Action number statistical method and device and computer equipment

Abstract

The invention relates to an action quantity statistical method and device and computer equipment. The method comprises the following steps: acquiring motion state data of a target object at different sampling moments in a sampling time period; for any sampling moment, determining a trunk inclination angle corresponding to the target object at the sampling moment according to the motion state data at the sampling moment; determining inclination angle change information according to the corresponding trunk inclination angle of the target object at each sampling moment; and determining the number of target actions completed by the target object in the sampling time period according to the inclination angle change information. By adopting the method, the completed target action quantity is determined according to the change information of the trunk inclination angles of the target object at different sampling moments, and the accuracy of action quantity statistics can be improved.

Description

Technical Field

[0001] This application relates to the field of smart wearable technology, and in particular to a method, apparatus and computer device for counting the number of actions. Background Technology

[0002] With the improvement of living standards and the growing popularity of fitness concepts, exercise has gradually evolved into an important activity in daily work and life. Whether it's fitness enthusiasts aiming for scientific fitness results or individuals seeking accurate test results in physical fitness assessments, the number of exercises performed has become a core indicator of interest.

[0003] Currently, the counting of fitness movements is usually done by the exerciser themselves during the exercise, or with the assistance of a third party (such as a fitness coach or tester).

[0004] However, manually counting the number of fitness movements is highly subjective and cannot accurately identify the effectiveness of the movements, leading to inaccurate movement counts. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, apparatus, and computer device for calculating the number of actions that can improve the accuracy of action counting, in order to address the aforementioned technical problems.

[0006] Firstly, this application provides a method for counting the number of actions, applied to smart wearable devices, including:

[0007] Acquire motion state data of the target object at different sampling times within the sampling period;

[0008] For any of the sampling times, the torso tilt angle of the target object at the sampling time is determined based on the motion state data at the sampling time.

[0009] Based on the torso tilt angle of the target object at each sampling time, determine the tilt angle change information;

[0010] Based on the tilt angle change information, the number of target actions completed by the target object during the sampling period is determined.

[0011] In one embodiment, determining the number of target actions completed by the target object during the sampling period based on the tilt angle change information includes: determining the angle change trajectory of the torso tilt angle during the sampling period based on the tilt angle change information; wherein the angle change trajectory is: the trajectory of the torso tilt angle changing from the initial angle to the peak angle and then returning to the initial angle from the peak angle; and determining the number of target actions of the target object during the sampling period based on the angle change trajectory corresponding to the torso tilt angle during the sampling period.

[0012] In one embodiment, determining the number of target actions completed by the target object within the sampling period based on the angle change trajectory corresponding to the torso tilt angle within the sampling period includes: obtaining the target duration consumed by any angle change trajectory; determining the angle change trajectory corresponding to the target duration that meets the target duration condition as the target change trajectory; and determining the number of target actions completed by the target object within the sampling period based on the number of target change trajectories.

[0013] In one embodiment, determining the number of target actions completed by the target object within the sampling period based on the angle change trajectory corresponding to the torso tilt angle within the sampling period includes: for any angle change trajectory, determining the motion state characteristics within the angle change trajectory based on the motion state data at each sampling time within the angle change trajectory; comparing the motion state characteristics with the standard state characteristics to determine the abnormal action detection result within the angle change trajectory; and determining the number of target actions completed by the target object within the sampling period based on the abnormal action detection results within each angle change trajectory.

[0014] In one embodiment, determining the number of target actions completed by the target object within the sampling period based on the abnormal action detection results within each of the angle change trajectories includes: for any angle change trajectory, if the abnormal action detection results within the angle change trajectory indicate that there are no abnormal actions, determining the angle change trajectory as a target change trajectory; and determining the number of target actions completed by the target object within the sampling period based on the number of determined target change trajectories.

[0015] In one embodiment, determining the angle change trajectory corresponding to the target duration that satisfies the target duration condition as the target change trajectory includes: for any angle change trajectory, if the target duration of the angle change trajectory satisfies the target duration condition, obtaining the rising duration and falling duration consumed by the angle change trajectory; wherein, the rising duration is: the time consumed for the torso tilt angle to change from the starting angle to the peak angle within the angle change trajectory, and the falling duration is: the time consumed for the torso tilt angle to recover from the peak angle to the starting angle within the angle change trajectory; if the rising duration is within the rising duration range and the falling duration is within the falling duration range, the angle change trajectory is determined as the target change trajectory.

[0016] In one embodiment, the rising duration includes a first duration for the trunk tilt angle to change from the initial angle to a rising angle threshold, and a second duration for the trunk tilt angle to change from the rising angle threshold to the peak angle; the falling duration includes a third duration for the trunk tilt angle to change from the peak angle to a falling angle threshold, and a fourth duration for the trunk tilt angle to change from the falling angle threshold to the initial angle; the step of determining the angle change trajectory as the target change trajectory if the rising duration belongs to the rising duration range and the falling duration belongs to the falling duration range includes: if the first duration belongs to the first duration range, the second duration belongs to the second duration range, the third duration belongs to the third duration range, and the fourth duration belongs to the fourth duration range, then the angle change trajectory is determined as the target change trajectory.

[0017] In one embodiment, the smart wearable device is worn on a target part of the target object; determining the torso tilt angle of the target object at the sampling time based on the motion state data at the sampling time includes: processing the motion state data at the sampling time to obtain the posture angle of the target part at the sampling time; converting the posture angle of the target part according to a preset conversion coefficient between the posture angle and the torso tilt angle to obtain the torso tilt angle of the target object at the sampling time.

[0018] In one embodiment, the method further includes: determining a motion amplitude score based on the minimum and maximum torso tilt angles contained in each angle change trajectory; determining a motion stability score based on the similarity between different angle change trajectories; determining a motion speed score based on the target duration consumed by each angle change trajectory; and determining a comprehensive motion score for the target object based on the motion amplitude score, motion stability score, and motion speed score.

[0019] In one embodiment, the step of comparing the motion state features and the standard state features to determine the abnormal action detection result within the angle change trajectory includes: comparing the motion state features and the standard state features to obtain deviation features in the motion state features that are different from the standard state features; if the degree of matching between the deviation features and the motion features of any abnormal action is higher than the degree of matching threshold, the abnormal action detection result within the angle change trajectory is determined to indicate the presence of an abnormal action.

[0020] Secondly, this application also provides a motion counting device, deployed in a smart wearable device, comprising:

[0021] The acquisition module is used to acquire motion state data of the target object at different sampling times within the sampling period;

[0022] The processing module is used to determine the torso tilt angle of the target object at any of the sampling times, based on the motion state data at the sampling time.

[0023] The information determination module is used to determine the tilt angle change information based on the torso tilt angle of the target object at each sampling time.

[0024] The quantity determination module is used to determine the number of target actions completed by the target object during the sampling period based on the tilt angle change information.

[0025] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the various method embodiments provided in the first aspect above.

[0026] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the various method embodiments provided in the first aspect above.

[0027] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the various method embodiments provided in the first aspect above.

[0028] The aforementioned method, apparatus, and computer equipment for counting the number of actions acquire motion state data of the target object at different sampling times within a sampling period through a smart wearable device. This allows for comprehensive and accurate capture of the target object's motion trajectory when performing the target action. Furthermore, for any given sampling time, based on the motion state data at that time, the torso tilt angle of the target object is determined. Based on the torso tilt angles at various sampling times, the tilt angle change information is determined, and thus, the number of target actions performed by the target object within the sampling period is determined based on the tilt angle change information. By determining the number of completed target actions based on the torso tilt angle changes of the target object at different sampling times, this method accurately identifies whether the target actions performed by the target object meet the specifications, technically avoiding the subjective bias caused by manual statistics, thereby improving the accuracy of the determined number of target actions. Moreover, the number of target actions can be counted using a smart wearable device without relying on third-party assistance, overcoming the time and space limitations of manual statistics and accurately counting actions even in single-person motion scenarios. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a diagram illustrating the application environment of the action count method in one embodiment;

[0031] Figure 2 This is a flowchart illustrating a method for counting the number of actions in one embodiment;

[0032] Figure 3 This is a flowchart illustrating the steps for determining the target number of actions in one embodiment;

[0033] Figure 4 This is a flowchart illustrating the steps for determining the target number of actions in another embodiment;

[0034] Figure 5 This is a flowchart illustrating the steps for determining the target number of actions in yet another embodiment;

[0035] Figure 6 This is a flowchart illustrating the action scoring steps in one embodiment;

[0036] Figure 7 This is a flowchart illustrating the action quantity counting method in another embodiment;

[0037] Figure 8 This is a structural block diagram of an action count counting device in one embodiment;

[0038] Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0040] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0041] The design concept of this application is introduced below:

[0042] With the improvement of living standards and the growing popularity of fitness concepts, exercise has gradually evolved into an important activity in daily work and life. Whether it's fitness enthusiasts aiming for scientific fitness results or individuals seeking accurate test results in physical fitness assessments, the number of exercises performed has become a core indicator of interest.

[0043] Currently, the counting of fitness exercises is usually done either by the exerciser themselves during the workout or with the assistance of a third party (such as a fitness coach or tester). However, this manual counting method is highly subjective and cannot accurately identify the effectiveness of the exercises, leading to inaccurate counts.

[0044] In view of this, a method, device, and computer equipment for counting the number of actions are proposed. By acquiring motion state data of a target object at different sampling times within a sampling period using a smart wearable device, the motion trajectory of the target object during the completion of the target action can be captured comprehensively and accurately. Then, for any sampling time, based on the motion state data at that time, the torso tilt angle of the target object at that sampling time is determined. Based on the torso tilt angle of the target object at each sampling time, the tilt angle change information is determined, and thus, based on the tilt angle change information, the number of target actions completed by the target object within the sampling period is determined. By determining the number of completed target actions based on the torso tilt angle change information of the target object at different sampling times, the method can accurately identify whether the target actions completed by the target object meet the specifications, technically avoiding the subjective bias caused by manual statistics, thereby improving the accuracy of the determined number of target actions. Furthermore, the number of target actions can be counted using a smart wearable device without relying on third-party assistance, breaking through the time and space limitations of manual statistics, and accurately counting the number of actions even in single-person motion scenarios.

[0045] The action quantity counting method provided in this application embodiment can be applied to, for example, Figure 1 In the application environment shown, the smart wearable device 102 communicates with the server 104 via a network. The data storage system can store the data that the server 104 needs to process, such as the number of target actions completed by the user. The data storage system can be integrated on the server 104 or placed in the cloud or on another network server. In one embodiment, the smart wearable device 102 acquires motion state data of the target object at different sampling times within a sampling period and sends this motion state data to the server 104. For any given sampling time, the server 104 determines the torso tilt angle of the target object at that sampling time based on the motion state data, determines the tilt angle change information based on the torso tilt angle of the target object at each sampling time, and determines the number of target actions completed by the target object within the sampling period based on the tilt angle change information. The smart wearable device 102 can be, but is not limited to, a smartwatch, smart bracelet, or head-mounted device. The server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0046] In one exemplary embodiment, such as Figure 2 As shown, a method for counting the number of actions is provided, which can be applied to... Figure 1 Taking smart wearable devices as an example, this will be explained, including:

[0047] S201, Obtain motion state data of the target object at different sampling times within the sampling period.

[0048] Motion state data is used to characterize the motion and behavioral features of the target object at the sampling time. For example, motion state data may include acceleration data of the target object's body movement and angular velocity data of its rotation. The smart wearable device can use multiple sensors to collect motion state data. Optionally, the smart wearable device is equipped with an accelerometer to collect linear acceleration data and a gyroscope to collect angular velocity data. For example, the smart wearable device may be equipped with a three-axis accelerometer and a three-axis gyroscope, with the three-axis accelerometer having a range of ±8 standard gravitational acceleration and the three-axis gyroscope having a range of ±500 degrees / second. Optionally, the smart wearable device may also be equipped with a three-axis magnetometer for assisting in device orientation and a heart rate sensor for monitoring exercise intensity. It should be noted that the sensors listed above for smart wearable devices are only illustrative examples; the specific sensors installed in the smart wearable device can be configured according to actual needs, and no specific limitations are made here.

[0049] Each sensor can acquire data at a preset sampling frequency, which is set according to actual needs and is not specifically limited here. Optionally, for the data acquired by a single sensor, firstly, bandpass filtering is performed on the acquired data to remove high-frequency noise and gravity component interference, resulting in processed acquired data. Then, a sliding window segmentation technique is used, employing a 2-second data window to slide across the processed acquired data. The average value of the data within a data window is taken as the motion state data corresponding to the center time of the data window. A 50% overlap rate is set between adjacent data windows to ensure data continuity. Finally, signal standardization processing is performed on the motion state data corresponding to any center time to eliminate the influence of device differences, and the standardized motion state data corresponding to the center time is taken as the motion state data corresponding to the sampling time.

[0050] S202, for any sampling time, determine the torso tilt angle of the target object at the sampling time based on the motion state data at the sampling time.

[0051] The trunk tilt angle describes the angle by which the target object's trunk deviates from the midline. The midline can be set according to the target action to be statistically analyzed during the sampling period. For example, if the target action is a sit-up, the midline can be the straight line where the target object's trunk lies when it is in a supine position. Optionally, the motion state data at the sampling time includes acceleration data and angular velocity data. Based on the acceleration data and angular velocity data, the trunk tilt angle of the target object at the sampling time is determined.

[0052] S203, determine the tilt angle change information based on the torso tilt angle of the target object at each sampling time.

[0053] The tilt angle change information is used to describe the change of the trunk tilt angle at each sampling time. For example, the tilt angle change information may include the trunk tilt angle change curve, the maximum tilt angle and the minimum tilt angle at each sampling time, etc., without specific limitations.

[0054] S204, Based on the tilt angle change information, determine the number of target actions completed by the target object during the sampling period.

[0055] Understandably, during the repeated completion of the target action, the target subject's trunk inclination angle should exhibit a certain pattern of change. Optional target actions include sit-ups, crunches, and sit-and-reach exercises. Taking sit-ups as an example, the trunk inclination angle should increase from 0° to a peak angle and then decrease back to 0° after each completed target action. Therefore, the number of target actions completed by the target subject can be determined based on the inclination angle change information. The following explanation primarily uses sit-ups as an example.

[0056] In the above method, the action quantity counting method, device, and computer equipment acquire motion state data of the target object at different sampling times within a sampling period through a smart wearable device. This allows for comprehensive and accurate capture of the target object's motion trajectory when performing the target action. Furthermore, for any given sampling time, based on the motion state data at that time, the torso tilt angle of the target object is determined. Based on the torso tilt angles of the target object at each sampling time, tilt angle change information is determined. Therefore, based on the tilt angle change information, the number of target actions performed by the target object within the sampling period is determined. By determining the number of completed target actions based on the torso tilt angle changes of the target object at different sampling times, the method accurately identifies whether the target actions performed by the target object meet the specifications, technically avoiding the subjective bias caused by manual statistics, thereby improving the accuracy of the determined target action count. Moreover, the target action count can be completed through a smart wearable device without relying on third-party assistance, breaking through the time and space limitations of manual statistics and accurately completing action quantity counting even in single-person motion scenarios.

[0057] In one embodiment, the smart wearable device is worn on a target part of the target object; the step of determining the torso tilt angle of the target object is further refined to include:

[0058] The motion state data at the sampling time is processed to obtain the attitude angle of the target part at the sampling time; the attitude angle of the target part is converted according to the preset conversion coefficient between the attitude angle and the torso tilt angle to obtain the torso tilt angle of the target object at the sampling time.

[0059] The target body part worn by the smart wearable device is set according to the type of target action. For example, if the target action is a sit-up, the target body part can be the wrist. It is understandable that, since the smart wearable device is worn on the target body part, only the posture angle of the target body part can be calculated based on motion data. However, during the sit-up, the relative position between the target body part and the torso is fixed; that is, the target's hands must always be behind their head. Therefore, after calculating the posture angle of the target body part, it can be converted into a torso tilt angle according to a preset conversion coefficient. This preset conversion coefficient can be set based on experience, multiple trials, and actual needs, and is not specifically limited here. Optionally, the preset conversion coefficient can be determined as follows: First, guide the user to maintain a standard supine posture, and calculate the first posture angle of the target body part in the standard supine posture based on the motion data collected by the smart wearable device. Then, guide the user to maintain a standard sitting posture, and calculate the second posture angle of the target body part in the standard sitting posture based on the motion data collected by the smart wearable device. Finally, determine the first and second posture angles, as well as the trunk tilt angle in the standard supine posture and the trunk tilt angle in the standard sitting posture, and determine the preset conversion coefficient. The trunk tilt angle in the standard supine posture can be obtained through actual measurement.

[0060] In the above embodiments, the posture angle of the target part is converted into the torso tilt angle by a preset conversion coefficient, which improves the accuracy of the torso tilt angle. In turn, the number of target actions completed by the target object can be counted more accurately based on the torso tilt angle, thereby improving the accuracy of the number of target actions.

[0061] In one embodiment, such as Figure 3 As shown, the steps for determining the number of target actions in S204 are further refined, including:

[0062] S301, based on the tilt angle change information, determine the trajectory of the torso tilt angle change during the sampling period.

[0063] The angle change trajectory is the change in trunk tilt angle from the initial angle to the peak angle, and then back to the initial angle. In other words, within the sampling period, each time a change in trunk tilt angle from the initial angle to the peak angle and back to the initial angle is detected, one angle change trajectory can be determined. If the sampling period is long enough, multiple angle change trajectories may be detected within that period. The initial angle and peak angle can be set based on experience with the target movement, multiple trials, and actual needs; no specific limitations are set here. For example, if the target movement is a sit-up, the initial angle can be 0 degrees, and the peak angle can be 90 degrees.

[0064] S302, determine the number of target actions of the target object during the sampling period based on the angle change trajectory corresponding to the torso tilt angle during the sampling period.

[0065] Understandably, during the process of a target object completing a target action, the determined trajectory of a torso tilt angle represents the target object's attempt to complete a target action. Optionally, the number of determined angle change trajectories can be used as the number of target actions of the target object within the sampling period. Furthermore, to improve the accuracy of the determined number of target actions, the angle change trajectories can be filtered, and the number of filtered angle change trajectories can be used as the number of target actions. For example, based on the duration of the angle change trajectories, the trajectories can be filtered, retaining those with a duration less than a duration threshold. The duration threshold can be set based on experience, multiple trials, and actual needs.

[0066] In the above embodiments, the number of target movements is determined based on the angular change trajectory corresponding to the torso tilt angle. The number of target movements can be counted without human intervention, which enables the counting of target movements to be completed in a single-person exercise scenario, thus improving the flexibility of fitness exercises.

[0067] In one embodiment, such as Figure 4 As shown, the steps for determining the number of target actions in S302 are further refined, including:

[0068] S401: For any angle change trajectory, obtain the target duration of the angle change trajectory.

[0069] The target duration represents the time required for the torso tilt angle to change from the initial angle to the peak angle and then back to the initial angle.

[0070] S402, the angle change trajectory corresponding to the target duration that meets the target duration condition is determined as the target change trajectory.

[0071] The target duration condition is used to filter the trajectory of angle change. For example, the target duration condition can be that the target duration belongs to a preset duration range. The preset duration range can be set according to experience, multiple tests and actual needs. For example, the preset duration range can be 3 seconds to 5 seconds, or it can be greater than 3 seconds or less than 5 seconds.

[0072] S403, based on the number of target change trajectories, determine the number of target actions completed by the target object within the sampling period.

[0073] Specifically, the number of target change trajectories can be used as the number of target actions. For example, if the number of target change trajectories is 15, then the number of target actions completed by the target object during the sampling period is determined to be 15.

[0074] In the above embodiments, the angle change trajectory is determined as the target change trajectory only if the target time consumed by the angle change trajectory meets the target time condition. This ensures the effectiveness of the target action within the target change trajectory, eliminates invalid actions that are completed too quickly or too slowly, and improves the accuracy of the number of target actions.

[0075] In one embodiment, the step of determining the target's changing trajectory in S402 is further refined, including:

[0076] For any angle change trajectory, if the target duration of the angle change trajectory meets the target duration condition, obtain the rising duration and falling duration consumed by the angle change trajectory; if the rising duration is within the rising duration range and the falling duration is within the falling duration range, determine the angle change trajectory as the target change trajectory.

[0077] Specifically, for an angle change trajectory, to improve the accuracy of the determined target change trajectory, if the target duration of the angle change trajectory meets the target condition, the rising and falling durations of the angle change trajectory are further judged. The rising duration is the time taken for the torso tilt angle to change from the initial angle to the peak angle within the angle change trajectory, and the falling duration is the time taken for the torso tilt angle to recover from the peak angle to the initial angle within the angle change trajectory. The sum of the rising and falling durations constitutes the target duration of the angle change trajectory. In other words, the process of the target object completing the target action is divided into two parts: the rising process and the falling process. Not only does the total duration of the target object completing the target action (i.e., the target duration) need to meet the target duration condition, but the duration of the target object completing the rising process (i.e., rising duration) and the duration of the target object completing the falling process (i.e., falling duration) also need to meet their respective corresponding duration ranges (i.e., rising duration range and falling duration range). The rising duration range and falling duration range can be set based on experience, multiple trials, and actual needs. For example, the target duration condition is that the target duration is less than or equal to 6 seconds, the rising duration range is less than 3.5 seconds, the falling duration range is less than 3 seconds, and the target duration of angle change trajectory 1 is 6 seconds, which meets the target duration condition. Further, the rising duration is obtained as 4 seconds and the falling duration is 2 seconds. Since the rising duration does not fall within the rising duration range, angle change trajectory 1 cannot be determined as the target change trajectory.

[0078] In the above embodiments, the target duration is further divided into rising duration and falling duration. When the rising duration and falling duration are within their respective duration ranges, a target action is completed within the recorded angle change trajectory. This ensures the effectiveness of the target action, and the action recognition accuracy can reach over 95%, improving the accuracy of the target action count.

[0079] Based on dividing the target duration into rising duration and falling duration, the rising duration and falling duration can be further divided. In one embodiment, the rising duration includes the first duration taken for the trunk tilt angle to change from the starting angle to the rising angle threshold, and the second duration taken for the trunk tilt angle to change from the rising angle threshold to the peak angle; the falling duration includes the third duration taken for the trunk tilt angle to change from the peak angle to the falling angle threshold, and the fourth duration taken for the trunk tilt angle to change from the falling angle threshold to the starting angle.

[0080] The ascent and descent angle thresholds can be set based on experience, multiple trials, and actual needs. For example, the ascent angle threshold can be 45 degrees, and the descent angle threshold can be 15 degrees. In other words, the target duration of the angle change trajectory can be divided into four durations: a first duration, a second duration, a third duration, and a fourth duration. Each duration corresponds to a different stage in the target object's completion of the target action. For example, the stage where the torso tilt angle changes from the initial angle to the ascent angle threshold is considered the stationary stage; the stage where the torso tilt angle changes from the ascent angle threshold to the peak angle is considered the ascent stage; the stage where the torso tilt angle changes from the peak angle to the descent angle is considered the peak holding stage; and the stage where the torso tilt angle changes from the descent angle threshold to the initial angle is considered the descent stage. By constraining the duration of these four stages, the accuracy of determining the target change trajectory can be further improved. Optionally, further limitations can be made to determine the angle change trajectory as the target change trajectory, including:

[0081] If the first duration falls within the first duration range, the second duration falls within the second duration range, the third duration falls within the third duration range, and the fourth duration falls within the fourth duration range, then the angle change trajectory is determined as the target change trajectory.

[0082] The first, second, third, and fourth duration ranges can be set according to actual needs. For example, the first duration range can be less than 1.5 seconds, the second duration range less than 3 seconds, the third duration range less than 3 seconds, and the fourth duration range less than 1.5 seconds. The first duration of angle change trajectory 2 is 1 second, the second duration is 2 seconds, the third duration is 2 seconds, and the fourth duration is 1 second. It is determined that the first duration belongs to the first duration range, the second duration belongs to the second duration range, the third duration belongs to the third duration range, and the fourth duration belongs to the fourth duration range. Angle change trajectory 2 is then used as the target change trajectory.

[0083] In the above embodiments, the target duration is further divided into a first duration, a second duration, a third duration, and a fourth duration. By dividing the process of completing the target action into more refined stages, the entire trajectory of the sit-up movement can be fully tracked, thereby more accurately identifying effective movements and improving the accuracy of the number of target movements.

[0084] In one embodiment, such as Figure 5 As shown, the steps for determining the number of target actions in S302 are further refined, including:

[0085] S501, for any angle change trajectory, determine the motion state characteristics within the angle change trajectory based on the motion state data at each sampling time within the angle change trajectory.

[0086] Among them, motion state features are used to characterize the behavior state of the target object at the sampling time. Optionally, motion state features may include at least one of time domain features, frequency domain features and action features. Time domain features may include data such as mean, variance, peak value and zero crossing rate. Frequency domain features obtain data such as energy distribution, frequency peak value and spectral features through fast Fourier transform. Action features may include data such as action period, amplitude and speed.

[0087] S502, compare motion state features with standard state features to determine the abnormal action detection results within the trajectory of angle change.

[0088] It is understandable that standard state features are used to characterize the features of the target object in accurately completing the target action. Therefore, by comparing motion state features and standard state features, it is possible to determine whether there are abnormal actions within the angle change trajectory, and obtain abnormal action detection results. Abnormal action detection results are used to characterize whether there are abnormal actions within the angle change trajectory.

[0089] S503 determines the number of target actions completed by the target object within the sampling period based on the abnormal action detection results within the trajectory of each angle change.

[0090] Specifically, based on the abnormal action detection results within each angle change trajectory, it can be determined whether the target object has completed the target action within each angle change trajectory. Optionally, the number of first trajectories where no abnormal action exists, represented by the corresponding abnormal action detection results, can be used as the number of target actions completed by the target object within the sampling period. For example, if the sampling period includes angle change trajectories 1-20, and the abnormal action detection results for angle change trajectories 5, 12, and 14 all indicate the presence of abnormal actions, then the number of first trajectories is 17, meaning the number of target actions completed by the target object within the sampling period is 17.

[0091] The system can also comprehensively determine the number of target actions completed by the target object within the sampling period based on the abnormal action detection results within each angle change trajectory. For example, if the number of second trajectories indicating abnormal actions in the corresponding abnormal action detection results is greater than a threshold, then the number of target actions completed by the target object within the sampling period is determined to be zero. If the number of second trajectories is not greater than the threshold, then the number of first trajectories indicating no abnormal actions in the corresponding abnormal action detection results is taken as the number of target actions completed by the target object within the sampling period. The threshold can be set based on experience, multiple trials, and actual needs. It is understandable that if the number of second trajectories is too large, the reliability of angle change trajectories that do not detect abnormal actions is also low. Therefore, if the number of second trajectories is greater than the threshold, the number of target actions can be directly set to 0.

[0092] It should be noted that the above examples of determining the number of target actions based on the abnormal action detection results within the trajectory of angle change are for illustrative purposes only. The specific methods can be set according to actual needs and are not specifically limited here.

[0093] In the above embodiments, by comparing the motion state characteristics and standard state characteristics within the angle change trajectory, it is possible to identify whether there are abnormal movements within the angle change trajectory, thereby determining the number of target movements. This can effectively distinguish between sit-up-specific movement patterns and daily activities, improve the accuracy of the determined number of target movements, ensure the authenticity and reliability of the number of target movements, and achieve a counting error rate of less than 3%.

[0094] In one embodiment, the step of determining the abnormal action detection result in S502 is further refined to include:

[0095] By comparing the motion state features with the standard state features, deviation features that differ from the standard state features are obtained from the motion state features. If the degree of matching between the deviation features and the motion features of any abnormal action is higher than the degree of matching threshold, the abnormal action detection result within the angle change trajectory is determined to be the presence of an abnormal action.

[0096] Understandably, if the target object performs the target action with standard movements, then the motion state characteristics and standard state characteristics should be highly similar. However, if the target object's movements are not standard, the motion state characteristics will show deviation characteristics that differ from the standard state characteristics. These deviation characteristics can include time-domain and frequency-domain deviation characteristics. Therefore, by comparing the matching degree between the deviation characteristics and the motion characteristics of the abnormal movements, it can be determined whether there are abnormal movements within the angular change trajectory. The matching degree threshold can be set based on experience, multiple trials, and actual needs. Optionally, abnormal movements can include neck compensation movements, leg-assisted movements, incomplete movements, inertial-assisted movements, and abnormal device wearing movements, etc., which can be set according to actual needs.

[0097] Optionally, for detecting neck compensation movements, abnormal head shaking can be identified by analyzing high-frequency vibration patterns and compared with standard abdominal force exertion patterns to detect neck muscle compensation behavior. For example, for neck compensation movements, abnormal head shaking motion features can be pre-extracted, and the presence of neck compensation movements within the angle change trajectory can be determined by comparing the matching degree of deviation features with the abnormal head shaking motion features. For detecting leg-assisted movements, abnormal acceleration features generated by leg assistance can be monitored to identify violations such as leg assistance and hip lift. For example, for leg-assisted movements, leg assistance motion features can be pre-extracted, and the presence of leg assistance movements within the angle change trajectory can be determined by comparing the matching degree of deviation features with the leg assistance motion features. For detecting incomplete movements, it can be verified whether the movement amplitude meets the standard requirements, detecting half-range movement patterns, and analyzing movement integrity indicators for identification. For example, half-range movement motion features can be pre-extracted, and the presence of incomplete movements within the angle change trajectory can be determined by comparing the matching degree of deviation features with the half-range movement motion features. For detecting inertial-assisted movements, it's possible to identify swinging patterns utilizing inertia, analyze abnormal characteristics in the relationship between speed and angle, and detect momentum abuse. For example, by pre-extracting the motion features of inertial swinging, and comparing the matching degree between deviation features and the motion features of inertial swinging, it can be determined whether inertial-assisted movements exist within the trajectory of angle changes. For detecting abnormal device wearing movements, it's possible to monitor changes in the wearing position of smart wearable devices. For example, by pre-extracting the motion features of smart wearable devices worn on the legs, and by comparing the matching degree between deviation features and the motion features of abnormal device wearing movements, it can be determined whether abnormal device wearing movements exist within the trajectory of angle changes.

[0098] In the above embodiments, based on the matching degree between the deviation features in the motion state features and the motion features of abnormal actions, it is determined whether there are abnormal actions within the trajectory of angle change. By analyzing the motion biomechanical features to identify abnormal action patterns, different types of abnormal actions can be comprehensively identified, improving the accuracy of abnormal action detection results and thus improving the accuracy of the number of target actions. By analyzing the motion biomechanical features to identify violation action patterns, when abnormal actions are detected, real-time feedback can also be provided to guide users to correct their actions immediately.

[0099] In one embodiment, the step of determining the number of target actions in S503 is further refined, including:

[0100] For any angle change trajectory, if the abnormal action detection result within the angle change trajectory indicates that there is no abnormal action, the angle change trajectory is determined as the target change trajectory; based on the number of determined target change trajectories, the number of target actions completed by the target object within the sampling period is determined.

[0101] If the abnormal action detection result within the angle change trajectory indicates that there is no abnormal action, then it can be considered that the target object has completed a target action within the angle change trajectory. The angle change trajectory is then taken as the target change trajectory, and the number of target change trajectories is taken as the number of target actions.

[0102] In the above embodiments, if there is no abnormal action within the angle change trajectory, the target object is recorded as having completed a target action, thus avoiding the inclusion of abnormal actions in the target action count. This can improve the accuracy of the target action count, and the abnormal behavior detection rate can reach over 90%.

[0103] In one embodiment, while counting the number of target actions completed by the target object during the sampling period, a standardized score can also be applied to the actions completed by the target object, such as... Figure 6 As shown, the standardization of the target object's actions is scored using the following methods:

[0104] S601 determines the range of motion score based on the minimum and maximum torso tilt angles contained in each angular change trajectory.

[0105] The motion amplitude score reflects the completion and effectiveness of the target object's action. For each angle change trajectory, the minimum and maximum torso tilt angles are extracted, reflecting the motion amplitude of the target object within the trajectory. Optionally, for each angle change trajectory, the amplitude score is determined based on the differences between the minimum and first reference tilt angles, and between the maximum and second reference tilt angles. The motion amplitude score is then determined based on the amplitude scores for each trajectory. The first and second reference tilt angles can be set based on experience, multiple trials, and actual needs, and are not specifically limited here. It is understood that the smaller the difference between the minimum and first reference tilt angles, and the smaller the difference between the maximum and second reference tilt angles, the more standard the target object's action is, and the higher the amplitude score for the angle change trajectory. Consequently, the higher the amplitude score for each angle change trajectory, the higher the motion amplitude score.

[0106] S602 determines the motion stability score based on the similarity between trajectories changing at different angles.

[0107] Among them, the motion stability score is used to reflect the motion control ability and quality of the target object. Optionally, for different angle change trajectories, the similarity between each pair of angle change trajectories can be determined, and the mean of the determined similarity can be used as the motion stability score. Alternatively, different angle change trajectories can be randomly divided into two groups, the average change trajectory of each group can be determined, and the similarity between the average change trajectories of the two groups can be used as the motion stability score.

[0108] S603 determines the motion speed score based on the target time consumed by the trajectory of each angle change.

[0109] The motion speed score reflects the rhythm and efficiency of the target object's movements. For example, the motion speed score can be determined based on the similarity between the target durations taken for different angle change trajectories. Alternatively, for each angle change trajectory, an individual speed score can be determined based on the first, second, third, and fourth durations taken for that trajectory, and the average of these individual speed scores can be used to determine the overall motion speed score. The specific method for determining the motion speed score can be set according to actual needs and is not specifically limited here.

[0110] S604 determines the overall motion score of the target object based on the motion amplitude score, motion stability score, and motion speed score.

[0111] The comprehensive exercise score is used to characterize the target subject's overall performance in performing a target action, considering factors such as range of motion, stability, and speed. Optionally, different weights can be assigned to the range of motion, stability, and speed scores, and a weighted sum can be calculated based on these weights to determine the target subject's comprehensive exercise score. For example, if the weights for range of motion, stability, and speed are 0.4, 0.3, and 0.4 respectively, and the scores are 80, 90, and 80 points respectively, then the comprehensive exercise score would be 32 + 27 + 32 = 91 points. Optionally, in a physical fitness test scenario, after obtaining the target subject's comprehensive exercise score, it can be converted into a standard score based on physical fitness standards and personalized factors such as the target subject's age and gender. This standard score can then be used to evaluate the target subject's physical fitness. By incorporating personalized factors into the standard score conversion, the system can adapt to the different exercise characteristics of various users and maintain high recognition accuracy.

[0112] In the above embodiments, by scoring the target object in multiple dimensions during the process of the target object completing the target action, the movement of the target object can be comprehensively reflected, helping the target object to improve and enhance.

[0113] Based on the above embodiments, the motion count method in this application can be applied to count sit-ups in both real and simulated physical fitness test scenarios. Specifically, in both scenarios, the sampling period is set to one minute, and parameters such as the starting angle and peak angle can be set according to physical fitness test standards. This allows users to conduct accurate self-assessment and targeted training before the actual physical fitness test, verifying the correctness of the starting posture and providing an experience environment highly consistent with the real test environment. Furthermore, a predictive model can be built based on machine learning technology. Input features are constructed based on the user's historical training data, current performance trends, physiological parameters, and environmental factors. These features are then input into the predictive model, which outputs the expected physical fitness test score, the probability of passing, and an analysis of weaknesses, providing accurate training guidance for the user. The accuracy rate of physical fitness test score prediction can reach over 85%. Optionally, an efficient data processing pipeline can be established, starting from the acquisition of raw sensor data, through cache queue management, digital filtering, feature extraction calculation, and finally motion recognition, with the processing results updated in real time on the display interface. In simulated physical fitness training scenarios, the system can monitor movement quality in real time during training, provide timely correction suggestions, and generate targeted training plans based on users' historical data and goal settings to help users scientifically improve training results. It can also record users' long-term training data, visualize the progress trajectory, and provide users with the motivation to continue training.

[0114] Based on the above embodiments, the motion counting method in this application can also be applied to the counting of sit-ups in a multi-player competitive system, supporting multiple participants to compete in sit-ups in real time. Specifically, a distributed architecture is adopted, with each participant's smart wearable device independently counting motions locally and uploading the target number of motions completed by the participant to the cloud competitive server in real time via an encrypted transmission protocol. The server aggregates the data and calculates real-time rankings, feeding back the results to all participants. During the motion counting process of each smart wearable device, the influence of hardware differences can be eliminated through standardized processing of sensor data between devices, and a personalized motion mode adaptive learning algorithm can be adopted to adapt to the movement characteristics of different participants. An environmental factor compensation mechanism is introduced to ensure the fairness of the competition. To further improve the fairness of the competition, each smart wearable device can also identify abnormal motions during the competition through abnormal motion detection results, identify abnormal fluctuations in performance through historical performance comparison verification, establish a reputation scoring system, maintain a fair competitive environment, and achieve accurate identification, standardized evaluation, and interactive competition of sit-up motions. The aforementioned real-time multiplayer competitive activities, through optimized data transmission and ranking algorithms, provide a smooth competitive experience and establish a social sharing, team challenge, and achievement reward mechanism based on real sports data. This can significantly improve user participation and training persistence, and can also provide personalized encouragement and guidance based on user performance, thereby enhancing user stickiness and training motivation.

[0115] Therefore, the motion count method in this application has significant effects in many aspects, such as motion recognition accuracy, anti-cheating ability, standardized evaluation, social interaction and intelligent guidance, providing a more comprehensive, accurate and interesting sit-up training experience.

[0116] Based on the above embodiments, in an exemplary embodiment, such as Figure 7 As shown, the action count method in this embodiment can be applied to smart wearable devices, and includes the following steps:

[0117] S701: Obtain motion state data of the target object at different sampling times within the sampling period.

[0118] Among them, smart wearable devices are worn on the target parts of the target object.

[0119] S702: For any sampling time, process the motion state data at the sampling time to obtain the attitude angle of the target part at the sampling time; according to the preset conversion coefficient between the attitude angle and the torso tilt angle, convert the attitude angle of the target part to obtain the torso tilt angle of the target object at the sampling time.

[0120] S703: Determine the tilt angle change information based on the torso tilt angle of the target object at each sampling time.

[0121] S704: Based on the tilt angle change information, determine the trajectory of the trunk tilt angle change during the sampling period.

[0122] The trajectory of the angle change is as follows: the trajectory of the torso tilt angle changing from the initial angle to the peak angle, and then returning from the peak angle to the initial angle.

[0123] S705: For any angle change trajectory, obtain the target duration of the angle change trajectory. If the target duration meets the target duration condition, obtain the first duration of the torso tilt angle changing from the starting angle to the rising angle threshold, the second duration of the torso tilt angle changing from the rising angle threshold to the peak angle, the third duration of the torso tilt angle changing from the peak angle to the falling angle threshold, and the fourth duration of the torso tilt angle changing from the falling angle threshold to the starting angle within the angle change trajectory.

[0124] S706: For any angle change trajectory, determine the motion state characteristics within the angle change trajectory based on the motion state data at each sampling time within the angle change trajectory; compare the motion state characteristics with the standard state characteristics to determine the abnormal action detection results within the angle change trajectory.

[0125] S707: For any angle change trajectory, if the first duration corresponding to the angle change trajectory belongs to the first duration range, the second duration belongs to the second duration range, the third duration belongs to the third duration range, and the fourth duration belongs to the fourth duration range, the angle change trajectory is determined as the target change trajectory. Furthermore, if the abnormal action detection result within the angle change trajectory indicates that there is no abnormal action, the angle change trajectory is determined as the target change trajectory.

[0126] S708: Based on the number of target change trajectories determined, determine the number of target actions completed by the target object within the sampling period.

[0127] The specific implementation methods of S701-S708 are the same as those in the above method embodiments, and will not be repeated here.

[0128] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0129] Based on the same inventive concept, this application also provides a motion quantity counting device for implementing the motion quantity counting method described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more motion quantity counting device embodiments provided below can be found in the limitations of the motion quantity counting method above, and will not be repeated here.

[0130] In one exemplary embodiment, such as Figure 8 As shown, an action quantity counting device is provided, including: an acquisition module 801, a processing module 802, an information determination module 803, and a quantity determination module 804, wherein:

[0131] The acquisition module 801 is used to acquire motion state data of the target object at different sampling times within the sampling period;

[0132] The processing module 802 is used to determine the torso tilt angle of the target object at any sampling time based on the motion state data at the sampling time.

[0133] The information determination module 803 is used to determine the tilt angle change information based on the torso tilt angle of the target object at each sampling time.

[0134] The quantity determination module 804 is used to determine the number of target actions completed by the target object during the sampling period based on the tilt angle change information.

[0135] In one embodiment, the quantity determination module 804 includes:

[0136] The first determining unit is used to determine the angle change trajectory of the torso tilt angle during the sampling period based on the tilt angle change information; wherein, the angle change trajectory is: the trajectory of the torso tilt angle changing from the initial angle to the peak angle and then returning from the peak angle to the initial angle;

[0137] The second determining unit is used to determine the number of target actions of the target object during the sampling period based on the angle change trajectory corresponding to the torso tilt angle during the sampling period.

[0138] In one embodiment, the second determining unit is specifically used to: for any angle change trajectory, obtain the target duration of the angle change trajectory; determine the angle change trajectory corresponding to the target duration that meets the target duration condition as the target change trajectory; and determine the number of target actions completed by the target object within the sampling period based on the number of target change trajectories.

[0139] In one embodiment, the second determining unit is specifically used to: for any angle change trajectory, determine the motion state characteristics within the angle change trajectory based on the motion state data at each sampling time within the angle change trajectory; compare the motion state characteristics with the standard state characteristics to determine the abnormal action detection results within the angle change trajectory; and determine the number of target actions completed by the target object within the sampling period based on the abnormal action detection results within each angle change trajectory.

[0140] In one embodiment, the second determining unit is specifically used to: for any angle change trajectory, if the abnormal action detection result within the angle change trajectory indicates that there is no abnormal action, determine the angle change trajectory as the target change trajectory; and determine the number of target actions completed by the target object within the sampling period based on the number of determined target change trajectories.

[0141] In one embodiment, the second determining unit is specifically used to: for any angle change trajectory, if the target duration of the angle change trajectory meets the target duration condition, obtain the rising duration and falling duration of the angle change trajectory; wherein, the rising duration is: the time taken for the torso tilt angle to change from the initial angle to the peak angle within the angle change trajectory, and the falling duration is: the time taken for the torso tilt angle to recover from the peak angle to the initial angle within the angle change trajectory; if the rising duration is within the rising duration range and the falling duration is within the falling duration range, determine the angle change trajectory as the target change trajectory.

[0142] In one embodiment, the rising duration includes a first duration for the trunk tilt angle to change from the initial angle to the rising angle threshold, and a second duration for the trunk tilt angle to change from the rising angle threshold to the peak angle; the falling duration includes a third duration for the trunk tilt angle to change from the peak angle to the falling angle threshold, and a fourth duration for the trunk tilt angle to change from the falling angle threshold to the initial angle; the second determining unit is specifically used to: if the first duration belongs to the first duration range, the second duration belongs to the second duration range, the third duration belongs to the third duration range, and the fourth duration belongs to the fourth duration range, determine the angle change trajectory as the target change trajectory.

[0143] In one embodiment, the smart wearable device is worn on the target part of the target object; the processing module 802 is specifically used to: process the motion state data at the sampling time to obtain the posture angle of the target part at the sampling time; and convert the posture angle of the target part according to the preset conversion coefficient between the posture angle and the torso tilt angle to obtain the torso tilt angle of the target object at the sampling time.

[0144] In one embodiment, the quantity determination module 804 includes a scoring unit for: determining a motion amplitude score based on the minimum and maximum torso tilt angles contained in each angle change trajectory; determining a motion stability score based on the similarity between different angle change trajectories; determining a motion speed score based on the target duration consumed by each angle change trajectory; and determining a comprehensive motion score for the target object based on the motion amplitude score, motion stability score, and motion speed score.

[0145] In one embodiment, the second determining unit is specifically used to: compare motion state features and standard state features to obtain deviation features that are different from standard state features in motion state features; if the degree of matching between the deviation features and the motion features of any abnormal action is higher than the degree of matching threshold, determine that the abnormal action detection result within the angle change trajectory is that there is an abnormal action.

[0146] Each module in the aforementioned action quantity counting device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0147] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a method for counting the number of actions. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0148] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0149] In one exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to: acquire motion state data of a target object at different sampling times within a sampling period; for any sampling time, determine the torso tilt angle of the target object at the sampling time based on the motion state data at the sampling time; determine tilt angle change information based on the torso tilt angle of the target object at each sampling time; and determine the number of target actions completed by the target object within the sampling period based on the tilt angle change information.

[0150] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon. When executed by a processor, the computer program performs the following: acquiring motion state data of a target object at different sampling times within a sampling period; determining the torso tilt angle of the target object at any sampling time based on the motion state data at that sampling time; determining tilt angle change information based on the torso tilt angle of the target object at each sampling time; and determining the number of target actions completed by the target object within the sampling period based on the tilt angle change information.

[0151] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following: acquiring motion state data of a target object at different sampling times within a sampling period; determining, for any sampling time, the torso tilt angle of the target object at the sampling time based on the motion state data at that sampling time; determining tilt angle change information based on the torso tilt angle of the target object at each sampling time; and determining the number of target actions completed by the target object within the sampling period based on the tilt angle change information.

[0152] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0153] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0154] The technical features of the above embodiments can be combined in any way. For the sake of brevity, 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 application.

[0155] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method of counting the number of actions, characterized by, The method is applied to a smart wearable device, and comprises: obtaining motion state data of a target object at different sampling time points in a sampling period; for any sampling time point, determining a trunk inclination angle of the target object at the sampling time point according to the motion state data of the sampling time point; determining inclination angle change information according to the trunk inclination angles of the target object at the sampling time points; determining a target action quantity completed by the target object in the sampling period according to the inclination angle change information.

2. The method of claim 1, wherein, The determination of the target action quantity completed by the target object in the sampling period according to the inclination angle change information comprises: determining an angle change trajectory of the trunk inclination angle in the sampling period according to the inclination angle change information; wherein the angle change trajectory is a change trajectory in which the trunk inclination angle changes from a starting angle to a peak angle and then returns to the starting angle; determining the target action quantity of the target object in the sampling period according to the angle change trajectory of the trunk inclination angle in the sampling period.

3. The method of claim 2, wherein, The determination of the target action quantity completed by the target object in the sampling period according to the angle change trajectory of the trunk inclination angle in the sampling period comprises: for any angle change trajectory, obtaining a target time length consumed by the angle change trajectory; determining an angle change trajectory corresponding to a target time length that meets a target time length condition as a target change trajectory; determining the target action quantity completed by the target object in the sampling period according to the number of target change trajectories.

4. The method of claim 2, wherein, The determination of the target action quantity completed by the target object in the sampling period according to the angle change trajectory of the trunk inclination angle in the sampling period comprises: for any angle change trajectory, determining a motion state feature in the angle change trajectory according to the motion state data at the sampling time points in the angle change trajectory; comparing the motion state feature with a standard state feature to determine an abnormal action detection result in the angle change trajectory; determining the target action quantity completed by the target object in the sampling period according to the abnormal action detection results in the angle change trajectories.

5. The method of claim 4, wherein, The determination of the target action quantity completed by the target object in the sampling period according to the abnormal action detection results in the angle change trajectories comprises: for any angle change trajectory, if the abnormal action detection result in the angle change trajectory indicates that there is no abnormal action, determining the angle change trajectory as a target change trajectory; determining the target action quantity completed by the target object in the sampling period according to the number of determined target change trajectories.

6. The method of claim 3, wherein, The determination of the angle change trajectory corresponding to the target time length that meets the target time length condition as the target change trajectory comprises: For any angle change trajectory, if a target time length of the angle change trajectory satisfies a target time length condition, an ascending time length and a descending time length consumed by the angle change trajectory are obtained; the ascending time length is a time length consumed by the trunk inclination angle changing from the starting angle to the peak angle in the angle change trajectory, and the descending time length is a time length consumed by the trunk inclination angle changing from the peak angle to the starting angle in the angle change trajectory; If the ascending time length belongs to an ascending time length range and the descending time length belongs to a descending time length range, the angle change trajectory is determined as a target change trajectory.

7. The method of claim 6, wherein, The ascending time length includes a first time length consumed by the trunk inclination angle changing from the starting angle to an ascending angle threshold and a second time length consumed by the trunk inclination angle changing from the ascending angle threshold to the peak angle, and the descending time length includes a third time length consumed by the trunk inclination angle changing from the peak angle to a descending angle threshold and a fourth time length consumed by the trunk inclination angle changing from the descending angle threshold to the starting angle; The step of determining the angle change trajectory as the target change trajectory if the ascending time length belongs to the ascending time length range and the descending time length belongs to the descending time length range includes: The angle change trajectory is determined as the target change trajectory if the first time length belongs to a first time length range, the second time length belongs to a second time length range, the third time length belongs to a third time length range, and the fourth time length belongs to a fourth time length range.

8. The method according to any one of claims 1 to 7, characterized in that, The intelligent wearable device is worn on a target part of the target object; The step of determining the trunk inclination angle corresponding to the target object at the sampling time according to the motion state data at the sampling time includes: Processing the motion state data at the sampling time to obtain a posture angle of the target part at the sampling time; Converting the posture angle of the target part according to a preset conversion coefficient between the posture angle and the trunk inclination angle to obtain the trunk inclination angle corresponding to the target object at the sampling time.

9. An action number counting apparatus characterized by comprising: The device is deployed on an intelligent wearable device, and the device includes: An acquisition module configured to acquire motion state data of a target object at different sampling times in a sampling period; A processing module configured to, for any sampling time, determine a trunk inclination angle corresponding to the target object at the sampling time according to the motion state data at the sampling time; An information determination module configured to determine inclination angle change information according to the trunk inclination angles corresponding to the target object at the sampling times; A quantity determination module configured to determine a target action quantity completed by the target object in the sampling period according to the inclination angle change information. 10.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor implements the steps of the method in any one of claims 1 to 8 when executing the computer program.

Images