action;

[0033] Furthermore, step S300 includes:

[0034] Step S310: According to the smart wearable device, force collection is performed on each skeletal force point of the first initial training action to generate a data set of each skeletal force point, wherein the data set of each skeletal force point includes a set of force magnitudes of each skeletal force and a set of force directions of each skeletal force;

[0035] Step S320: performing single calculation of each bone posture vector on the set of each bone force magnitude and the set of each bone force direction to generate a set of calculation results of each bone vector;

[0036] Step S330: performing data analysis on the calculation result sets of the bone vectors to generate the first bone posture collection.

[0037] Specifically, after obtaining the first initial training action of the first target user, the force collection of each skeletal force point of the first initial training action can be performed according to the smart wearable device. Specifically, when the user is performing the first initial training action, the bones of each key part of his limbs will be subjected to force to maintain the balance of the user's limbs. For example, when the user performs the above example action, there is a first force on the cervical spine bones, a second force on the left upper limb bones, a third force on the right upper limb bones, a fourth force on the trunk bones, a fifth force on the left lower limb bones, and a sixth force on the right lower limb bones, etc. Further, the first force, the second force, and up to the sixth force constitute the data set of each skeletal force point. It should be noted that the data set of each skeletal force point includes at least six forces, and the forces on the bones of other parts of the user's body are not described one by one.

[0038] Further, the data set of each bone force point includes a set of force magnitudes of each bone and a set of force directions of each bone. As the name implies, the set of force magnitudes of each bone is a set of force magnitudes, and data is collected through a wearable device. The set of force directions of each bone is a set of force directions, and data is also collected through a wearable device. When a user performs rehabilitation balance training, the strength and force direction of the rehabilitation training action need to meet scientific standards in order to better complete the rehabilitation training action. By calculating the force magnitudes of each bone and the force direction of each bone, a single vector calculation of each bone posture is performed. The fifth force of the left lower limb bone is used as an example for explanation. When the user's left leg is naturally bent at 90°, the bone at the knee needs to be forced in the left direction, rather than turning over and forcing in front of the user. Then, the vector calculation is performed on the left direction of the bone at the knee and the bone force magnitude in this direction. The force magnitude of the user's kneecap in the left direction can be calculated. By analogy, the force magnitude in the force direction of each bone is calculated respectively, thereby forming the set of calculation results of each bone vector.

[0039] Then, data analysis is performed on the calculation result set of each bone vector, that is, by performing data analysis on the force magnitude in the force direction of each bone, it can be judged whether the various bones of the body are scientifically stressed when the user completes the first initial training action, and then the user's human skeleton representation map is located by using the force information of each bone in the form of images or videos as input data. Furthermore, the first bone posture collection is generated based on the located human skeleton representation map, and the first bone posture collection truly reflects the posture information of each bone of the user when completing the first initial training action.

[0040] Step S400: constructing a human posture balance correction database, and the human posture balance correction database is included in the human posture balance capture system;

[0041] Further, such as Figure 3 As shown, step S400 includes:

[0042] Step S410: collecting historical training data of a second user set according to user data access rights to generate a historical training data set, wherein the second posture balance feature of the second user set is different from the first posture balance feature of the first target user;

[0043] Step S420: dividing the historical training data set by data weight values ​​to generate a historical training data set and a historical verification data set;

[0044] Step S430: performing data training on the historical training data set to construct a human posture balance correction-data training layer;

[0045] Step S440: Based on the historical verification data set, data verification is performed on the constructed human posture balance correction-data training layer to generate a first data verification result;

[0046] Step S450: According to the first data verification result, the human body posture balance correction-data training layer is hierarchically optimized.

[0047] Specifically, after obtaining the user's real skeletal posture information when completing the first initial training action, in order to conduct force evaluation on the real skeletal postures, a human body posture balance correction database can be constructed for evaluation and correction, wherein the human body posture balance correction database is included in the human body posture balance capture system.

[0048] Specifically, in order to construct the human posture balance correction database, the historical training data of the second user set can be collected according to the user data access rights, wherein the second user set can be characterized as a set of users who have completed balance rehabilitation training movements and whose training movements have met scientific standards. When collecting training data for the second user set, historical training data collection can be performed according to the user data access rights after the user's access rights are passed. The historical training data set is the collected data set, which includes the balance rehabilitation training movements of each user in the second user set, as well as the specific bone force data set of each movement, etc.

[0049] Furthermore, the historical training data set can be divided according to data weight values, that is, the historical training data set is divided into the historical training data set and the historical verification data set. The historical training data set is used to train and construct the human posture balance correction database, and the historical verification data set is used to verify the constructed human posture balance correction database, and then timely optimize the problems with defects, so that the constructed database can accurately balance the user's actual training movements. It should be noted that the historical training data set is the main part of the training, and should occupy at least half of the weight value in the historical training data set, and the rest is the historical verification data set.

[0050] By performing data training on the historical training data set, a human body posture balance correction-data training layer can be constructed, wherein the human body posture balance correction-data training layer is a preliminary hierarchical level of the human body posture balance correction database, which is used to train the historical training data set, and then the constructed human body posture balance correction-data training layer is data verified through the historical verification data set, that is, to verify whether the human body posture balance correction-data training layer can accurately evaluate the user's training movements. If it can be accurately evaluated, the user's actual training movements are evaluated and corrected according to the human body posture balance correction-data training layer. Otherwise, this level is optimized.

[0051] Step S500: inputting the first initial training action into the human posture balance correction database for training screening to obtain first preset skeletal posture information corresponding to the first initial training action;

[0052] Step S600: determining whether the first skeleton posture collection satisfies the first preset skeleton posture information;

[0053] Step S700: If the first skeleton posture collection does not satisfy the first preset skeleton posture information, the human body posture balance capture system generates a first compensation instruction to perform motion compensation on the first initial training action.

[0054] Specifically, after the human posture balance correction database is constructed, in order to evaluate the actual training movements of the target user, the first initial training movement can be input into the human posture balance correction database for training screening. Because the human posture balance correction database is built from historical training data of a set of users who have completed balance rehabilitation training movements and whose training movements have met scientific standards, the first initial training movement can be trained according to the human posture balance correction database to obtain the first preset skeletal posture information corresponding to the first initial training movement, wherein the first preset skeletal posture information can be trained based on the actual training movements of the user until standard skeletal posture information that meets the actual training movements is parsed, and the first preset skeletal posture information can be represented as standard skeletal posture information.

[0055] Furthermore, it is necessary to determine whether the first bone posture collection satisfies the first preset bone posture information, that is, whether the actual bone posture of the user reaches the above-mentioned standard bone posture information. If not, the first initial training action can be compensated according to the first compensation instruction. For example, if the force direction of the bones of the left leg of the user does not reach the predetermined left direction when completing the first initial training action, the user can be reminded according to the first compensation instruction, and the existing actual action can be compensated until the force direction of the bones of the left leg reaches the predetermined left direction, so that the single training action of the target user can reach a scientific and standard state, thereby improving the training efficiency of the target user.

[0056] Further, such as Figure 2 As shown, the embodiment of the present application also includes:

[0057] Step S810: obtaining a second training action in the training action collection;

[0058] Step S820: obtaining first transition action information according to the first initial training action and the second training action;

[0059] Step S830: performing transition action environment recognition and collection on the first transition action information according to the smart wearable device of the first target user, and generating transition action-environment collection data;

[0060] Step S840: obtaining a first limb transition center of gravity point of the transition action based on the transition action-environment collection data;

[0061] Step S850: determining whether the first ground distance information of the transition center of gravity of the first limb reaches the preset ground distance information;

[0062] Step S860: If the first ground distance information of the first limb transition center of gravity point reaches the preset ground distance information, a first recovery instruction is generated to perform action recovery on the first initial training action.

[0063] Specifically, since the individual training movements in the training movement collection are in a coherent state, in order to ensure the stability and coherence of the target user in completing this series of movements, the transition movements between the training movements may be evaluated.

[0064] Further, the second training action can be characterized as: on the basis of the target user standing upright, the target user puts his hands together, raises them upright over his head, and then bends his right leg naturally to 90°, and maintains this posture for about one minute. Wherein, when the first initial training action is converted to the second training action, data collection can be performed on its transition action. The first transition action information can be understood as the transition of the hands from the state of being together to the state of being naturally drooping, and the left leg is put down to present a standard standing posture. At the same time, on the basis of the standard standing posture, the right leg is slowly raised, so that the right leg is naturally bent to 90°. This transition action can be identified and collected according to the smart wearable device. Specifically, it can be understood that the surrounding environment is collected according to the smart glasses part in the smart wearable device, that is, it is determined whether the surrounding environment will affect the target user's action. For example, when the target user is in a rehabilitation training venue with a large number of people, the target user may be nervous, which affects the standard execution of his action. The transition action-environment collection data is the result of data collection of the surrounding environment.

[0065] Furthermore, based on the transition action-environment collection data, the first limb transition center of gravity point of the transition action can be obtained. As shown in the above example, when the target user fails to fully perform the first transition action due to nervousness, his limbs may tilt, causing the original body center of gravity to shift. The first limb transition center of gravity point is the body center of gravity point after the shift. Furthermore, if the body center of gravity shifts significantly, it may cause the target user to fall down or other unexpected situations. Therefore, it can be determined whether the first ground distance information of the first limb transition center of gravity point reaches the preset ground distance information, wherein the first ground distance information is the distance from the target user's shifted body center of gravity to the ground, and the preset ground distance information can be understood as a preset dangerous distance from the user's body center of gravity to the ground, and this dangerous distance may cause the user to fall down.

[0066] If the first ground distance information of the first limb transition center of gravity point reaches the preset ground distance information, that is, the target user may be at risk of falling due to the low center of gravity of the body, the target user can be reminded to return to the first initial training action according to the first recovery instruction, thereby achieving a stable transition of the target user after completing each consecutive action, ensuring the balanced capture of the target user's human posture.

[0067] Furthermore, the step S840 of obtaining the first limb transition center of gravity of the transition action includes:

[0068] Step S841: performing data analysis on the upper limb force information of the first target user according to the transition action-environment collection data to generate upper limb force data;

[0069] Step S842: performing data analysis on the trunk force information of the first target user according to the transition action-environment collection data to generate trunk force data;

[0070] Step S843: performing data analysis on the lower limb force information of the first target user according to the transition action-environment collection data to generate lower limb force data;

[0071] Step S844: Performing computer algorithm data processing on the upper limb force data, the trunk force data and the lower limb force data to generate the first limb transition center of gravity point.

[0072] Specifically, when obtaining the first limb transition center of gravity point of the transition action, specifically, data can be collected based on the transition action-environment data, and data analysis can be performed on the force information of various parts of the target user's body, and then data processing can be performed on the analysis results.

[0073] Furthermore, based on the transition action-environment collection data, data analysis can be performed on the upper limb force information of the first target user, that is, when the target user is at risk of falling, the upper limb force data collected by the wearable device can be used to perform data analysis on the upper limb force information, and the upper limb force data is the result of the analysis; similarly, based on the transition action-environment collection data, data analysis can be performed on the torso force information of the first target user, that is, the main body force data collected by the wearable device can be used to perform data analysis on the main body force information, and the torso force data can be used to perform the analysis; similarly, based on the transition action-environment collection data, data analysis can be performed on the lower limb force information of the first target user, that is, the lower limb force data collected by the wearable device can be used to perform data analysis on the lower limb force information, and the lower limb force data can be used to perform the analysis.

[0074] Then, the upper limb force data, the trunk force data and the lower limb force data are processed by computer algorithms. For example, when the target user's body tilts, the upper limb force data will drop appropriately in space, the trunk force data will tilt appropriately in space, and the lower limb force data will drop appropriately in space. Then, the three force data are processed by computer algorithms. Through a series of data analysis, data fusion, etc., the final transition center of gravity of the first limb is obtained. In theory, the transition center of gravity of the first limb is offset compared to the center point of the first limb when the target user is in the first initial training action.

[0075] Furthermore, the step S430 of constructing the human body posture balance correction-data training layer includes:

[0076] Step S431: obtaining a training posture data set of each user in the second user set according to the historical training data set;

[0077] Step S432: splitting the training posture data set to obtain each skeletal posture information set corresponding to each training posture;

[0078] Step S433: performing longitudinal matching mapping on each training posture and each skeleton posture information set to construct the human posture balance correction-data training layer.

[0079] Specifically, when constructing the human posture balance correction-data training layer, specifically, the training posture data set of each user in the second user set can be obtained according to the historical training data set, wherein the training posture data set includes each skeletal posture information set corresponding to each training posture, wherein each training posture is each sub-action when the target user performs balance rehabilitation training, and each skeletal posture information set can be understood as a standard skeletal posture information set for each sub-action.

[0080] Then, a longitudinal matching mapping is performed on the training postures and the skeletal posture information sets. That is, in terms of splitting, the first training posture in the training postures corresponds to the first training action, which corresponds to the skeletal posture information sets in the first training action, and so on, to complete the longitudinal matching mapping of all postures in the training postures, and then according to each distributed longitudinal matching mapping, the human body posture balance correction-data training layer is constructed, and the human body posture balance correction-data training layer reflects the mapping relationship between the user's training actions and their standard skeletal postures.

[0081] Furthermore, the step S450 of performing hierarchical optimization on the human body posture balance correction-data training layer includes:

[0082] Step S451: determining whether the first data verification result matches a preset data verification result;

[0083] Step S452: if the first data verification result does not match the preset data verification result, constructing a human posture balance correction-data optimization layer according to the historical verification data set;

[0084] Step S453: According to the human body posture balance correction-data optimization layer, the human body posture balance correction-data training layer is hierarchically optimized.

[0085] Specifically, when performing hierarchical optimization on the human body posture balance correction-data training layer, it can be determined first whether the first data verification result is consistent with the preset data verification result, wherein the first data verification result is a verification result for verifying whether the human body posture balance correction-data training layer can accurately evaluate the user's training movements, and the preset data verification result is a verification result under a preset ideal state that the human body posture balance correction-data training layer can accurately evaluate the user's training movements.

[0086] If the first data verification result does not match the preset data verification result, it means that the human body posture balance correction-data training layer cannot accurately evaluate the user's training movements, and then a human body posture balance correction-data optimization layer can be constructed based on the historical verification data set. The human body posture balance correction-data optimization layer is a deep optimization layer of the human body posture balance correction database, and the human body posture balance correction-data training layer can be hierarchically optimized according to the verification data until the training layer can accurately evaluate the target user's training movements.

[0087] Compared with the prior art, the present invention has the following beneficial effects:

[0088] 1. Obtain a first training posture of a first target user, wherein the first training posture includes a training action collection, and the individual training actions in the training action collection are in a coherent state; obtain a first initial training action of the first target user according to the training action collection, wherein the training action collection all meets the first posture requirement; based on a smart wearable device, collect skeletal posture data of the first initial training action of the first target user to generate a first skeletal posture collection of the first initial training action; construct a human posture balance correction database, and the human posture balance correction database is included in the human posture balance capture system; input the first initial training action into the human posture balance correction database for training screening to obtain the first preset skeletal posture information corresponding to the first initial training action; determine whether the first skeletal posture collection meets the first preset skeletal posture information; if the first skeletal posture collection does not meet the first preset skeletal posture information, the human posture balance capture system generates a first compensation instruction to perform action compensation on the first initial training action. Through the intelligent wearable device, the training posture data of the target user is traversed and collected. By building a human posture balance correction database, the skeletal posture collection corresponding to the actual training movements of the target user is compared with the expected standard skeletal posture collection, and then the substandard training movements are compensated, so that the individual training movements of the target user are all in a scientific and standard state, achieving the technical effect of all-round monitoring and capturing of the human posture balance of the target user, and providing real-time motion compensation reminders for substandard training movements, thereby improving the efficiency of the human posture balance training of the target user.

[0089] 2. By evaluating the transition movements between each training movement, the stability and continuity of the target user in completing this series of movements are ensured. When the center of gravity of the target user's body shifts, the target user is reminded to return to the first initial training movement, thereby achieving a stable transition of the target user in completing each coherent movement and ensuring the balanced capture of the target user's human posture.

[0090] 3. By weighting the historical training data set, a human posture balance correction-data training layer and a human posture balance correction-data optimization layer are constructed respectively, thereby achieving accurate training of user training movements. When the human posture balance correction-data training layer cannot accurately evaluate the user's training movements, a human posture balance correction-data optimization layer can be constructed, and the human posture balance correction-data training layer is hierarchically optimized according to the verification data until the training layer can accurately evaluate the target user's training movements.

[0091] Embodiment 2

[0092] Based on the same inventive concept as the method for capturing a human posture balance based on a smart wearable device in the aforementioned embodiment, the present invention also provides a human posture balance capturing system based on a smart wearable device, such as Figure 4 As shown, the system comprises:

[0093] First obtaining unit 11: The first obtaining unit 11 is used to obtain a first training posture of a first target user, wherein the first training posture includes a training action collection, and the individual training actions in the training action collection are in a coherent state;

[0094] The second obtaining unit 12 is used to obtain the first initial training action of the first target user according to the training action collection, wherein the training action collection all meets the first expression requirement;

[0095] The first collecting unit 13: The first collecting unit 13 is used to collect skeletal posture data of the first initial training action of the first target user based on the smart wearable device, and generate a first skeletal posture collection of the first initial training action;

[0096] The first construction unit 14: the first construction unit 14 is used to construct a human posture balance correction database, and the human posture balance correction database is included in the human posture balance capture system;

[0097] First input unit 15: The first input unit 15 is used to input the first initial training action into the human posture balance correction database for training screening, and obtain first preset bone posture information corresponding to the first initial training action;

[0098] The first judging unit 16: The first judging unit 16 is used to judge whether the first skeleton posture collection satisfies the first preset skeleton posture information;

[0099] The first generating unit 17: The first generating unit 17 is used for the human body posture balance capturing system to generate a first compensation instruction to perform motion compensation on the first initial training motion if the first skeleton posture collection does not satisfy the first preset skeleton posture information.

[0100] Furthermore, the system also includes:

[0101] A third obtaining unit: the third obtaining unit is used to obtain a second training action in the training action collection;

[0102] Fourth obtaining unit: the fourth obtaining unit is used to obtain first transition action information according to the first initial training action and the second training action;

[0103] Second collection unit: The second collection unit is used to perform environment recognition and collection of the first transition action information according to the smart wearable device of the first target user, and generate transition action-environment collection data;

[0104] Fifth obtaining unit: the fifth obtaining unit is used to obtain the first limb transition center of gravity point of the transition action based on the transition action-environment collection data;

[0105] The second judging unit is used to judge whether the first ground distance information of the transition center of gravity of the first limb reaches the preset ground distance information;

[0106] Second generating unit: The second generating unit is used to generate a first recovery instruction to perform action recovery on the first initial training action if the first ground distance information of the transition center of gravity point of the first limb reaches the preset ground distance information.

[0107] Furthermore, the system also includes:

[0108] First analysis unit: The first analysis unit is used to collect data based on the transition action-environment, perform data analysis on the upper limb force information of the first target user, and generate upper limb force data;

[0109] Second analysis unit: The second analysis unit is used to collect data based on the transition action-environment, perform data analysis on the trunk force information of the first target user, and generate trunk force data;

[0110] The third analysis unit: the third analysis unit is used to collect data based on the transition action-environment, perform data analysis on the force information of the lower limbs of the first target user, and generate lower limb force data;

[0111] The first processing unit: The first processing unit is used to perform data processing on the upper limb force data, the trunk force data and the lower limb force data using a computer algorithm to generate the first limb transition center of gravity point.

[0112] Furthermore, the system also includes:

[0113] The third collection unit is used to collect the force of each bone force point of the first initial training action according to the smart wearable device, and generate a data set of each bone force point, wherein the data set of each bone force point includes a set of force magnitudes of each bone and a set of force directions of each bone;

[0114] The first calculation unit: the first calculation unit is used to perform a single calculation of each bone posture vector on the set of each bone force magnitude and the set of each bone force direction, and generate a set of calculation results of each bone vector;

[0115] The first parsing unit: The first parsing unit is used to perform data parsing on the calculation result sets of each bone vector to generate the first bone posture collection.

[0116] Furthermore, the system also includes:

[0117] Fourth collection unit: the fourth collection unit is used to collect data on the historical training data of the second user set according to the user data access permission to generate a historical training data set, wherein the second posture balance feature of the second user set is different from the first posture balance feature of the first target user;

[0118] First partitioning unit: The first partitioning unit is used to perform data weight value partitioning on the historical training data set to generate a historical training data set and a historical verification data set;

[0119] The first training unit: The first training unit is used to perform data training on the historical training data set to construct a human posture balance correction-data training layer;

[0120] First verification unit: The first verification unit is used to perform data verification on the constructed human posture balance correction-data training layer based on the historical verification data set to generate a first data verification result;

[0121] First optimization unit: The first optimization unit is used to perform hierarchical optimization on the human body posture balance correction-data training layer according to the first data verification result.

[0122] Furthermore, the system also includes:

[0123] Sixth obtaining unit: the sixth obtaining unit is used to obtain a training posture data set of each user in the second user set according to the historical training data set;

[0124] First splitting unit: The first splitting unit is used to split the training posture data set to obtain each skeletal posture information set corresponding to each training posture;

[0125] The first mapping unit: The first mapping unit is used to perform longitudinal matching mapping on each training posture and each skeleton posture information set to construct the human posture balance correction-data training layer.

[0126] Furthermore, the system also includes:

[0127] The third judgment unit: the third judgment unit is used to judge whether the first data verification result is consistent with the preset data verification result;

[0128] Second construction unit: The second construction unit is used to construct a human posture balance correction-data optimization layer according to the historical verification data set if the first data verification result does not match the preset data verification result;

[0129] Second optimization unit: The second optimization unit is used to perform hierarchical optimization on the human body posture balance correction-data training layer according to the human body posture balance correction-data optimization layer.

[0130] The foregoing Figure 1 The various variations and specific examples of the human body posture balance capture method based on a smart wearable device in Example 1 are also applicable to a human body posture balance capture system based on a smart wearable device in this embodiment. Through the above detailed description of a human body posture balance capture method based on a smart wearable device, those skilled in the art can clearly know the implementation method of a human body posture balance capture system based on a smart wearable device in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0131] Embodiment 3

[0132] Reference below Figure 5 The computer device of the embodiment of the present application is described as follows. The computer device may be an application version management server or a terminal, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for processing an application package is implemented.

[0133] When the computer device is a terminal, the computer device may further include a display screen and an input device. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touch pad provided on the computer device housing, or an external keyboard, touch pad or mouse, etc.

[0134] Those skilled in the art will understand that Figure 5The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0135] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0136] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0137] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in the above-mentioned method embodiments.

[0138] The embodiment of the present application provides a human posture balance capture method based on a smart wearable device, characterized in that the method is applied to a human posture balance capture system, and the method includes: obtaining a first training posture of a first target user, wherein the first training posture includes a training action collection, and the single training actions in the training action collection are in a coherent state; obtaining a first initial training action of the first target user according to the training action collection, wherein the training action collection all meets the first demeanor requirement; based on the smart wearable device, performing skeletal posture data collection on the first initial training action of the first target user to generate a first skeletal posture collection of the first initial training action; constructing a human posture balance correction database, and the human posture balance correction database is included in the human posture balance capture system; inputting the first initial training action into the human posture balance correction database for training screening to obtain first preset skeletal posture information corresponding to the first initial training action; judging whether the first skeletal posture collection meets the first preset skeletal posture information; if the first skeletal posture collection does not meet the first preset skeletal posture information, the human posture balance capture system generates a first compensation instruction to perform action compensation on the first initial training action.

[0139] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0140] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0141] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for capturing human posture balance based on a smart wearable device, characterized in that: The method is applied to a human posture balance capture system, and the method comprises: Obtaining a first training posture of a first target user, wherein the first training posture includes a training action collection, and individual training actions in the training action collection are in a coherent state; Obtaining a first initial training action of the first target user according to the training action collection, wherein the training action collection all meets the first posture requirement; Based on the smart wearable device, collecting skeletal posture data of the first initial training action of the first target user to generate a first skeletal posture collection of the first initial training action; Constructing a human posture balance correction database, wherein the human posture balance correction database is included in a human posture balance capture system; Inputting the first initial training action into the human body posture balance correction database for training screening to obtain first preset skeletal posture information corresponding to the first initial training action; Determine whether the first skeleton posture collection satisfies the first preset skeleton posture information; If the first skeleton posture collection does not satisfy the first preset skeleton posture information, the human body posture balance capture system generates a first compensation instruction to perform motion compensation on the first initial training action; Obtaining a second training action in the training action collection; Obtaining first transition action information according to the first initial training action and the second training action; According to the smart wearable device of the first target user, performing environment recognition and collection of the transition action on the first transition action information to generate transition action-environment collection data; Based on the transition action-environment collection data, obtaining a first limb transition center of gravity point of the transition action; Determine whether the first ground distance information of the transition center of gravity of the first limb reaches the preset ground distance information; If the first ground distance information of the first limb transition center of gravity point reaches the preset ground distance information, a first recovery instruction is generated to perform action recovery on the first initial training action.

2. The method according to claim 1, characterized in that The step of obtaining the first limb transition center of gravity point of the transition action comprises: Performing data analysis on upper limb force information of the first target user according to the transition action-environment collection data to generate upper limb force data; According to the transition action-environment collection data, performing data analysis on the torso force information of the first target user to generate torso force data; According to the transition action-environment collection data, data analysis is performed on the lower limb force information of the first target user to generate lower limb force data; The upper limb force data, the trunk force data and the lower limb force data are processed by computer algorithm to generate the first limb transition center of gravity point.

3. The method according to claim 1, characterized in that The step of generating a first skeleton posture collection of the first initial training action includes: According to the smart wearable device, force is collected on each skeletal force point of the first initial training action to generate a data set of each skeletal force point, wherein the data set of each skeletal force point includes a set of force magnitudes of each skeletal force and a set of force directions of each skeletal force; Performing a single bone posture vector calculation on the set of force magnitudes of each bone and the set of force directions of each bone to generate a set of calculation results of each bone vector; Data analysis is performed on the calculation result sets of each bone vector to generate the first bone posture collection.

4. The method according to claim 1, characterized in that The method of constructing a human posture balance correction database comprises: According to the user data access permission, data collection is performed on the historical training data of the second user set to generate a historical training data set, wherein the second posture balance feature of the second user set is different from the first posture balance feature of the first target user; Performing data weight value division on the historical training data set to generate a historical training data set and a historical verification data set; Performing data training on the historical training data set to construct a human posture balance correction-data training layer; Based on the historical verification data set, data verification is performed on the constructed human posture balance correction-data training layer to generate a first data verification result; According to the first data verification result, the human body posture balance correction-data training layer is hierarchically optimized.

5. The method according to claim 4, characterized in that The constructing of the human posture balance correction-data training layer includes: Obtaining a training posture data set for each user in the second user set according to the historical training data set; Splitting the training posture data set to obtain each skeletal posture information set corresponding to each training posture; The training postures and the skeleton posture information sets are longitudinally matched and mapped to construct the human posture balance correction-data training layer.

6. The method according to claim 4, characterized in that The step of performing hierarchical optimization on the human posture balance correction-data training layer includes: Determining whether the first data verification result is consistent with a preset data verification result; If the first data verification result does not match the preset data verification result, constructing a human posture balance correction-data optimization layer according to the historical verification data set; According to the human body posture balance correction-data optimization layer, the human body posture balance correction-data training layer is hierarchically optimized.

7. A human posture balance capture system based on a smart wearable device, characterized in that: The system comprises: First obtaining unit: The first obtaining unit is used to obtain a first training posture of a first target user, wherein the first training posture includes a training action collection, and the individual training actions in the training action collection are in a coherent state; Second obtaining unit: The second obtaining unit is used to obtain the first initial training action of the first target user according to the training action collection, wherein the training action collection all meets the first expression requirement; First collection unit: The first collection unit is used to collect skeletal posture data of the first initial training action of the first target user based on the smart wearable device, and generate a first skeletal posture collection of the first initial training action; First construction unit: the first construction unit is used to construct a human posture balance correction database, and the human posture balance correction database is included in the human posture balance capture system; First input unit: The first input unit is used to input the first initial training action into the human posture balance correction database for training screening, and obtain first preset bone posture information corresponding to the first initial training action; First judging unit: The first judging unit is used to judge whether the first skeleton posture collection satisfies the first preset skeleton posture information; The first generating unit is used for generating a first compensation instruction by the human body posture balance capturing system to perform action compensation on the first initial training action if the first bone posture collection does not satisfy the first preset bone posture information; The system further comprises: A third obtaining unit: the third obtaining unit is used to obtain a second training action in the training action collection; Fourth obtaining unit: the fourth obtaining unit is used to obtain first transition action information according to the first initial training action and the second training action; Second collection unit: The second collection unit is used to perform environment recognition and collection of the first transition action information according to the smart wearable device of the first target user, and generate transition action-environment collection data; Fifth obtaining unit: the fifth obtaining unit is used to obtain the first limb transition center of gravity point of the transition action based on the transition action-environment collection data; The second judging unit is used to judge whether the first ground distance information of the transition center of gravity of the first limb reaches the preset ground distance information; Second generating unit: The second generating unit is used to generate a first recovery instruction to perform action recovery on the first initial training action if the first ground distance information of the transition center of gravity point of the first limb reaches the preset ground distance information.

8. A human posture balance capture device based on a smart wearable device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method according to any one of claims 1 to 6 are implemented.

Images