Training action recognition method, device, equipment and medium

Abstract

The embodiment of the present invention relates to the technical field of limb joint rehabilitation therapy, and discloses a training action recognition method, device, equipment and medium. The method includes: controlling the muscles of the target training part to be trained according to the guided action, and obtaining the training action recognition data in real time; the training action recognition data includes multiple training action recognition signal quantities; obtaining the change rate of the training action recognition signal quantity according to the multiple training action recognition signal quantities; and recognizing the training action according to the change rate of the training action recognition signal quantity. The embodiment of the present invention recognizes the training action by the change rate of the action recognition signal quantity, abandons the cumbersome operation of calibrating the voltage value of the guided action, and can greatly reduce the operating burden of patients and medical staff.

Description

Technical Field

[0001] The present invention relates to the technical field of limb joint rehabilitation therapy, and in particular to a training action recognition method, device, equipment and medium. Background Art

[0002] According to relevant literature research, there are currently 70 million stroke patients in my country, with more than 2 million new cases each year. More than 50% of stroke patients have varying degrees of limb dysfunction, such as hand function disorder.

[0003] Clinical rehabilitation treatment of hand function movement disorders can be achieved through a hand function rehabilitation trainer. The hand function rehabilitation trainer includes a rehabilitation training host, training gloves and sensor gloves. Mirror training is a rehabilitation training method used by the hand function rehabilitation trainer, which identifies the flexion and extension movements of the healthy hand (as the guide hand) wearing the sensor glove, and then the trainer controls the affected hand wearing the training glove to perform active and passive training according to the guiding movements of the guide hand. In the mirror training method, the recognition of the guiding hand's movements is the key to the success of this method.

[0004] The training devices currently on the market recognize the movements of the guiding hand by obtaining the movement data of the guiding hand through the flexible resistance sensor set in the guiding glove. When the guiding hand flexes and extends, the guiding glove flexes and extends along with the guiding hand. The absolute value of the voltage output by the flexible resistance sensor can represent the degree of flexion and extension, so that the voltage values ​​of the hand when it is fully open, fully closed, and in the intermediate process from opening to closing can be collected, and then the hand movement of the guiding hand can be recognized based on the collected voltage values. At present, when recognizing the movements of the guiding hand, it is necessary to calibrate the voltage values ​​for the different degrees of flexion and extension during the hand movement process one by one, and then recognize the training movements of the training hand based on the calibrated voltage values. The operation is cumbersome and brings great inconvenience to doctors and patients. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide a training action recognition method, device, equipment and medium, which recognizes the training action through the change rate of the action recognition signal, eliminates the tedious operation of voltage value calibration for the guiding action, and can greatly reduce the operating burden of patients and medical staff.

[0006] In order to solve the above technical problems, in a first aspect, an embodiment of the present invention provides a training action recognition method, which is applied to limb joint muscle rehabilitation training, comprising:

[0007] Controlling the muscles of the target training part to be trained according to the guiding action, and acquiring training action recognition data in real time; the training action recognition data includes a plurality of training action recognition signal quantities;

[0008] Obtaining a change rate of a training action recognition signal amount according to the plurality of training action recognition signal amounts;

[0009] The training action is identified according to the change rate of the training action recognition signal amount and the relationship between the change rate of the action recognition signal amount and the action.

[0010] In addition, the identifying the training action according to the change rate of the training action recognition signal and the relationship between the change rate of the action recognition signal and the action includes:

[0011] If the change rate of the training movement recognition signal quantity within the first preset time length is greater than 0, and the values ​​of the change rate of the training movement recognition signal quantity are all within the first reference value interval, and the number of change rates of the training movement recognition signal quantity is greater than or equal to the first threshold, then the training movement is determined to be a flexion movement.

[0012] In addition, the identifying the training action according to the change rate of the training action recognition signal amount and the relationship between the change rate of the action recognition signal amount and the action further includes:

[0013] If the change rate of the training action recognition signal quantity within the second preset time length is less than 0, and the absolute value of the change rate of the training action recognition signal quantity is within the second reference value interval, and the number of change rates of the training action recognition signal quantity is greater than the second threshold, then the training action is determined to be a stretching action.

[0014] In addition, the identifying the training action according to the change rate of the training action recognition signal amount and the relationship between the change rate of the action recognition signal amount and the action further includes:

[0015] The degree of completion of the training action is determined according to the rate of change of the training action recognition signal and the calibrated rate of change determined based on the guiding action signal.

[0016] In addition, the determining the degree of completion of the training action according to the change rate of the training action recognition signal and the calibrated change rate determined based on the guiding action signal comprises:

[0017] Calculating the ratio of the change rate of each training action recognition signal to the calibrated change rate;

[0018] The degree of completion of the training action is determined according to the ratio.

[0019] In addition, the method further comprises:

[0020] The calibration change rate is obtained in the following manner:

[0021] Collecting guidance action identification data; the guidance action identification data includes identification data of multiple guidance actions; the identification data of each guidance action includes: at least one signal quantity of the guidance action;

[0022] Obtaining a rate of change of the signal quantity of each guiding action according to at least one signal quantity of each guiding action;

[0023] Obtaining the calibration change rate according to the change rate of the signal quantity of the multiple guiding actions;

[0024] Optionally, the calibrated change rate is an average value of the change rates of the signal quantities of the multiple guiding actions.

[0025] In addition, the first reference value interval and the second reference value interval are updated using a preset strategy;

[0026] The method further includes updating the change rate boundary values ​​of the first reference value interval and the second reference value interval in the following manner:

[0027] Recording the maximum and minimum values ​​of the rate of change of the signal amount of each flexion action in the flexion action and the maximum and minimum values ​​of the rate of change of the signal amount of each extension action in the extension action;

[0028] If the maximum value of the rate of change of the signal amount of each flexing action in the flexing actions counted for K consecutive times is less than the maximum value of the rate of change of the previous first reference value interval, or the minimum value of the rate of change of the signal amount of each flexing action in the flexing actions counted for K consecutive times is greater than the minimum value of the rate of change of the previous first reference value interval, then the average value of the maximum values ​​of the rate of change of the signal amount of the flexing action counted for the K consecutive times is used as the maximum value of the rate of change of the first reference value interval after update, and the average value of the minimum values ​​of the rate of change of the signal amount of the flexing action counted for the K consecutive times is used as the minimum value of the rate of change of the first reference value interval after update;

[0029] If the maximum value of the rate of change of the signal amount of each stretching action in the stretching action counted for K consecutive times is less than the maximum value of the rate of change of the second reference value interval in the previous time, or the minimum value of the rate of change of the signal amount of each stretching action in the stretching action counted for K consecutive times is greater than the minimum value of the rate of change of the first reference value interval in the previous time, then the average value of the maximum value of the rate of change of the signal amount of the stretching action counted for the K times is used as the maximum value of the rate of change of the second reference value interval after update, and the average value of the minimum value of the rate of change of the signal amount of the stretching action counted for the K times is used as the minimum value of the rate of change of the second reference value interval after update; K is a natural number greater than 1;

[0030] Optionally, the method further comprises:

[0031] Collecting guidance action recognition data; the guidance action recognition data includes recognition data of multiple guidance actions; the recognition data of each guidance action includes: multiple signal quantities of flexion action and multiple signal quantities of extension action;

[0032] Obtaining a maximum value and a minimum value of a rate of change of the signal quantity of each flexion action according to multiple signal quantities of the flexion action, and obtaining a maximum value and a minimum value of a rate of change of a first initial reference value interval of the first reference value interval according to the maximum value and the minimum value of the rate of change of the signal quantity of multiple flexion actions;

[0033] Obtaining a maximum value and a minimum value of a rate of change of the signal quantity of each stretching action according to the multiple signal quantities of the stretching action, and obtaining a maximum value and a minimum value of a rate of change of a second initial reference value interval of the second reference value interval according to the maximum value and the minimum value of the rate of change of the signal quantity of the multiple stretching actions;

[0034] Optionally, the target training part includes a hand, a wrist joint, an elbow joint or a knee joint;

[0035] Optionally, the action recognition signal is a voltage signal that can change with the flexion and extension of the target training part.

[0036] In a second aspect, an embodiment of the present invention provides a training action recognition device, which is applied to limb joint muscle rehabilitation training, comprising:

[0037] A signal acquisition module is used to control the muscles of the target training part to be trained according to the guided action, and to obtain training action recognition data in real time; the training action recognition data includes a plurality of training action recognition signal quantities;

[0038] A change rate calculation module, used for obtaining a change rate of the training action recognition signal amount according to the plurality of training action recognition signal amounts;

[0039] The identification module is used to identify the training action according to the change rate of the training action identification signal and a preset judgment strategy.

[0040] In a third aspect, an embodiment of the present invention further provides a training action recognition device, comprising: a memory and a processor, the memory storing a computer program, and the processor running the computer program to implement the training action recognition method as described above.

[0041] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the training action recognition method as described in any embodiment of the present invention.

[0042] Compared with the prior art, the technical solution provided by the embodiment of the present invention has at least the following positive effects:

[0043] The embodiment of the present invention acquires training motion recognition data in real time, the training motion recognition data includes multiple training motion recognition signal quantities, obtains the change rate of the training motion recognition signal quantity based on the multiple training motion recognition signal quantities, and then recognizes the training motion based on the change rate of the training motion recognition signal quantity, thereby eliminating the cumbersome signal quantity calibration problem that exists when motion recognition is directly performed based on the training motion recognition signal quantity, and reduces the operational burden on patients and medical staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. It can be understood that the drawings in the following description are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0045] Figure 1 is a flow chart of a training action recognition method provided by an embodiment of the present invention;

[0046] Figure 2 is a schematic diagram of the structure of a training action recognition device provided by an embodiment of the present invention;

[0047] Figure 3 It is a structural diagram of a training action recognition device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described through implementation methods with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] The inventor found that the current hand function rehabilitation training device needs to calibrate the voltage value for the different flexion and extension degrees of the hand movement for the calibration of the guided action during the mirror rehabilitation training, and then identify the training action of the training hand according to the calibrated voltage value. The operation of calibrating the voltage value one by one is cumbersome, which brings great inconvenience to doctors and patients. To this end, the inventor proposed to calculate the action recognition data, such as the change rate of multiple voltage values, so as to perform action recognition according to the relationship between the change rate and the action, abandoning the method of directly performing action recognition by calibrating the voltage value, which can greatly reduce the burden on patients and medical staff.

[0050] The training action recognition method provided in the embodiment of the present invention is applicable to limb joint muscle rehabilitation training, and is particularly applicable to the mirror training method. Limb joints include but are not limited to hand, wrist, elbow or knee joints and other limb joints that have movement disorders due to muscle damage. The method is applicable to related training equipment, including but not limited to hand function electromyographic stimulation rehabilitation training equipment. Figure 1 As shown, the training action recognition method of this embodiment includes the following steps:

[0051] Step 101: Control the muscles of the target training part to perform training according to the guided action, and obtain the training action recognition data in real time.

[0052] In the mirror training method, the guiding action can be a training guiding action performed by the patient's healthy hand, but it is not limited thereto, and the guiding action can also be obtained by other means. The method of controlling the target training part to perform mirror training according to the guiding action is well known to those skilled in the art and will not be described in detail here.

[0053] The target training part may be a limb joint that needs to be trained, such as a hand, elbow joint, wrist joint or knee joint. This embodiment takes the training of the hand as an example.

[0054] In order to monitor the training movements of the hand, it is necessary to collect the training movement recognition data when the hand moves. The training movement recognition data includes a plurality of training movement recognition signal quantities. The training movement recognition signal quantity can be a voltage signal that can change with the flexion and extension of the target training part.

[0055] Specifically, sensor gloves can be used to collect training motion recognition data of the hand. For example, the voltage value that changes with the hand training motion can be obtained through the flexible resistance sensor in the sensor gloves. Preferably, the inductive sensor gloves provided by the applicant in the patent with application number 2021114250623 and the name of gloves for hand rehabilitation training and coil inductance for hand rehabilitation training can also be used to collect training motion recognition data, and different voltage values ​​generated by the training motion can also be collected. And the inductive sensor gloves also have the advantages of fast output signal response speed, low cost, and easy installation.

[0056] Step 102: Obtain the change rate of the training action recognition signal quantity according to the plurality of training action recognition signal quantities.

[0057] Taking the voltage value collected by the gloves as an example of the training motion recognition signal, the voltage increases continuously when the hand is fully opened and fully clenched; conversely, the voltage gradually decreases when the hand is fully clenched and fully opened. The curve of the voltage value of a complete fist clenching and opening action over time is similar to a smooth sawtooth wave. The shape of the waveform changes with the speed of the action. The faster the action speed, the steeper the waveform, and the slower the action speed, the smoother the waveform. If the action is paused during flexion and extension, the waveform remains fixed.

[0058] The calculation method of the change rate of the training action recognition signal is to calculate the ratio of the difference between the signal collected at the latter moment and the signal collected at the previous moment to the time length between the previous and next moments.

[0059] Specifically, the sensing signal voltage output by the glove sensor is recorded as u, and the voltage value collected at any time i-1, i, i+1 is recorded as u i-1 、u i 、u i+1 , where i ranges from 1 to N; the rate of change (i.e. slope) at time i is denoted by k i , in is the sampling time interval.

[0060] In step 102, the change rate of the training action recognition signal quantity can be calculated in real time based on the training action recognition signal quantity obtained in real time.

[0061] Step 103: Identify the training action according to the change rate of the training action recognition signal and the relationship between the change rate of the action recognition signal and the action.

[0062] The relationship between the change rate of the motion recognition signal and the motion is explained by taking the voltage value collected by the gloves as an example. In the process of the hand from fully opening to fully clenching, the voltage increases continuously. During this process, the change rate of the motion recognition signal of the hand motion is always greater than 0; conversely, the voltage gradually decreases from fully clenching to fully opening. During this process, the change rate of the motion recognition signal of the hand motion is always less than 0. When the change rate changes from greater than 0 to less than 0, the fist-clenching action ends and switches to the opening action. Conversely, when the change rate changes from less than 0 to greater than 0, the opening action switches to the fist-clenching action. During the fist-clenching and opening actions, the change rate can fluctuate within a certain range. When the sign of the change rate changes, the number of change rates within a certain time period TS can be counted, so that the number of change rates greater than 0 during the fist-clenching action process and the number of change rates less than 0 during the opening process can be counted. Therefore, by determining whether the rate of change is greater than or less than 0, the number of rates of change, and whether the size of the rate of change is within a certain range of values, it is possible to identify whether the hand movement is a flexion or extension movement, or an opening movement. The degree of completion of the hand movement can also be further identified, such as the degree of completion of a fist or the degree of opening.

[0063] Identifying training movements based on the rate of change of the training movement recognition signal and the relationship between the rate of change of the movement recognition signal and the movement may specifically include: if the rate of change of the training movement recognition signal within a first preset time period is greater than 0, and the values ​​of the rate of change of the training movement recognition signal are within a first reference value interval, and the number of the rates of change of the training movement recognition signal is greater than or equal to a first threshold, then determining that the training movement is a flexion movement.

[0064] Among them, the first preset duration can be the duration of the guided action to complete the flexion action. The first reference value interval can be determined according to the rate of change of the action recognition signal when the guided action completes the flexion action. The numerical range of the first reference value interval can be specifically determined according to the numerical range of the rate of change of the action recognition signal when the guided action completes the flexion action. Similarly, the first threshold can also be determined according to the number of change rates of the action recognition signal when the guided action completes the flexion action, that is, when the sampling frequencies of the action recognition signals of the training action and the guided action are consistent, the action recognition is performed in accordance with the principle that the number of change rates of the two is also consistent.

[0065] Identifying the training action based on the rate of change of the training action recognition signal and the relationship between the rate of change of the action recognition signal and the action may also specifically include: if the rate of change of the training action recognition signal is less than 0 within the second preset time period, and the absolute value of the rate of change of the training action recognition signal is within the second reference value interval, and the number of the rate of change of the training action recognition signal is greater than the second threshold, then the training action is determined to be a stretching action. The second preset time period, the second reference value interval, and the second threshold are determined in a similar manner to the first preset time period, the first reference value interval, and the second threshold, and can be determined accordingly according to the opening action of the guiding hand, which will not be repeated here.

[0066] Specifically, when it is detected that the sign of the change rate of the training action recognition signal quantity changes, for example, the change rate changes from less than 0 to greater than 0, the number of change rates of the training action recognition signal quantity within a certain time period TS (first preset time length) is calculated, and the size of the change rate value is determined at the same time. If the number of change rate values ​​is greater than or less than the first threshold value N_KT, and the change rate values ​​are all within the first reference value interval, it is considered that the hand performs a fisting action. When performing the hand opening action, the judgment method is similar and will not be repeated here.

[0067] The training action is identified according to the change rate of the training action recognition signal and the relationship between the change rate of the action recognition signal and the action, and may also include: determining the completion of the training action according to the change rate of the training action recognition signal and the calibrated change rate determined based on the guiding action signal.

[0068] Specifically, determining the completion of the training action based on the change rate of the training action recognition signal and the calibrated change rate determined based on the guiding action signal may include: calculating the ratio of the change rate of each training action recognition signal to the calibrated change rate, and determining the completion of the training action based on the ratio.

[0069] The judgment of the completion of the training action is to judge whether the amplitude and speed of the training hand's fist-clenching or opening action are consistent with the amplitude and speed of the guiding hand. Of course, the action completion can also be the consistency of the amplitude of the training hand's action and the amplitude of the guiding hand's action.

[0070] The method further comprises obtaining the calibration change rate in the following manner: collecting guidance action recognition data, the guidance action recognition data comprising recognition data of multiple guidance actions; the recognition data of each guidance action comprising: at least one signal quantity of the guidance action; obtaining the change rate of the signal quantity of each guidance action according to the at least one signal quantity of each guidance action; obtaining the calibration change rate according to the change rates of the signal quantities of multiple guidance actions. Optionally, the calibration change rate is the average of the change rates of the signal quantities of multiple guidance actions.

[0071] The following is an example of how to obtain the calibration change rate:

[0072] a. Perform the calibration action. The patient's healthy hand performs the action from fully open to fully clenched fist, and the rate of change of the signal during the action is recorded, which is recorded as the maximum value k max_l , and its calculation formula is Where l ranges from 0 to M-1; k max_l That is the rate of change of the signal quantity of a complete fist-clenching action.

[0073] b. Repeat step a. M times, and the rate of change each time is recorded as k max_0 , k max_1 , …, k max_(M-1) ;

[0074] c. The calibration change rate is equal to the change rate k max_l The average value of k max_mean =(k max_0 +k max_1 +…+k max_(M-1) ) / M, where M is any positive integer.

[0075] After obtaining the calibrated change rate, the action completion degree p is obtained in the following way during training:

[0076] p=|100×k i / k max_mean |, where p is the action completion degree parameter.

[0077] Furthermore, the first reference value interval and the second reference value interval can also be updated using a preset strategy. As the hand fatigue increases during training, the flexion and extension angle range will also become smaller accordingly. Therefore, the boundary values ​​of the reference value interval, that is, the maximum and minimum values ​​of the value interval, can be dynamically adjusted during training.

[0078] The method may further include updating the change rate boundary value of the first reference value interval and the second reference value interval in the following manner:

[0079] The maximum and minimum values ​​of the rate of change of the signal amount in each flexion movement and the maximum and minimum values ​​of the rate of change of the signal amount in each extension movement are recorded.

[0080] If the maximum value of the rate of change of the signal amount of each flexion action in K consecutive statistical flexion actions is less than the maximum value of the rate of change of the previous first reference value interval, or the minimum value of the rate of change of the signal amount of each flexion action in K consecutive statistical flexion actions is greater than the minimum value of the rate of change of the previous first reference value interval, then the average value of the maximum values ​​of the rate of change of the signal amount of the K statistical flexion actions is used as the maximum value of the rate of change of the updated first reference value interval, and the average value of the minimum values ​​of the rate of change of the signal amount of the K statistical flexion actions is used as the minimum value of the rate of change of the updated first reference value interval.

[0081] If the maximum value of the rate of change of the signal amount of each stretching action in the stretching actions counted for K consecutive times is less than the maximum value of the rate of change of the previous second reference value interval, or the minimum value of the rate of change of the signal amount of each stretching action in the stretching actions counted for K consecutive times is greater than the minimum value of the rate of change of the previous first reference value interval, then the average value of the maximum value of the rate of change of the signal amount of the stretching actions counted for K times is used as the maximum value of the rate of change of the updated second reference value interval, and the average value of the minimum value of the rate of change of the signal amount of the stretching actions counted for K times is used as the minimum value of the rate of change of the updated second reference value interval. K is a natural number greater than 1.

[0082] Optionally, the method may also include: collecting guiding action identification data; the guiding action identification data includes identification data of multiple guiding actions; the identification data of each guiding action includes: multiple signal quantities of flexion action and multiple signal quantities of extension action; obtaining the maximum and minimum values ​​of the rate of change of the signal quantity of each flexion action based on the multiple signal quantities of the flexion action, and obtaining the maximum and minimum values ​​of the rate of change of the first initial reference value interval of the first reference value interval based on the maximum and minimum values ​​of the rate of change of the signal quantity of multiple flexion actions; obtaining the maximum and minimum values ​​of the rate of change of the signal quantity of each extension action based on the multiple signal quantities of the extension action, and obtaining the maximum and minimum values ​​of the rate of change of the second initial reference value interval of the second reference value interval based on the maximum and minimum values ​​of the rate of change of the signal quantity of multiple extension actions.

[0083] During the training process, just repeat the recognition of the movements of clenching and opening the fist.

[0084] Compared with the prior art, the embodiment of the present invention identifies the training action through the change rate of the action recognition signal, abandons the cumbersome operation of calibrating the voltage value of the guiding action, and can greatly reduce the operating burden of patients and medical staff.

[0085] The embodiment of the present invention provides a training action recognition device, which is applied to limb joint muscle rehabilitation training and can be configured in a training action recognition device. Figure 2As shown, the device includes: a signal acquisition module 301, a change rate calculation module 302 and an identification module 303.

[0086] The signal acquisition module 301 is used to control the muscles of the target training part to be trained according to the guided action, and to obtain the training action recognition data in real time; the training action recognition data includes a plurality of training action recognition signal quantities.

[0087] The change rate calculation module 302 is used to obtain the change rate of the training action recognition signal amount according to the multiple training action recognition signal amounts;

[0088] The identification module 303 is used to identify the training action according to the change rate of the training action identification signal amount and a preset judgment strategy.

[0089] Optionally, the identification module 303 is specifically used to determine that the training movement is a flexion movement if the change rate of the training movement identification signal quantity within a first preset time period is greater than 0, and the values ​​of the change rate of the training movement identification signal quantity are all within a first reference value interval, and the number of change rates of the training movement identification signal quantity is greater than or equal to a first threshold.

[0090] The identification module 303 is also used to determine that the training action is a stretching action if the change rate of the training action recognition signal quantity within a second preset time period is less than 0, and the absolute value of the change rate of the training action recognition signal quantity is within a second reference value interval, and the number of change rates of the training action recognition signal quantity is greater than a second threshold.

[0091] The recognition module 303 is further configured to determine the degree of completion of the training action according to the change rate of the training action recognition signal and the calibrated change rate determined based on the guiding action signal.

[0092] The recognition module 303 is specifically used to calculate the ratio of the change rate of each training action recognition signal quantity to the calibrated change rate; and determine the completion degree of the training action according to the ratio.

[0093] Optionally, the device may also include a change rate calibration module, which is used to obtain the calibrated change rate in the following manner: collecting guidance action identification data; the guidance action identification data includes identification data of multiple guidance actions; the identification data of each guidance action includes: at least one signal quantity of the guidance action; obtaining the change rate of the signal quantity of each guidance action based on at least one signal quantity of each guidance action; obtaining the calibrated change rate based on the change rate of the signal quantities of the multiple guidance actions.

[0094] Optionally, the calibrated change rate is an average value of the change rates of the signal quantities of the multiple guiding actions.

[0095] Optionally, the device may further include a reference value interval updating module, configured to update the first reference value interval and the second reference value interval using a preset strategy.

[0096] The reference value interval updating module is specifically used to update the change rate boundary values ​​of the first reference value interval and the second reference value interval in the following manner: record the maximum and minimum values ​​of the change rate of the signal amount of each flexion action in the flexion action and the maximum and minimum values ​​of the change rate of the signal amount of each stretching action in the stretching action; if the maximum value of the change rate of the signal amount of each flexion action in the flexion action counted for K consecutive times is less than the maximum value of the change rate of the previous first reference value interval, or the minimum value of the change rate of the signal amount of each flexion action in the flexion action counted for K consecutive times is greater than the minimum value of the change rate of the previous first reference value interval, then take the average value of the maximum value of the change rate of the signal amount of the flexion action counted for K consecutive times as the maximum value of the change rate of the updated first reference value interval, and take the average value of the maximum value of the change rate of the signal amount of the flexion action counted for K consecutive times as the maximum value of the change rate of the updated first reference value interval, and take the The average value of the minimum value of the rate of change of the signal amount of the flexion action counted K times is used as the minimum value of the rate of change of the first reference value interval after update; if the maximum value of the rate of change of the signal amount of each stretching action in the stretching action counted K times in succession is less than the maximum value of the rate of change of the previous second reference value interval, or the minimum value of the rate of change of the signal amount of each stretching action in the stretching action counted K times in succession is greater than the minimum value of the rate of change of the previous first reference value interval, then the average value of the maximum value of the rate of change of the signal amount of the stretching action counted K times is used as the maximum value of the rate of change of the second reference value interval after update, and the average value of the minimum value of the rate of change of the signal amount of the stretching action counted K times is used as the minimum value of the rate of change of the second reference value interval after update; K is a natural number greater than 1.

[0097] The device may also include: an initial reference value interval acquisition module, used to collect guiding action identification data; the guiding action identification data includes identification data of multiple guiding actions; the identification data of each guiding action includes: multiple signal quantities of flexion action and multiple signal quantities of extension action; the maximum value and minimum value of the rate of change of the signal quantity of each flexion action are obtained according to the multiple signal quantities of the flexion action, and the maximum value and minimum value of the rate of change of the first initial reference value interval of the first reference value interval are obtained according to the maximum value and minimum value of the rate of change of the signal quantity of multiple flexion actions; the maximum value and minimum value of the rate of change of the signal quantity of each extension action are obtained according to the multiple signal quantities of the extension action, and the maximum value and minimum value of the rate of change of the second initial reference value interval of the second reference value interval are obtained according to the maximum value and minimum value of the rate of change of the signal quantity of multiple extension actions.

[0098] Optionally, the target training parts include hands, wrists, elbows or knees.

[0099] Optionally, the action recognition signal is a voltage signal that can change with the flexion and extension of the target training part.

[0100] Compared with the prior art, the device of the embodiment of the present invention identifies the training action through the rate of change of the action recognition signal, eliminating the cumbersome operation of calibrating the voltage value of the guided action, which can greatly reduce the operating burden of patients and medical staff.

[0101] Figure 3 The structure diagram of the training action recognition device provided in the third embodiment of the present invention is as follows. The training action recognition device 30 includes a memory 31, a processor 32, and a computer program stored in the memory 31 and executable on the processor 32. When the processor 32 executes the program, the technical solution described in the above method is implemented.

[0102] Compared with the prior art, the device of the embodiment of the present invention identifies the training action through the rate of change of the action recognition signal, eliminating the cumbersome operation of calibrating the voltage value of the guided action, which can greatly reduce the operating burden of patients and medical staff.

[0103] Embodiment 4 of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a computer processor, it is used to execute the technical solution of any method embodiment.

[0104] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, a read-only memory (ROM), a random access memory (RAM), a flash memory (FLASH), a hard disk or an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a grid device, etc.) to execute the methods described in each embodiment of the present invention.

[0105] It is worth noting that in the embodiment of the above-mentioned training action recognition device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0106] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A training action recognition device, applied to limb joint muscle rehabilitation training, characterized in that: include: The signal acquisition module is used to control the muscles of the target training part to be trained according to the guided action and obtain the training action recognition data in real time; The training action recognition data includes a plurality of training action recognition signal quantities; A change rate calculation module, used for obtaining a change rate of the training action recognition signal amount according to the plurality of training action recognition signal amounts; An identification module, used for identifying the training action according to the change rate of the training action identification signal and a preset judgment strategy; The recognition module is specifically configured to determine that the training action is a flexion action if the change rate of the training action recognition signal quantity within a first preset time period is greater than 0, the values ​​of the change rate of the training action recognition signal quantity are within a first reference value interval, and the number of the change rates of the training action recognition signal quantity is greater than or equal to a first threshold value; The recognition module is further configured to determine that the training action is a stretching action if the change rate of the training action recognition signal quantity is less than 0 within a second preset time period, and the absolute value of the change rate of the training action recognition signal quantity is within a second reference value interval, and the number of the change rates of the training action recognition signal quantity is greater than a second threshold value; The device also includes a reference value interval updating module, which is used to update the first reference value interval and the second reference value interval using a preset strategy; The reference value interval updating module is specifically used to update the change rate boundary values ​​of the first reference value interval and the second reference value interval in the following manner: record the maximum value and minimum value of the change rate of the signal amount of each flexion action in the flexion action and the maximum value and minimum value of the change rate of the signal amount of each extension action in the extension action; if the maximum value of the change rate of the signal amount of each flexion action in the flexion action counted for K consecutive times is less than the maximum value of the change rate of the previous first reference value interval, or the minimum value of the change rate of the signal amount of each flexion action in the flexion action counted for K consecutive times is greater than the minimum value of the change rate of the previous first reference value interval, then the average value of the maximum value of the change rate of the signal amount of the flexion action counted for the K times is used as the maximum value of the change rate of the first reference value interval after update, and the average value of the minimum value of the change rate of the signal amount of the flexion action counted for the K times is used as the minimum value of the change rate of the first reference value interval after update; If the maximum value of the rate of change of the signal amount of each stretching action in the stretching actions counted for K consecutive times is less than the maximum value of the rate of change of the previous second reference value interval, or the minimum value of the rate of change of the signal amount of each stretching action in the stretching actions counted for K consecutive times is greater than the minimum value of the rate of change of the previous first reference value interval, then the average value of the maximum values ​​of the rate of change of the signal amount of the stretching action counted for the K times is used as the maximum value of the rate of change of the second reference value interval after update, and the average value of the minimum values ​​of the rate of change of the signal amount of the stretching action counted for the K times is used as the minimum value of the rate of change of the second reference value interval after update; K is a natural number greater than 1.

2. The training action recognition device according to claim 1, characterized in that: include: The recognition module is also used to determine the degree of completion of the training action according to the change rate of the training action recognition signal and the calibrated change rate determined based on the guiding action signal.

3. The training action recognition device according to claim 2, characterized in that: include: The recognition module is specifically used to calculate the ratio of the change rate of each training action recognition signal quantity to the calibrated change rate; The degree of completion of the training action is determined according to the ratio.

4. The training action recognition device according to claim 2, characterized in that: The device also includes a change rate calibration module, which is used to obtain the calibration change rate in the following manner: Collecting guidance action recognition data; the guidance action recognition data includes recognition data of multiple guidance actions; The identification data of each guiding action includes: at least one signal quantity of the guiding action; the change rate of the signal quantity of each guiding action obtained according to the at least one signal quantity of each guiding action; and the calibration change rate obtained according to the change rate of the signal quantity of the multiple guiding actions.

5. The training action recognition device according to claim 4, characterized in that: The calibrated change rate is an average value of the change rates of the signal quantities of the multiple guiding actions.

6. The training action recognition device according to claim 1, characterized in that: The device also includes: an initial reference value interval acquisition module, which is used to collect guiding action identification data; the guiding action identification data includes identification data of multiple guiding actions; the identification data of each guiding action includes: multiple signal quantities of flexion action and multiple signal quantities of extension action; the maximum value and minimum value of the change rate of the signal quantity of each flexion action are obtained according to the multiple signal quantities of the flexion action, and the maximum value and minimum value of the change rate of the first initial reference value interval of the first reference value interval are obtained according to the maximum value and minimum value of the change rate of the signal quantity of multiple flexion actions; the maximum value and minimum value of the change rate of the signal quantity of each extension action are obtained according to the multiple signal quantities of the extension action, and the maximum value and minimum value of the change rate of the second initial reference value interval of the second reference value interval are obtained according to the maximum value and minimum value of the change rate of the signal quantity of multiple extension actions.

7. The training action recognition device according to claim 6, characterized in that: The target training parts include hands, wrist joints, elbow joints or knee joints.

8. The training action recognition device according to claim 6, characterized in that: The action recognition signal quantity is a voltage signal that can change with the flexion and extension of the target training part.

9. A training motion recognition device, characterized in that: include: A memory and a processor, the memory stores a computer program, and the processor runs the computer program to implement a training action recognition method, the method comprising: Controlling the muscles of the target training part to be trained according to the guiding action, and acquiring training action recognition data in real time; the training action recognition data includes a plurality of training action recognition signal quantities; Obtaining a change rate of a training action recognition signal amount according to the plurality of training action recognition signal amounts; Identify the training action according to the rate of change of the training action recognition signal and the relationship between the rate of change of the action recognition signal and the action; The identifying the training action according to the change rate of the training action recognition signal and the relationship between the change rate of the action recognition signal and the action includes: If the change rate of the training movement recognition signal amount within the first preset time length is greater than 0, and the value of the change rate of the training movement recognition signal amount is within the first reference value interval, and the number of change rates of the training movement recognition signal amount is greater than or equal to the first threshold, then the training movement is determined to be a flexion movement; The identifying the training action according to the change rate of the training action recognition signal and the relationship between the change rate of the action recognition signal and the action also includes: If the change rate of the training action recognition signal quantity within the second preset time length is less than 0, and the absolute value of the change rate of the training action recognition signal quantity is within the second reference value interval, and the number of change rates of the training action recognition signal quantity is greater than the second threshold, then the training action is determined to be a stretching action; The first reference value interval and the second reference value interval are updated using a preset strategy; The method further includes updating the change rate boundary values ​​of the first reference value interval and the second reference value interval in the following manner: Recording the maximum and minimum values ​​of the rate of change of the signal amount of each flexion action in the flexion action and the maximum and minimum values ​​of the rate of change of the signal amount of each extension action in the extension action; If the maximum value of the rate of change of the signal amount of each flexing action in the flexing actions counted for K consecutive times is less than the maximum value of the rate of change of the previous first reference value interval, or the minimum value of the rate of change of the signal amount of each flexing action in the flexing actions counted for K consecutive times is greater than the minimum value of the rate of change of the previous first reference value interval, then the average value of the maximum values ​​of the rate of change of the signal amount of the flexing action counted for the K consecutive times is used as the maximum value of the rate of change of the first reference value interval after update, and the average value of the minimum values ​​of the rate of change of the signal amount of the flexing action counted for the K consecutive times is used as the minimum value of the rate of change of the first reference value interval after update; If the maximum value of the rate of change of the signal amount of each stretching action in the stretching action counted for K consecutive times is less than the maximum value of the rate of change of the previous second reference value interval, or the minimum value of the rate of change of the signal amount of each stretching action in the stretching action counted for K consecutive times is greater than the minimum value of the rate of change of the previous first reference value interval, then the average value of the maximum value of the rate of change of the signal amount of the stretching action counted for the K times is used as the maximum value of the rate of change of the updated second reference value interval, and the average value of the minimum value of the rate of change of the signal amount of the stretching action counted for the K times is used as the minimum value of the rate of change of the updated second reference value interval; K is a natural number greater than 1.

10. The training motion recognition device according to claim 9, characterized in that: The step of determining the degree of completion of the training action according to the change rate of the training action recognition signal and the calibrated change rate determined based on the guiding action signal comprises: Calculating the ratio of the change rate of each training action recognition signal to the calibrated change rate; The degree of completion of the training action is determined according to the ratio.

11. The training motion recognition device according to claim 10, characterized in that: The step of determining the degree of completion of the training action according to the change rate of the training action recognition signal and the calibrated change rate determined based on the guiding action signal comprises: Calculating the ratio of the change rate of each training action recognition signal to the calibrated change rate; The degree of completion of the training action is determined according to the ratio.

12. The training motion recognition device according to claim 10, characterized in that: The method further comprises: The calibration change rate is obtained in the following manner: Collecting guidance action identification data; the guidance action identification data includes identification data of multiple guidance actions; the identification data of each guidance action includes: at least one signal quantity of the guidance action; Obtaining a rate of change of the signal quantity of each guiding action according to at least one signal quantity of each guiding action; The calibrated change rate is obtained according to the change rate of the signal quantity of the multiple guiding actions.

13. The training motion recognition device according to claim 12, characterized in that: The calibrated change rate is an average value of the change rates of the signal quantities of the multiple guiding actions.

14. The training motion recognition device according to claim 9, wherein the method further comprises: Collecting guidance action recognition data; The guiding action identification data includes identification data of multiple guiding actions; The identification data of each guided action includes: a plurality of signal quantities of flexion action and a plurality of signal quantities of extension action; Obtaining a maximum value and a minimum value of a rate of change of the signal quantity of each flexion action according to multiple signal quantities of the flexion action, and obtaining a maximum value and a minimum value of a rate of change of a first initial reference value interval of the first reference value interval according to the maximum value and the minimum value of the rate of change of the signal quantity of multiple flexion actions; The maximum and minimum values ​​of the change rate of the signal quantity of each stretching action are obtained according to the multiple signal quantities of the stretching action, and the maximum and minimum values ​​of the change rate of the second initial reference value interval of the second reference value interval are obtained according to the maximum and minimum values ​​of the change rate of the signal quantity of multiple stretching actions.

15. The training motion recognition device according to claim 14, wherein the method further comprises: The target training parts include hands, wrist joints, elbow joints or knee joints.

16. The training motion recognition device according to claim 14, wherein the method further comprises: The action recognition signal quantity is a voltage signal that can change with the flexion and extension of the target training part.

17. A computer-readable storage medium having a computer program stored thereon, characterized in that When the program is executed by the processor, a training action recognition method is implemented, and the method includes: Controlling the muscles of the target training part to be trained according to the guiding action, and acquiring training action recognition data in real time; the training action recognition data includes a plurality of training action recognition signal quantities; Obtaining a change rate of a training action recognition signal amount according to the plurality of training action recognition signal amounts; Identify the training action according to the rate of change of the training action recognition signal and the relationship between the rate of change of the action recognition signal and the action; The identifying the training action according to the change rate of the training action recognition signal and the relationship between the change rate of the action recognition signal and the action includes: If the change rate of the training movement recognition signal amount within the first preset time length is greater than 0, and the value of the change rate of the training movement recognition signal amount is within the first reference value interval, and the number of change rates of the training movement recognition signal amount is greater than or equal to the first threshold, then the training movement is determined to be a flexion movement; The identifying the training action according to the change rate of the training action recognition signal and the relationship between the change rate of the action recognition signal and the action also includes: If the change rate of the training action recognition signal quantity within the second preset time length is less than 0, and the absolute value of the change rate of the training action recognition signal quantity is within the second reference value interval, and the number of change rates of the training action recognition signal quantity is greater than the second threshold, then the training action is determined to be a stretching action; The first reference value interval and the second reference value interval are updated using a preset strategy; The method further includes updating the change rate boundary values ​​of the first reference value interval and the second reference value interval in the following manner: Recording the maximum and minimum values ​​of the rate of change of the signal amount of each flexion action in the flexion action and the maximum and minimum values ​​of the rate of change of the signal amount of each extension action in the extension action; If the maximum value of the rate of change of the signal amount of each flexing action in the flexing actions counted for K consecutive times is less than the maximum value of the rate of change of the previous first reference value interval, or the minimum value of the rate of change of the signal amount of each flexing action in the flexing actions counted for K consecutive times is greater than the minimum value of the rate of change of the previous first reference value interval, then the average value of the maximum values ​​of the rate of change of the signal amount of the flexing action counted for the K consecutive times is used as the maximum value of the rate of change of the first reference value interval after update, and the average value of the minimum values ​​of the rate of change of the signal amount of the flexing action counted for the K consecutive times is used as the minimum value of the rate of change of the first reference value interval after update; If the maximum value of the rate of change of the signal amount of each stretching action in the stretching action counted for K consecutive times is less than the maximum value of the rate of change of the previous second reference value interval, or the minimum value of the rate of change of the signal amount of each stretching action in the stretching action counted for K consecutive times is greater than the minimum value of the rate of change of the previous first reference value interval, then the average value of the maximum value of the rate of change of the signal amount of the stretching action counted for the K times is used as the maximum value of the rate of change of the updated second reference value interval, and the average value of the minimum value of the rate of change of the signal amount of the stretching action counted for the K times is used as the minimum value of the rate of change of the updated second reference value interval; K is a natural number greater than 1.

18. The computer-readable storage medium according to claim 17, characterized in that: The step of determining the degree of completion of the training action according to the change rate of the training action recognition signal and the calibrated change rate determined based on the guiding action signal comprises: Calculating the ratio of the change rate of each training action recognition signal to the calibrated change rate; The degree of completion of the training action is determined according to the ratio.

19. The computer-readable storage medium according to claim 18, characterized in that The step of determining the degree of completion of the training action according to the change rate of the training action recognition signal and the calibrated change rate determined based on the guiding action signal comprises: Calculating the ratio of the change rate of each training action recognition signal to the calibrated change rate; The degree of completion of the training action is determined according to the ratio.

20. The computer-readable storage medium according to claim 18, wherein: The method further comprises: The calibration change rate is obtained in the following manner: Collecting guidance action identification data; the guidance action identification data includes identification data of multiple guidance actions; the identification data of each guidance action includes: at least one signal quantity of the guidance action; Obtaining a rate of change of the signal quantity of each guiding action according to at least one signal quantity of each guiding action; The calibrated change rate is obtained according to the change rate of the signal quantity of the multiple guiding actions.

21. The computer-readable storage medium according to claim 20, characterized in that The calibrated change rate is an average value of the change rates of the signal quantities of the multiple guiding actions.

22. The computer-readable storage medium according to claim 17, wherein: The method further comprises: Collecting guidance action recognition data; the guidance action recognition data includes recognition data of multiple guidance actions; the recognition data of each guidance action includes: multiple signal quantities of flexion action and multiple signal quantities of extension action; Obtaining a maximum value and a minimum value of a rate of change of the signal quantity of each flexion action according to multiple signal quantities of the flexion action, and obtaining a maximum value and a minimum value of a rate of change of a first initial reference value interval of the first reference value interval according to the maximum value and the minimum value of the rate of change of the signal quantity of multiple flexion actions; The maximum and minimum values ​​of the change rate of the signal quantity of each stretching action are obtained according to the multiple signal quantities of the stretching action, and the maximum and minimum values ​​of the change rate of the second initial reference value interval of the second reference value interval are obtained according to the maximum and minimum values ​​of the change rate of the signal quantity of multiple stretching actions.

23. The computer-readable storage medium according to claim 22, characterized in that The method also includes: the target training part includes a hand, a wrist joint, an elbow joint or a knee joint.

24. The computer-readable storage medium according to claim 22, characterized in that The method also includes: the action recognition signal quantity is a voltage signal that can change with the flexion and extension of the target training part.

Images