A method and system for correcting counting errors and dynamically classifying exercise intensity in intelligent jump ropes.

Abstract

This application relates to the field of intelligent sports equipment technology, and provides a method and system for correcting counting errors and dynamically classifying exercise intensity in intelligent jump rope. The method involves acquiring three-dimensional motion data during the jump rope process; extracting features from the three-dimensional motion data to obtain jump rope motion features; identifying target data segments in the three-dimensional motion data using these features; acquiring the time interval and displacement trajectory repeatability of two adjacent movements within the target data segment; determining invalid counts and removing segments when the time interval exceeds the normal jump rope frequency range or the displacement trajectory repeatability is below a preset threshold; correcting abnormal jump rope counting results; and calculating the current exercise intensity index for dynamic classification by combining basic information and real-time collected jump rope frequency, continuous exercise time, cumulative jump rope counts, and heart rate data. This application provides users with a more accurate and personalized exercise experience.

Description

Technical Field

[0001] This application relates to the field of intelligent sports equipment technology, and in particular to a method and system for correcting counting errors and dynamically classifying exercise intensity in an intelligent jump rope. Background Technology

[0002] With the increasing awareness of fitness among the general public, smart jump ropes, as exercise equipment that combines portability and data-driven functions, are widely used. The core functions of existing smart jump ropes focus on counting jumps and basic exercise data statistics, but they have significant shortcomings in terms of counting accuracy and the precision of exercise intensity assessment.

[0003] The counting error problem is prominent: Traditional smart jump ropes mostly rely on a single accelerometer or angular velocity sensor to collect data, identifying jump rope movements solely through the periodic characteristics of the handle's motion trajectory. However, during actual jump rope use, users may generate interference data due to fatigue, distorted movements, or non-jumping motions (such as adjusting rope length or natural arm swings), leading to miscounting or missed counts. For example, existing solutions do not incorporate the correlation between jump rope movements and the user's landing (such as foot contact signals), making it difficult to effectively distinguish between abnormal movements such as "empty swing without jumping" and "jump without swinging the rope," resulting in a high counting error rate.

[0004] Exercise intensity assessment is often crude and lacks personalized dynamic grading: Current technologies for judging exercise intensity are mostly based on a single indicator (such as heart rate or rope skipping frequency), and the grading standards are fixed, failing to dynamically adjust to individual user differences (age, weight, fitness level) and multiple exercise goals (such as fat loss, endurance training, and explosive power training). For example, for users aiming to lose fat, current solutions simply divide heart rate zones, ignoring the impact of rope skipping frequency stability on energy expenditure; for explosive power training, there are no quantitative standards for the magnitude and duration of frequency spikes, resulting in users not receiving accurate exercise guidance and making it difficult to guarantee training effectiveness.

[0005] Insufficient multi-source data fusion: Existing smart jump ropes have limited sensor configurations (usually only an accelerometer built into the handle), failing to fully utilize multi-dimensional data such as rope tension sensors and foot pressure sensors for cross-validation. For example, changes in rope tension directly reflect the force of rope swing and the rope's motion state, while foot contact signals accurately characterize the occurrence of a jump. However, existing solutions lack the fusion processing of this data, resulting in poor robustness of the counting logic and strength assessment model.

[0006] Therefore, a method is urgently needed to solve at least one of the above problems. Summary of the Invention

[0007] This application provides a method and system for correcting counting errors and dynamically classifying exercise intensity in smart jump ropes. It aims to address the problem that existing smart jump ropes, whose core functions focus on jump rope counting and basic exercise data statistics, have significant shortcomings in terms of counting accuracy and the refinement of exercise intensity assessment.

[0008] In a first aspect, embodiments of this application provide a method for correcting counting errors and dynamically classifying exercise intensity in a smart jump rope, applied to a smart jump rope; the method includes:

[0009] Acquire three-dimensional motion data during the rope skipping process. The three-dimensional motion data includes at least the spatial displacement information, rotation angle information, and motion acceleration information corresponding to the smart rope skipping.

[0010] Feature extraction is performed on the three-dimensional motion data to obtain rope skipping motion features; the target data segment corresponding to the valid rope skipping action is identified in the three-dimensional motion data using the rope skipping motion features.

[0011] The time interval and displacement trajectory repeatability of two adjacent actions are obtained from the target data segment. When the time interval exceeds the normal rope skipping frequency range or the displacement trajectory repeatability is lower than the preset threshold, it is determined to be an invalid count and discarded.

[0012] The system acquires basic user information and combines this information with real-time data on jump rope frequency, duration of exercise, cumulative number of jumps, and heart rate to calculate the current exercise intensity index. It then dynamically grades the current exercise intensity index according to a pre-defined grading rule base, which includes intensity grading standards corresponding to different exercise goals.

[0013] In some embodiments, acquiring three-dimensional motion data during the rope skipping process includes: real-time data acquisition via a motion sensor assembly installed on the smart rope. The motion sensor assembly includes at least a three-axis accelerometer, a three-axis angular velocity sensor, and a gyroscope installed on the corresponding handle of the smart rope, for acquiring translational acceleration data, rotational angular velocity data, and spatial attitude angle data of the smart rope in three-dimensional space. The motion sensor assembly also includes a tension sensor installed on the corresponding rope body of the smart rope for assisting in acquiring information on rope tension changes during the swinging of the smart rope.

[0014] In some embodiments, the rope skipping motion features include the periodic features, displacement amplitude features, and angular velocity change features of normal rope skipping movements; the feature extraction of the three-dimensional motion data to obtain the rope skipping motion features includes: preprocessing the original three-dimensional motion data, removing high-frequency noise and abnormal abrupt data through a sliding window filtering algorithm to generate a smooth and continuous motion data sequence; performing time-domain feature analysis on the preprocessed sequence to extract the time period of a single rope skipping movement, the spatial amplitude extreme value of the handle displacement, and the peak value of the angular velocity change; performing frequency-domain feature analysis to obtain the main frequency component and secondary frequency component of the rope skipping motion through Fourier transform; and constructing the rope skipping motion features by combining the time-domain and frequency-domain features.

[0015] In some embodiments, identifying the target data segment corresponding to a valid rope skipping action in the three-dimensional motion data through the rope skipping motion features includes: dynamically matching the real-time acquired three-dimensional motion data sequence with a preset normal rope skipping action feature template, wherein the feature template includes a period threshold range, a displacement amplitude threshold range, and an angular velocity change rate threshold range pre-trained for different user body types; when the period feature of any segment of data in the three-dimensional motion data sequence falls within the corresponding threshold range, the displacement amplitude feature reaches the minimum displacement threshold for bottoming out and jumping, and the angular velocity change feature conforms to the direction conversion law of the handle swing, it is determined to be the target data segment corresponding to a valid rope skipping action.

[0016] In some embodiments, obtaining the time interval and displacement trajectory repeatability of two adjacent actions in the target data segment includes: performing temporal segmentation on the continuous target data segment, extracting the start time points of two adjacent valid actions to calculate the time interval, the normal rope skipping frequency range being dynamically adjusted according to the user's age and physical condition, the normal rope skipping frequency including an interval formed by a preset proportion of fluctuation above and below the average frequency based on the user's historical rope skipping data statistics; measuring the displacement trajectory repeatability by calculating the Euclidean distance similarity of the handle spatial displacement trajectory in two adjacent actions, the preset threshold being preset according to the standard action model of rope skipping.

[0017] In some embodiments, before obtaining the user's basic information, the method further includes: cross-validating the jump rope counting results by combining the foot contact signals collected during the user's jump to correct abnormal jump rope counting results. This includes: collecting foot contact signals by a pressure sensor worn on the user's foot or a pressure sensing device integrated into the jump rope pedal, extracting the periodic and intensity characteristics of the contact signals; comparing the period of the contact signals with the period of the jump rope action in time synchronization; and determining abnormal counting and correcting it when the difference between the number of contact signals and the number of jump rope counts within the same time window exceeds a preset threshold, or when the intensity of the contact signals does not reach the minimum pressure threshold for landing.

[0018] In some embodiments, the basic information includes age, weight, height, and preset exercise goals; the calculation of the current exercise intensity index by combining the basic information with real-time collected data on rope skipping frequency, continuous exercise time, cumulative rope skipping count, and heart rate includes: calculating the theoretical maximum heart rate based on the user's age and height, and obtaining the heart rate reserve percentage by combining real-time heart rate data; calculating the frequency change coefficient based on the difference between the rope skipping frequency and the user's historical average frequency, and estimating the real-time energy consumption rate by combining the continuous exercise time and cumulative rope skipping count through a preset energy consumption model; and weighting and fusing the heart rate reserve percentage, frequency change coefficient, and energy consumption rate to generate a dimensionless current exercise intensity index.

[0019] In some embodiments, the exercise goals include at least fat loss, endurance training, and explosive power training. The intensity grading standard for each exercise goal is divided according to heart rate zones, rope skipping frequency change rate, and energy consumption rate. The current exercise intensity index is dynamically graded according to a preset grading rule base, which contains intensity grading standards corresponding to different exercise goals, including: for fat loss goals, the grading standard is divided into three levels: low intensity, medium intensity, and high intensity based on heart rate reserve percentage, where medium intensity corresponds to 60%-75% of maximum heart rate and requires the rope skipping frequency to be stable at 80%-110% of the base frequency; for endurance training goals, the grading standard combines continuous exercise time and frequency change coefficient, requiring high intensity levels to maintain a frequency change coefficient within ±5% and a duration of more than 40 minutes; for explosive power training goals, the grading standard is based on the ratio of peak rope skipping frequency to base frequency, requiring high intensity levels to have a frequency surge exceeding 30% of the base frequency and maintained for at least 10 seconds.

[0020] In some embodiments, after dynamically classifying the current exercise intensity index according to a preset classification rule base, the method further includes: feeding back the real-time exercise intensity level to the user through an associated smart terminal based on the current exercise intensity classification result, and adjusting the intensity suggestions for subsequent exercise according to a preset adaptive strategy, wherein the adaptive strategy includes rules for dynamically optimizing the classification threshold based on the user's historical exercise data and current physical condition.

[0021] Secondly, this application provides a counting error correction and dynamic exercise intensity grading system for smart jump ropes, applied to smart jump ropes, the system comprising:

[0022] The data acquisition unit is used to acquire three-dimensional motion data during the rope skipping process. The three-dimensional motion data includes at least the spatial displacement information, rotation angle information and motion acceleration information corresponding to the smart rope skipping.

[0023] The feature extraction unit is used to extract features from the three-dimensional motion data to obtain rope skipping motion features; and to identify the target data segment corresponding to the valid rope skipping action in the three-dimensional motion data through the rope skipping motion features.

[0024] An anomaly correction unit is used to obtain the time interval and displacement trajectory repeatability of two adjacent actions in the target data segment. When the time interval exceeds the normal rope skipping frequency range or the displacement trajectory repeatability is lower than a preset threshold, it is determined to be an invalid count and discarded.

[0025] The dynamic grading unit is used to acquire the user's basic information and, in conjunction with the basic information and real-time collected data on rope skipping frequency, continuous exercise time, cumulative rope skipping count, and heart rate, calculate the current exercise intensity index. The current exercise intensity index is dynamically graded according to a preset grading rule base, which contains intensity grading standards corresponding to different exercise goals.

[0026] This application provides a method and system for correcting counting errors and dynamically grading exercise intensity in a smart jump rope. It addresses the shortcomings of existing technologies, which lack a multi-source fusion counting correction method combining three-dimensional motion data, rope tension signals, and foot contact signals, and also fail to construct a dynamic intensity grading system based on user information and multiple exercise goals. Specifically, existing solutions have the following technical gaps: They fail to achieve multi-dimensional verification of jump rope movements: relying solely on handle trajectory recognition without cross-validating counting results using foot contact signals, thus failing to exclude abnormal scenarios such as "invalid swings" or "missed jumps"; they lack a dynamic grading rule base: failing to design differentiated intensity grading standards for different exercise goals such as fat loss, endurance, and explosive power, and especially failing to construct a comprehensive evaluation model incorporating multiple indicators such as heart rate zones, rate of change of frequency, and energy consumption rate; and they fail to consider individual user differences: grading thresholds are fixed and not dynamically adjusted based on user age, physical condition, and historical exercise data, resulting in a lack of personalized evaluation results.

[0027] Therefore, there is an urgent need for an intelligent rope skipping data processing method that can improve counting accuracy through multi-source data fusion and achieve dynamic intensity grading based on multiple motion targets, in order to solve the core problems of large counting errors and coarse evaluation in existing technologies.

[0028] Based on the above technical solutions, this invention effectively eliminates invalid movements such as "empty swings" and "missed jumps" by fusing multi-dimensional data from triaxial acceleration, angular velocity, and rope tension sensors, combined with cross-verification of foot contact signals, thus reducing the counting error rate. It constructs a differentiated grading rule base for different exercise goals (fat loss, endurance, explosive power), dynamically adjusting the intensity level based on user basic information and real-time exercise data (heart rate, frequency, energy consumption), making exercise guidance more personalized. Through time-domain-frequency domain feature analysis and dynamic threshold matching, it achieves effective movement recognition in complex exercise scenarios (such as high-frequency rope skipping and movement variations), improving the device's adaptability to different user groups. Based on the dynamic grading results, it provides real-time intensity feedback and optimizes the grading threshold according to the user's physical condition, forming a closed loop of "data collection-evaluation-feedback," significantly improving exercise efficiency and scientific rigor.

[0029] In summary, this invention fills the technological gap in the fields of counting correction and dynamic intensity grading of existing smart jump ropes, providing users with a more accurate and personalized exercise experience.

[0030] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0031] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic flowchart illustrating the steps of a method for correcting counting errors and dynamically classifying exercise intensity in an intelligent jump rope, as provided in an embodiment of this application.

[0033] Figure 2 This is a schematic diagram of the structure of a smart jump rope provided in an embodiment of this application;

[0034] Figure 3 This is a schematic block diagram of a smart jump rope counting error correction and dynamic exercise intensity grading system provided in an embodiment of this application;

[0035] Figure 4 This is a schematic block diagram of the structure of an intelligent jump rope provided in an embodiment of this application.

[0036] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Detailed Implementation

[0037] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0038] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0039] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0040] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0041] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0042] With the increasing awareness of fitness among the general public, smart jump ropes, as exercise equipment that combines portability and data-driven functions, are widely used. The core functions of existing smart jump ropes focus on counting jumps and basic exercise data statistics, but they have significant shortcomings in terms of counting accuracy and the precision of exercise intensity assessment.

[0043] The counting error problem is prominent: Traditional smart jump ropes mostly rely on a single accelerometer or angular velocity sensor to collect data, identifying jump rope movements solely through the periodic characteristics of the handle's motion trajectory. However, during actual jump rope sessions, users may generate interference data due to fatigue, distorted movements, or non-jumping motions (such as adjusting rope length or natural arm swings), leading to miscounting or missed counts. For example, existing solutions do not incorporate the correlation between jump rope movements and the user's landing (such as foot contact signals), making it difficult to effectively distinguish between abnormal movements such as "empty swing without jumping" and "jump without swinging the rope," resulting in a high counting error rate (especially at high-frequency jump rope sessions or when movements are not performed correctly, the error can reach 15%-20%).

[0044] Exercise intensity assessment is often crude and lacks personalized dynamic grading: Current technologies for judging exercise intensity are mostly based on a single indicator (such as heart rate or rope skipping frequency), and the grading standards are fixed, failing to dynamically adjust to individual user differences (age, weight, fitness level) and multiple exercise goals (such as fat loss, endurance training, and explosive power training). For example, for users aiming to lose fat, current solutions simply divide heart rate zones, ignoring the impact of rope skipping frequency stability on energy expenditure; for explosive power training, there are no quantitative standards for the magnitude and duration of frequency spikes, resulting in users not receiving accurate exercise guidance and making it difficult to guarantee training effectiveness.

[0045] Insufficient multi-source data fusion: Existing smart jump ropes have limited sensor configurations (usually only an accelerometer built into the handle), failing to fully utilize multi-dimensional data such as rope tension sensors and foot pressure sensors for cross-validation. For example, changes in rope tension directly reflect the force of rope swing and the rope's motion state, while foot contact signals accurately characterize the occurrence of a jump. However, existing solutions lack the fusion processing of this data, resulting in poor robustness of the counting logic and strength assessment model.

[0046] To address the aforementioned issues, existing technologies have not yet proposed a multi-source fusion counting correction method that combines three-dimensional motion data, rope tension signals, and foot contact signals, nor have they constructed a dynamic intensity grading system based on user information and multiple motion targets. Specifically, existing solutions have the following technical gaps: They fail to achieve multi-dimensional verification of jump rope movements: relying solely on handle motion trajectory to identify movements, they do not cross-validate the counting results using foot contact signals, and cannot eliminate abnormal scenarios such as "invalid swings" or "missed jumps."

[0047] Lack of dynamic grading rule base: There are no intensity grading standards designed for different sports goals such as fat loss, endurance, and explosive power. In particular, there is no comprehensive evaluation model that combines multiple indicators such as heart rate zone, rate of change of frequency, and energy consumption rate. Failure to consider individual differences among users: The grading thresholds are fixed and are not dynamically adjusted according to the user's age, physical condition, and historical exercise data, resulting in a lack of personalization in the evaluation results.

[0048] Therefore, there is an urgent need for an intelligent rope skipping data processing method that can improve counting accuracy through multi-source data fusion and achieve dynamic intensity grading based on multiple motion targets, in order to solve the core problems of large counting errors and coarse evaluation in existing technologies.

[0049] To resolve the above issues, please refer to... Figure 1 This application provides a method for correcting counting errors and dynamically classifying exercise intensity in a smart jump rope, applicable to, for example... Figure 2 The smart jump rope shown is an example. It should also be noted that all information involved in the method provided in this application was extracted with the authorization of the relevant user and in accordance with relevant regulations, and will not infringe on user privacy.

[0050] The provided method for correcting counting errors and dynamically classifying exercise intensity using a smart jump rope includes steps S101 to S104. Details are as follows:

[0051] Step S101. Obtain three-dimensional motion data during the rope skipping process. The three-dimensional motion data includes at least the spatial displacement information, rotation angle information, and motion acceleration information corresponding to the smart rope skipping.

[0052] Specifically, the smart jump rope uses built-in multi-sensor components to collect multi-dimensional physical signals during jump rope movement in real time, constructing a comprehensive dataset that includes spatial displacement, rotation angle, acceleration, and rope tension, providing the raw data foundation for subsequent motion recognition and counting correction.

[0053] Sensor hardware configuration: Handle sensors: The two handles of the smart jump rope integrate a three-axis accelerometer (collecting translational acceleration along the X / Y / Z axes) and a three-axis angular velocity sensor (gyroscope) (collecting rotational angular velocity around the X / Y / Z axes), respectively, to acquire real-time motion acceleration data of the handles in three-dimensional space (unit: m / s²). 2 ) and rotational angular velocity data (unit: rad / s). Rope sensor: A tension sensor is installed in the middle section or connection point of the jump rope to monitor the change in rope tension (unit: N) during swinging, reflecting the swinging force and rope tension. Auxiliary sensor: A pressure sensor is worn on the user's feet or a jump rope pedal with an integrated pressure sensing device is used to collect the foot contact signal when landing (used in conjunction with step S103 for cross-validation).

[0054] The data acquisition frequency is at a sampling frequency of no less than 100Hz to synchronously acquire data from each sensor, ensuring that motion details (such as high-frequency swinging and sudden acceleration changes) are fully captured.

[0055] By combining handle motion data (displacement, angle, acceleration) with rope tension signals, a multi-dimensional feature vector of jump rope movements is constructed. This avoids the data bias caused by motion deformation or environmental interference from a single sensor (such as an accelerometer alone), laying a data foundation for accurate identification of effective movements. High-frequency acquisition and multi-dimensional signal coverage can effectively capture the attenuation of user movements under fatigue (such as decreased tension and angular velocity fluctuations) or subtle trajectory changes during high-frequency jump rope, improving the system's robustness to complex motion scenarios.

[0056] Step S102. Extract features from the three-dimensional motion data to obtain rope skipping motion features; identify the target data segment corresponding to the valid rope skipping action in the three-dimensional motion data through the rope skipping motion features.

[0057] Specifically, by denoising and extracting features from the original three-dimensional motion data, a rope skipping motion feature model containing time-domain and frequency-domain features is constructed. By dynamically matching preset action feature templates, the target data segment corresponding to the effective rope skipping action is segmented from the continuous data.

[0058] Data preprocessing employs sliding window filtering algorithms (such as mean filtering and Kalman filtering) to reduce noise in the original data, removing high-frequency noise (such as hand tremor interference) and abnormal abrupt changes (such as impact noise when the rope accidentally hits the ground during a jump), generating a smooth motion data sequence (including acceleration, angular velocity, and tension signals).

[0059] Feature extraction includes: Time domain features: calculating the time period of a single rope skipping motion (the time interval between two adjacent handle bottoming swings), the spatial amplitude extremes of handle displacement (maximum displacement difference along the X / Y / Z axes), the peak value of angular velocity change (the abrupt change in angular velocity at the start / end of the swing), and the peak value of rope tension (the maximum tension during rope swing). Frequency domain features: performing spectral analysis on the preprocessed signal using Fourier transform to extract the dominant frequency component (corresponding to normal rope skipping frequencies, such as 1-3Hz) and secondary frequency components (such as harmonic frequencies of arm swings) of the rope skipping motion, used to distinguish between regular rope skipping motions and random interference motions.

[0060] Target data segment recognition: Pre-set dynamic feature templates: Based on the user's body type (height, arm span) and historical motion data, pre-train different user's periodic threshold range (e.g., children's jump rope period 0.4-0.6s, adults 0.2-0.4s), displacement amplitude threshold (e.g., displacement difference ≥20cm corresponding to the lowest swing height of the handle), and angular velocity change rate threshold (angular velocity change rate ≥50rad / s² during direction change). When the periodic feature of a continuous data segment falls within the corresponding threshold range, the displacement amplitude reaches the minimum displacement threshold for bottoming out and jumping (avoiding ineffective swings of the arm with small swings), and the angular velocity change feature conforms to the direction change pattern of the handle swing (e.g., alternating clockwise and counterclockwise), it is determined to be the target data segment corresponding to a valid action.

[0061] By combining time-frequency domain features with dynamic threshold matching, this method effectively distinguishes between "valid jump rope movements" and invalid movements such as "adjusting rope length" and "natural arm swinging," avoiding missegmentation caused by traditional single-cycle detection (e.g., misclassifying high-frequency, small-amplitude swings as valid movements). Based on user body shape and historical data, the feature template is dynamically adjusted to address the individual difference adaptation problem caused by the "one-size-fits-all" threshold in existing solutions (e.g., misjudgment when there are large differences in movement amplitude between children and adults), thus improving the recognition accuracy for different user groups.

[0062] Step S103. Obtain the time interval and displacement trajectory repeatability of two adjacent actions in the target data segment. When the time interval exceeds the normal rope skipping frequency range or the displacement trajectory repeatability is lower than the preset threshold, it is determined to be an invalid count and discarded.

[0063] Specifically, in the identified target data segment, valid counts are filtered by action time interval and displacement trajectory repeatability, and double verification is performed by combining foot contact signal to eliminate false counts caused by abnormal actions such as "empty swing without jumping" and "jump without swinging rope".

[0064] The temporal logic filtering includes: Time interval detection: Continuous target data segments are divided according to the start time of the action, and the time interval Δt between two adjacent actions is calculated. Dynamic adjustment of normal jump rope frequency range: Base frequency: Determined based on the user's age (e.g., high frequency for teenagers, low frequency for middle-aged and elderly) and physical condition (historical average frequency), allowing fluctuations of 10%-20% to form a dynamic range (e.g., historical average frequency 180 times / minute, allowable range 162-198 times / minute). If Δt exceeds this range, it is judged as an abnormal action (e.g., continuous rapid swinging without jumping, or missed jumps with excessively long intervals), and the corresponding count is removed. Displacement trajectory repeatability is determined by calculating the Euclidean distance similarity (average distance of the trajectory coordinate point set) of the handle spatial displacement trajectory in two adjacent actions. When a preset threshold is reached (e.g., similarity < 70%), it is judged as a movement deformation (e.g., unilateral arm swinging causing trajectory abnormality), and the count is removed.

[0065] Cross-verification of foot contact signals involves collecting contact signals using foot pressure sensors or jump rope pedal sensors, and extracting the signal period (jump landing frequency) and intensity (pressure value ≥30N to avoid misjudgment of slight contact).

[0066] Time synchronization comparison is performed by checking if the difference between the number of ground contact signals and the number of rope jumps is greater than 1 within the same time window (e.g., 1 second), or if there is no ground contact signal and the pressure value is less than the minimum threshold at a certain count time, then it is judged as an abnormal count (e.g., "rope swinging without jumping" or "jumping without swinging rope"), and the count result is corrected (deducted or added).

[0067] By employing triple verification of time intervals, trajectory repeatability, and ground contact signals, this solution addresses the issues of "counting empty swings" and "missed jumps" caused by relying solely on handle data in traditional methods. The correlation verification between ground contact signals and jump rope movements ensures that the counting results correspond only to the complete "rope swing + jump" motion, avoiding interference from invalid movements and improving data reliability (e.g., swings when the user adjusts the rope length are not mistakenly counted).

[0068] Step S104. Obtain the user's basic information, and combine the basic information with real-time collected data on rope skipping frequency, continuous exercise time, cumulative rope skipping count, and heart rate to calculate the current exercise intensity index; dynamically classify the current exercise intensity index according to a preset classification rule base, wherein the classification rule base contains intensity classification standards corresponding to different exercise goals.

[0069] Specifically, by combining basic user information and real-time exercise data, a multi-indicator evaluation model is constructed, which includes heart rate, rope skipping frequency, and energy consumption. Based on different exercise goals such as fat loss, endurance, and explosive power, the current exercise intensity level is dynamically classified and personalized guidance is provided.

[0070] Exercise intensity metrics include: Heart Rate Reserve (HRR%): Calculated based on the user's age (theoretical value = 220 - age). The percentage of the difference between real-time heart rate and resting heart rate relative to the difference between maximum and resting heart rate reflects cardiopulmonary load intensity. Frequency Variation Coefficient (Δf%): The percentage difference between the current jump rope frequency and the user's historical average frequency, reflecting fluctuations in exercise intensity (e.g., Δf% > +30% indicates a high-frequency burst). Energy Consumption Rate (kcal / min): Estimates real-time energy consumption using a preset model (combining weight, height, duration, and cumulative repetitions). Formula: Energy Consumption = 0.0175 × weight (kg) × jump rope frequency (times / minute) × duration (min). Weighted fusion generates a dimensionless intensity metric (0-100 points) by normalizing HRR%, Δf%, and energy consumption rate according to preset weights (e.g., 0.5:0.3:0.2).

[0071] The dynamic grading rule base includes: Fat loss goals: Low intensity: HRR% < 60%, frequency stable at 80%-110% of base frequency (focusing on sustained low-load fat burning); Medium intensity (optimal range): HRR% 60%-75%, frequency stable at 80%-110% of base frequency (maximizing fat burning efficiency); High intensity: HRR% > 75%, frequency fluctuation allowed ±15% (suitable for advanced fat loss). Endurance training: High intensity level requirements: frequency variation coefficient ≤ ±5% (high movement stability), duration > 40 minutes (challenging cardiovascular endurance). Explosive power training: High intensity level requirements: frequency surge > 30% of base frequency, and duration ≥ 10 seconds (stimulating rapid muscle contraction).

[0072] Real-time feedback and adaptive adjustment are provided by displaying the current intensity level in real time through smart terminals (APP / watch) and dynamically optimizing the grading threshold based on the user's historical exercise data (such as the duration requirement for improving endurance training after long-term training).

[0073] By combining individual differences such as age and weight with diverse exercise goals, this system breaks away from the traditional, crude grading based on a single indicator (such as heart rate alone). It provides guidance for fat-burning users to achieve optimal fat-burning zones, endurance users to improve movement stability, and explosive power users to clearly define sprint intensity and duration. The system dynamically adjusts grading thresholds to adapt to varying fitness levels (such as intensity increases from beginner to advanced users), avoiding "overtraining" or "undertraining" caused by fixed standards, thus improving the scientific nature and efficiency of exercise.

[0074] Steps S101-S104, through a closed-loop process of "multi-source data acquisition → precise feature extraction → multi-dimensional counting verification → personalized intensity grading," systematically solve the core problems of large counting errors and coarse intensity assessment in existing smart jump ropes, achieving a functional upgrade from "data acquisition tool" to "smart sports coach," significantly improving the user's exercise experience and training effect.

[0075] In some embodiments, acquiring three-dimensional motion data during the rope skipping process includes: real-time data acquisition via a motion sensor assembly installed on the smart rope. The motion sensor assembly includes at least a three-axis accelerometer, a three-axis angular velocity sensor, and a gyroscope installed on the corresponding handle of the smart rope, for acquiring translational acceleration data, rotational angular velocity data, and spatial attitude angle data of the smart rope in three-dimensional space. The motion sensor assembly also includes a tension sensor installed on the corresponding rope body of the smart rope for assisting in acquiring information on rope tension changes during the swinging of the smart rope.

[0076] By clearly defining the hardware sensor configuration of the smart jump rope, and by using multiple types of sensors to collaboratively collect three-dimensional motion data, a multidimensional dataset including translational acceleration, rotational angular velocity, spatial posture, and rope tension is constructed, providing raw signal input for subsequent motion analysis.

[0077] Sensor hardware deployment: Handheld sensor group: A three-axis accelerometer (such as ADXL345) and a three-axis angular velocity sensor (gyroscope, such as MPU6050) are integrated inside the left and right handheld controllers to collect translational acceleration data of the controllers on the X / Y / Z axes (unit: m / s). 2 The system collects angular velocity data (in rad / s) around the X / Y / Z axes and calculates spatial attitude angles (pitch, roll, yaw) using gyroscopes. A miniature tension sensor (such as a resistance strain gauge sensor) is embedded at the connection between the jump rope and the handle or in the middle of the rope to monitor real-time tension changes (in N) during rope swinging. The tension increases significantly when the rope is taut and decreases when it is slack.

[0078] The data acquisition mechanism involves each sensor connecting to the main control chip (such as STM32) via I2C or SPI bus to acquire data at a synchronous frequency of ≥100Hz, ensuring that details of high-frequency actions (such as rapid rope swinging) are not lost.

[0079] Unlike traditional single accelerometer solutions, this system adds a gyroscope to acquire rotational angular velocity and attitude angle, combined with a tension sensor to capture the rope's tautness. This creates a three-dimensional data input of "handle motion trajectory + rope mechanical signal," overcoming the blind spots of single sensors in identifying scenarios such as "empty swing without jump" or "insufficient rope swing force." Clearly defining sensor types, installation locations, and communication protocols provides a reusable hardware solution for mass production, reducing subsequent R&D costs.

[0080] In some embodiments, the rope skipping motion features include the periodic features, displacement amplitude features, and angular velocity change features of normal rope skipping movements; the feature extraction of the three-dimensional motion data to obtain the rope skipping motion features includes: preprocessing the original three-dimensional motion data, removing high-frequency noise and abnormal abrupt data through a sliding window filtering algorithm to generate a smooth and continuous motion data sequence; performing time-domain feature analysis on the preprocessed sequence to extract the time period of a single rope skipping movement, the spatial amplitude extreme value of the handle displacement, and the peak value of the angular velocity change; performing frequency-domain feature analysis to obtain the main frequency component and secondary frequency component of the rope skipping motion through Fourier transform; and constructing the rope skipping motion features by combining the time-domain and frequency-domain features.

[0081] By denoising and extracting features from the raw sensor data, and combining time-domain and frequency-domain analysis, the core features (period, amplitude, and frequency) of the rope skipping motion are separated from the noise signal, and a feature vector that can be used for motion recognition is constructed.

[0082] Data preprocessing employs sliding window mean filtering (window size 50ms) to reduce noise in acceleration, angular velocity, and tension signals, eliminating high-frequency vibration noise (such as slight hand tremors); median filtering is used to correct abrupt data (such as the impact peak when the rope hits the ground) to generate a smooth continuous data sequence.

[0083] Temporal feature extraction: Periodic features: Calculate the time interval between the lowest points of two adjacent handle swings to reflect the rope skipping frequency; Displacement amplitude: Extract the extreme difference of the handle displacement in the Z-axis (vertical direction) during a single action (highest point - lowest point) to characterize the swing amplitude; Peak angular velocity: Capture the maximum value of the sudden change in angular velocity when the rope swing direction changes (e.g., from clockwise to counterclockwise) to reflect the swing force.

[0084] Frequency domain feature extraction involves performing a Fast Fourier Transform (FFT) on the preprocessed signal to calculate the energy distribution in the 0.5-5Hz frequency band. The dominant frequency (corresponding to the main frequency of rope skipping) and secondary frequency (such as the second harmonic of arm swing) with the highest energy proportion are extracted to distinguish between regular rope skipping movements and random movements.

[0085] By filtering out environmental interference and motion distortion noise, the stability of feature extraction is ensured, and periodic misjudgment caused by slight hand shaking is avoided. By combining time domain (motion details) and frequency domain (frequency characteristics) features, a multi-dimensional motion descriptor is formed, which provides rich discrimination basis for the accurate segmentation of subsequent target data segments.

[0086] In some embodiments, identifying the target data segment corresponding to a valid rope skipping action in the three-dimensional motion data through the rope skipping motion features includes: dynamically matching the real-time acquired three-dimensional motion data sequence with a preset normal rope skipping action feature template, wherein the feature template includes a period threshold range, a displacement amplitude threshold range, and an angular velocity change rate threshold range pre-trained for different user body types; when the period feature of any segment of data in the three-dimensional motion data sequence falls within the corresponding threshold range, the displacement amplitude feature reaches the minimum displacement threshold for bottoming out and jumping, and the angular velocity change feature conforms to the direction conversion law of the handle swing, it is determined to be the target data segment corresponding to a valid rope skipping action.

[0087] Based on user-personalized motion feature templates, the system dynamically matches real-time collected motion data to identify data segments that meet the characteristics of "effective rope skipping motions" and excludes invalid swings (such as adjusting rope length) or non-standard movements (such as swinging one arm).

[0088] Feature template pre-training collects standard jump rope data from users of different body types (grouped by height and arm span) to train and obtain the periodic threshold range (e.g., 0.4-0.6s / time for children, 0.2-0.4s / time for adults), displacement amplitude threshold (e.g., vertical displacement of the handle ≥20cm to avoid misjudgment of small swings), and angular velocity change rate threshold (angular velocity change rate ≥50rad / s² during direction change to ensure effective switching of swing direction).

[0089] Real-time matching and judgment are performed by dividing the real-time data sequence into 500ms sliding windows and checking each window individually: whether the main frequency of the signal within the window falls within a preset frequency range (e.g., 1.5-5Hz, corresponding to 90-300 times / minute); whether the displacement amplitude is greater than or equal to the minimum threshold (a necessary condition for bottoming out and jumping); and whether there is an alternating direction of angular velocity change (a periodic change of clockwise → counterclockwise → clockwise). If all the above conditions are met, the data segment is determined to be a valid action target.

[0090] By using body type grouping pre-training templates, we can solve the common threshold misjudgment problem caused by differences in the range and frequency of movements between adults and children, and professional users and ordinary users. For example, we can prevent children from being missed due to insufficient arm strength and small swing amplitude. By combining the direction conversion law (alternating changes in angular velocity), we can eliminate "continuous swinging in a single direction" (such as invalid movements when adjusting the rope length) and ensure that only complete jump rope cycle movements are recognized.

[0091] In some embodiments, obtaining the time interval and displacement trajectory repeatability of two adjacent actions in the target data segment includes: performing temporal segmentation on the continuous target data segment, extracting the start time points of two adjacent valid actions to calculate the time interval, the normal rope skipping frequency range being dynamically adjusted according to the user's age and physical condition, the normal rope skipping frequency including an interval formed by a preset proportion of fluctuation above and below the average frequency based on the user's historical rope skipping data statistics; measuring the displacement trajectory repeatability by calculating the Euclidean distance similarity of the handle spatial displacement trajectory in two adjacent actions, the preset threshold being preset according to the standard action model of rope skipping.

[0092] Within the valid motion data segment, abnormal counts are further filtered by time interval and trajectory similarity to eliminate motion deformations caused by fatigue (such as sudden frequency changes or trajectory chaos) or non-skipping motions (such as temporary pauses for adjustment).

[0093] Dynamic calculation of time interval: Base frequency determination: The average frequency is calculated based on the user's historical jump rope data (such as the average frequency f_avg of the last 10 exercises), and a dynamic range (f_avg×0.85 to f_avg×1.15) is formed by allowing a fluctuation of 15% to avoid misjudgment of changes in user status (such as frequency decrease after fatigue) by fixed frequency thresholds; Anomaly detection: If the time interval Δt between adjacent actions exceeds the dynamic range, it is judged as an anomaly (such as Δt being too small corresponding to "high-frequency empty swing without jumping", Δt being too large corresponding to "missed jump").

[0094] Displacement trajectory repeatability calculation: The handle displacement trajectory of a single action is represented as a three-dimensional coordinate point sequence (x1, y1, z1)...(xn, yn, zn). The average distance between two adjacent action trajectory points is calculated using the Euclidean distance formula.

[0095] ;

[0096] Where (xi,yi,zi) are the three-dimensional spatial coordinates of the handle at the i-th sampling point during the first valid jump rope action (unit: meters or centimeters, dimensions must be consistent). The real-time position of the handle in the world coordinate system is obtained by combining the three-axis accelerometer and gyroscope built into the handle with spatial attitude calculation (such as complementary filtering algorithm). (The origin of the coordinate system needs to be determined through initial calibration, usually the initial position of the handle when the user is standing naturally is taken as the origin.)

[0097] (xi′,yi′,zi′) is the three-dimensional spatial coordinate of the handle at the i-th sampling point in the second consecutive valid rope skipping action (strictly aligned with the sampling point sequence of the first action).

[0098] Key requirements: The sampling points for the two actions must be time-synchronized (e.g., based on the start time of the action, sampling at the same time interval to ensure that the action phase corresponding to i is consistent, such as the lowest and highest points of the rope swing). The real-time position of the handle in the world coordinate system is obtained through the built-in three-axis accelerometer and gyroscope, combined with spatial attitude calculation (e.g., complementary filtering algorithm). (The origin of the coordinate system needs to be determined through initial calibration, usually using the initial position of the handle when the user is naturally standing as the origin.)

[0099] n is the total number of sampling points for a single rope skipping motion (i.e., the number of sampling points within one complete motion cycle). It is determined by the sensor sampling frequency and the motion cycle. For example, if the sampling frequency is 100Hz and the single motion cycle is 0.4 seconds, then n = 100 × 0.4 = 40 points. It needs to be dynamically adjusted through motion cycle detection (such as the change in cycle when skipping rope at different frequencies, to ensure that n always covers a complete motion cycle).

[0100] It is the three-dimensional Euclidean distance between the i-th corresponding sampling points in two adjacent actions, which measures the deviation of the handle position at that moment.

[0101] The maximum possible distance is the theoretical maximum single displacement that the handle can produce in three-dimensional space (used to normalize the distance, mapping the similarity results to the [0, 1] interval). By recording the maximum displacement range of the handle along the X / Y / Z axes through the user's maximum swing of the jump rope, the diagonal distance is calculated as a baseline (e.g., if the handle's maximum movement along the X-axis is ±0.3m, along the Y-axis ±0.2m, and along the Z-axis ±0.5m, then the maximum possible distance is:

[0102] );

[0103] Empirical value method: Based on ergonomics, the maximum displacement range of the handle is preset to 1.0m when an adult is jumping rope normally (this can be determined by statistically averaging a large amount of user data).

[0104] A preset similarity threshold (e.g., 70%) is used; values ​​below this threshold are considered abnormal trajectories (e.g., trajectory deviation caused by unilateral arm swinging). The frequency range is dynamically adjusted based on user historical data to address the adaptation issues of traditional fixed frequency thresholds (e.g., default 120-200 times / minute) to individual differences and state changes. For example, it allows users with good physical fitness to exceed the default high-frequency limit during high-intensity training. Geometric similarity is used to quantify movement regularity, eliminating invalid movements such as unilateral force exertion and trajectory confusion caused by fatigue, further improving counting accuracy (especially in scenarios involving distorted movements after prolonged exercise).

[0105] In some embodiments, before obtaining the user's basic information, the method further includes: cross-validating the jump rope counting results by combining the foot contact signals collected during the user's jump to correct abnormal jump rope counting results. This includes: collecting foot contact signals by a pressure sensor worn on the user's foot or a pressure sensing device integrated into the jump rope pedal, extracting the periodic and intensity characteristics of the contact signals; comparing the period of the contact signals with the period of the jump rope action in time synchronization; and determining abnormal counting and correcting it when the difference between the number of contact signals and the number of jump rope counts within the same time window exceeds a preset threshold, or when the intensity of the contact signals does not reach the minimum pressure threshold for landing.

[0106] By introducing foot contact signals as an external verification source, and through time synchronization comparison of "handle action data + foot pressure data", counting errors caused by sensor data conflicts such as "swinging the rope without jumping" and "jumping without swinging the rope" are corrected.

[0107] Ground contact signal acquisition: Wearable solution: Users wear sports shoes or ankle bracelets with integrated pressure sensors to collect pressure signals when the feet land (threshold set to ≥30N, excluding slight ground contact such as tiptoeing); Fixed solution: The jump rope pedal has a built-in pressure sensing module to detect the impact force when jumping and landing.

[0108] The cross-validation logic is aligned using a time window: with a 1-second window, the number of rope jumps (N1) and the number of ground contact signals (N2) are counted within the window; the anomaly judgment rules are: if |N1-N2|>1, it is determined that there is "overcounting" or "missed count" (e.g., N1=2, N2=0 corresponds to "two empty swings"); if there is no ground contact signal within 50ms before and after a certain counting time and the pressure value is <20N, it is determined that "rope swing without jump", and the count is deducted; if there is no rope jump count within 50ms before and after a certain ground contact signal time, it is determined that "jump without swinging rope", and one count is added (it is necessary to combine with other sensors to confirm whether it is a valid action).

[0109] Breaking away from the traditional closed loop that relies solely on controller data, this system establishes a two-way verification of "rope swinging action (controller) - jumping action (feet)" through direct evidence of the "jumping action" caused by the foot touching the ground. This completely solves the counting error caused by "incomplete actions" (such as in existing solutions where a user swings the controller once but does not jump, yet it is counted as one jump). It effectively identifies complex scenarios such as "single-leg hop" and "unstable landing," and eliminates accidental touch interference by using a pressure signal strength threshold (such as ≥30N), thereby improving the system's reliability in real-world motion environments.

[0110] In some embodiments, the basic information includes age, weight, height, and preset exercise goals; the calculation of the current exercise intensity index by combining the basic information with real-time collected data on rope skipping frequency, continuous exercise time, cumulative rope skipping count, and heart rate includes: calculating the theoretical maximum heart rate based on the user's age and height, and obtaining the heart rate reserve percentage by combining real-time heart rate data; calculating the frequency change coefficient based on the difference between the rope skipping frequency and the user's historical average frequency, and estimating the real-time energy consumption rate by combining the continuous exercise time and cumulative rope skipping count through a preset energy consumption model; and weighting and fusing the heart rate reserve percentage, frequency change coefficient, and energy consumption rate to generate a dimensionless current exercise intensity index.

[0111] By constructing a multi-dimensional exercise intensity assessment model that includes heart rate, rope skipping frequency, and energy consumption, and generating quantitative indicators through weighted fusion, the problem of the crudeness of traditional single-indicator (such as heart rate) assessment is solved.

[0112] Heart rate reserve calculation: Theoretical maximum heart rate = 220 - user age; resting heart rate is the user's average morning heart rate (requires initial user setting); Heart rate reserve percentage (HRR%) calculation formula: HRR% = 100% * (real-time heart rate - resting heart rate) / (maximum heart rate - resting heart rate); Frequency variation coefficient: Calculate the percentage difference Δf% between the current jump rope frequency f and the user's historical average frequency f_avg.

[0113] ;

[0114] Energy expenditure is estimated using an empirical formula that combines body weight (kg), duration (min), and cumulative number of times (N): Energy expenditure rate (kcal / min) = (0.0175 * body weight * N) / duration;

[0115] Weighted fusion normalizes HRR% (weight 0.5), Δf% (weight 0.3), and energy consumption rate (weight 0.2) (mapping to 0-100), and generates an intensity index through linear weighting: Intensity index = 0.5 × HRR%_norm + 0.3 × Δf%_norm + 0.2 × energy consumption_norm.

[0116] Heart rate reflects cardiopulmonary load, frequency changes reflect fluctuations in exercise intensity, and energy consumption is directly related to training effect. Combining these three factors avoids the one-sidedness of a single indicator (e.g., a low-intensity state can still be identified when the heart rate is normal but the frequency drops sharply). Based on the user's age, weight, and historical data, parameters are dynamically adjusted (e.g., resting heart rate, historical average frequency) to solve the problem of neglecting individual differences by traditional fixed formulas (e.g., estimating energy consumption only based on weight). For example, the difference in energy consumption between obese and lean users is accurately quantified.

[0117] In some embodiments, the exercise goals include at least fat loss, endurance training, and explosive power training. The intensity grading standard for each exercise goal is divided according to heart rate zones, rope skipping frequency change rate, and energy consumption rate. The current exercise intensity index is dynamically graded according to a preset grading rule base, which contains intensity grading standards corresponding to different exercise goals, including: for fat loss goals, the grading standard is divided into three levels: low intensity, medium intensity, and high intensity based on heart rate reserve percentage, where medium intensity corresponds to 60%-75% of maximum heart rate and requires the rope skipping frequency to be stable at 80%-110% of the base frequency; for endurance training goals, the grading standard combines continuous exercise time and frequency change coefficient, requiring high intensity levels to maintain a frequency change coefficient within ±5% and a duration of more than 40 minutes; for explosive power training goals, the grading standard is based on the ratio of peak rope skipping frequency to base frequency, requiring high intensity levels to have a frequency surge exceeding 30% of the base frequency and maintained for at least 10 seconds.

[0118] For three core exercise goals—fat loss, endurance, and explosive power—differentiated intensity grading standards are designed, forming a dynamic rule base that includes heart rate zones, frequency stability, and surge magnitude, providing precise training guidance.

[0119] Fat loss target levels: Low intensity: HRR% < 60%, and frequency between f_avg×80%-110% (focuses on long-term low-load fat burning, avoiding high intensity which leads to excessive glucose metabolism); Medium intensity: HRR% 60%-75% (optimal range for fat metabolism), and stable frequency (fluctuation ≤ ±10%) to ensure maximum energy consumption efficiency; High intensity: HRR% > 75%, allowing frequency fluctuations of ±15% (suitable for advanced users to increase metabolic rate).

[0120] Endurance training is categorized as follows: Basic intensity: Duration > 20 minutes, frequency fluctuation ≤ ±15%; High intensity: Duration > 40 minutes, frequency fluctuation ≤ ±5% (emphasizing movement stability and cardiorespiratory endurance).

[0121] Explosive power training levels: Basic intensity: frequency increase > 20%, duration ≥ 5 seconds; High intensity: frequency increase > 30% (more than f_avg × 1.3 times), duration ≥ 10 seconds (stimulates fast muscle fiber contraction and increases instantaneous power).

[0122] Different grading standards directly correspond to the principles of exercise physiology (such as the optimal HRR range for fat loss and the stability requirements for endurance training), avoiding blind training by users. For example, the dual conditions of "sudden increase in intensity + duration" in explosive power training ensure effective muscle stimulation. Abstract exercise goals (such as "improving explosive power") are transformed into measurable parameter combinations (a sudden increase in frequency of 30% + maintenance for 10 seconds), making training effects traceable and evaluable, and solving the problem of "fuzzy intensity grading" in existing solutions.

[0123] In some embodiments, after dynamically classifying the current exercise intensity index according to a preset classification rule base, the method further includes: feeding back the real-time exercise intensity level to the user through an associated smart terminal based on the current exercise intensity classification result, and adjusting the intensity suggestions for subsequent exercise according to a preset adaptive strategy, wherein the adaptive strategy includes rules for dynamically optimizing the classification threshold based on the user's historical exercise data and current physical condition.

[0124] By adding real-time feedback and adaptive mechanisms after intensity grading, the current intensity level is transmitted to the user through a smart terminal, and the grading threshold is dynamically optimized based on historical data, forming a closed loop of "assessment-feedback-adjustment".

[0125] The real-time feedback mechanism displays the current intensity level (such as "medium intensity for fat loss" or "high intensity for explosive power") on the LED screen of the jump rope handle, the accompanying APP, or a smartwatch, and provides prompts on whether the target has been achieved (such as "stable frequency, maintain current intensity" or "heart rate is low, it is recommended to speed up the rope swing").

[0126] Adaptive Strategy: Dynamic Threshold Adjustment: After every 10 workouts, the grading threshold is updated based on the user's average intensity data. For example, for users who are consistently at "high endurance intensity," the duration requirement is gradually increased from 40 minutes to 50 minutes. Fitness Status Awareness: If a user's heart rate reserve percentage decreases significantly (e.g., by more than 10%) during three consecutive workouts, fitness is assessed as improved, and the intensity threshold for each target is automatically increased (e.g., the lower limit of HRR for moderate intensity in fat loss is increased from 60% to 65%).

[0127] Real-time visual feedback helps users intuitively understand their training status and enhances their motivation to exercise. By learning from historical data, it solves the problem that traditional fixed thresholds cannot adapt to the user growth path of "beginner → intermediate → professional". For example, it avoids the stagnation of training effect caused by the unchanged standard after long-term training and realizes personalized adaptation of "intensity grading standards dynamically evolving with user ability".

[0128] Please see Figure 3 As shown, Figure 3 This is a schematic diagram of the intelligent jump rope counting error correction and dynamic exercise intensity grading system 200 provided in this application embodiment. The intelligent jump rope counting error correction and dynamic exercise intensity grading system 200 is used to execute the steps of the intelligent jump rope counting error correction and dynamic exercise intensity grading methods shown in the above embodiments. The intelligent jump rope counting error correction and dynamic exercise intensity grading system 200 can be a single server or a server cluster, or it can be a terminal, such as a handheld terminal, a laptop computer, a wearable device, or a robot.

[0129] like Figure 3 As shown, the intelligent jump rope counting error correction and dynamic exercise intensity grading system 200 includes:

[0130] The data acquisition unit 201 is used to acquire three-dimensional motion data during the rope skipping process. The three-dimensional motion data includes at least the spatial displacement information, rotation angle information and motion acceleration information corresponding to the smart rope skipping.

[0131] Feature extraction unit 202 is used to extract features from three-dimensional motion data to obtain rope skipping motion features; and to identify the target data segment corresponding to the valid rope skipping action in the three-dimensional motion data through the rope skipping motion features.

[0132] The anomaly correction unit 203 is used to obtain the time interval and displacement trajectory repeatability of two adjacent actions in the target data segment. When the time interval exceeds the normal rope skipping frequency range or the displacement trajectory repeatability is lower than a preset threshold, it is determined to be an invalid count and discarded.

[0133] The dynamic grading unit 204 is used to acquire the user's basic information, and combine the basic information with real-time collected data on rope skipping frequency, continuous exercise time, cumulative rope skipping count and heart rate to calculate the current exercise intensity index; and to dynamically grade the current exercise intensity index according to a preset grading rule library, which contains intensity grading standards corresponding to different exercise goals.

[0134] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the intelligent jump rope counting error correction and dynamic exercise intensity grading system and each module described above can be referred to the corresponding content in the various embodiments of the intelligent jump rope counting error correction and dynamic exercise intensity grading method, and will not be repeated here.

[0135] The aforementioned method for correcting counting errors and dynamically grading exercise intensity in intelligent jump ropes can be implemented as a computer program. This computer program can be used in various ways, such as... Figure 3 It runs on the device shown.

[0136] Please see Figure 4 , Figure 4 This is a schematic block diagram of the structure of the smart jump rope provided in an embodiment of this application. The smart jump rope includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and internal memory.

[0137] The storage medium can store operating devices and computer programs. The computer program includes program instructions that, when executed, cause the processor to perform any method for correcting counting errors and dynamically grading exercise intensity in a smart jump rope.

[0138] The processor provides computing and control capabilities to support the operation of the entire smart jump rope.

[0139] The internal memory provides an environment for the execution of computer programs in non-volatile storage media. When the computer program is executed by the processor, it enables the processor to perform any intelligent jump rope counting error correction and dynamic exercise intensity grading method.

[0140] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 4The 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 terminal to which the present application is applied. A specific smart jump rope may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0141] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0142] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps:

[0143] Acquire three-dimensional motion data during the rope skipping process. The three-dimensional motion data includes at least the spatial displacement information, rotation angle information, and motion acceleration information corresponding to the smart rope skipping.

[0144] Feature extraction is performed on the three-dimensional motion data to obtain rope skipping motion features; the target data segment corresponding to the valid rope skipping action is identified in the three-dimensional motion data using the rope skipping motion features.

[0145] The time interval and displacement trajectory repeatability of two adjacent actions are obtained from the target data segment. When the time interval exceeds the normal rope skipping frequency range or the displacement trajectory repeatability is lower than the preset threshold, it is determined to be an invalid count and discarded.

[0146] The system acquires basic user information and combines this information with real-time data on jump rope frequency, duration of exercise, cumulative number of jumps, and heart rate to calculate the current exercise intensity index. It then dynamically grades the current exercise intensity index according to a pre-defined grading rule base, which includes intensity grading standards corresponding to different exercise goals.

[0147] In some embodiments, acquiring three-dimensional motion data during the rope skipping process includes: real-time data acquisition via a motion sensor assembly installed on the smart rope. The motion sensor assembly includes at least a three-axis accelerometer, a three-axis angular velocity sensor, and a gyroscope installed on the corresponding handle of the smart rope, for acquiring translational acceleration data, rotational angular velocity data, and spatial attitude angle data of the smart rope in three-dimensional space. The motion sensor assembly also includes a tension sensor installed on the corresponding rope body of the smart rope for assisting in acquiring information on rope tension changes during the swinging of the smart rope.

[0148] In some embodiments, the rope skipping motion features include the periodic features, displacement amplitude features, and angular velocity change features of normal rope skipping movements; the feature extraction of the three-dimensional motion data to obtain the rope skipping motion features includes: preprocessing the original three-dimensional motion data, removing high-frequency noise and abnormal abrupt data through a sliding window filtering algorithm to generate a smooth and continuous motion data sequence; performing time-domain feature analysis on the preprocessed sequence to extract the time period of a single rope skipping movement, the spatial amplitude extreme value of the handle displacement, and the peak value of the angular velocity change; performing frequency-domain feature analysis to obtain the main frequency component and secondary frequency component of the rope skipping motion through Fourier transform; and constructing the rope skipping motion features by combining the time-domain and frequency-domain features.

[0149] In some embodiments, identifying the target data segment corresponding to a valid rope skipping action in the three-dimensional motion data through the rope skipping motion features includes: dynamically matching the real-time acquired three-dimensional motion data sequence with a preset normal rope skipping action feature template, wherein the feature template includes a period threshold range, a displacement amplitude threshold range, and an angular velocity change rate threshold range pre-trained for different user body types; when the period feature of any segment of data in the three-dimensional motion data sequence falls within the corresponding threshold range, the displacement amplitude feature reaches the minimum displacement threshold for bottoming out and jumping, and the angular velocity change feature conforms to the direction conversion law of the handle swing, it is determined to be the target data segment corresponding to a valid rope skipping action.

[0150] In some embodiments, obtaining the time interval and displacement trajectory repeatability of two adjacent actions in the target data segment includes: performing temporal segmentation on the continuous target data segment, extracting the start time points of two adjacent valid actions to calculate the time interval, the normal rope skipping frequency range being dynamically adjusted according to the user's age and physical condition, the normal rope skipping frequency including an interval formed by a preset proportion of fluctuation above and below the average frequency based on the user's historical rope skipping data statistics; measuring the displacement trajectory repeatability by calculating the Euclidean distance similarity of the handle spatial displacement trajectory in two adjacent actions, the preset threshold being preset according to the standard action model of rope skipping.

[0151] In some embodiments, before obtaining the user's basic information, the method further includes: cross-validating the jump rope counting results by combining the foot contact signals collected during the user's jump to correct abnormal jump rope counting results. This includes: collecting foot contact signals by a pressure sensor worn on the user's foot or a pressure sensing device integrated into the jump rope pedal, extracting the periodic and intensity characteristics of the contact signals; comparing the period of the contact signals with the period of the jump rope action in time synchronization; and determining abnormal counting and correcting it when the difference between the number of contact signals and the number of jump rope counts within the same time window exceeds a preset threshold, or when the intensity of the contact signals does not reach the minimum pressure threshold for landing.

[0152] In some embodiments, the basic information includes age, weight, height, and preset exercise goals; the calculation of the current exercise intensity index by combining the basic information with real-time collected data on rope skipping frequency, continuous exercise time, cumulative rope skipping count, and heart rate includes: calculating the theoretical maximum heart rate based on the user's age and height, and obtaining the heart rate reserve percentage by combining real-time heart rate data; calculating the frequency change coefficient based on the difference between the rope skipping frequency and the user's historical average frequency, and estimating the real-time energy consumption rate by combining the continuous exercise time and cumulative rope skipping count through a preset energy consumption model; and weighting and fusing the heart rate reserve percentage, frequency change coefficient, and energy consumption rate to generate a dimensionless current exercise intensity index.

[0153] In some embodiments, the exercise goals include at least fat loss, endurance training, and explosive power training. The intensity grading standard for each exercise goal is divided according to heart rate zones, rope skipping frequency change rate, and energy consumption rate. The current exercise intensity index is dynamically graded according to a preset grading rule base, which contains intensity grading standards corresponding to different exercise goals, including: for fat loss goals, the grading standard is divided into three levels: low intensity, medium intensity, and high intensity based on heart rate reserve percentage, where medium intensity corresponds to 60%-75% of maximum heart rate and requires the rope skipping frequency to be stable at 80%-110% of the base frequency; for endurance training goals, the grading standard combines continuous exercise time and frequency change coefficient, requiring high intensity levels to maintain a frequency change coefficient within ±5% and a duration of more than 40 minutes; for explosive power training goals, the grading standard is based on the ratio of peak rope skipping frequency to base frequency, requiring high intensity levels to have a frequency surge exceeding 30% of the base frequency and maintained for at least 10 seconds.

[0154] In some embodiments, after dynamically classifying the current exercise intensity index according to a preset classification rule base, the method further includes: feeding back the real-time exercise intensity level to the user through an associated smart terminal based on the current exercise intensity classification result, and adjusting the intensity suggestions for subsequent exercise according to a preset adaptive strategy, wherein the adaptive strategy includes rules for dynamically optimizing the classification threshold based on the user's historical exercise data and current physical condition.

[0155] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement the steps of the intelligent jump rope counting error correction and dynamic exercise intensity grading method provided in any embodiment of this application.

[0156] The computer-readable storage medium can be the internal storage unit of the smart jump rope described in the foregoing embodiments, such as the hard drive or memory of the smart jump rope. Alternatively, the computer-readable storage medium can be an external storage device of the smart jump rope, such as a plug-in hard drive, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the smart jump rope.

[0157] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for correcting counting errors and dynamically grading exercise intensity in an intelligent jump rope, characterized in that, Applications in smart jump ropes; including: Acquire three-dimensional motion data during the rope skipping process. The three-dimensional motion data includes at least the spatial displacement information, rotation angle information, and motion acceleration information corresponding to the smart rope skipping. Feature extraction is performed on three-dimensional motion data to obtain rope skipping motion features. These features include the periodicity, displacement amplitude, and angular velocity change characteristics of normal rope skipping movements. The rope skipping motion features are used to identify target data segments corresponding to valid rope skipping movements in the three-dimensional motion data. This includes dynamically matching the real-time acquired three-dimensional motion data sequence with a preset normal rope skipping movement feature template. The feature template includes pre-trained periodicity threshold ranges, displacement amplitude threshold ranges, and angular velocity change rate threshold ranges for different user body types. When the periodicity of any segment in the three-dimensional motion data sequence falls within the corresponding threshold range, the displacement amplitude reaches the minimum displacement threshold for bottoming out and jumping, and the angular velocity change characteristic conforms to the direction change law of the handle swing, it is determined to be the target data segment corresponding to a valid rope skipping movement. The time interval and displacement trajectory repeatability of two adjacent actions are obtained in the target data segment. When the time interval exceeds the normal rope skipping frequency range or the displacement trajectory repeatability is lower than the preset threshold, it is determined to be an invalid count and discarded. The rope skipping count results are cross-validated by combining the foot contact signal collected during the user's jump to correct abnormal rope skipping count results. The method acquires basic user information, including age, weight, height, and preset exercise goals. Combining this basic information with real-time collected data on jump rope frequency, duration of exercise, cumulative jump count, and heart rate, it calculates the current exercise intensity index. The current exercise intensity index is then dynamically graded according to a preset grading rule base. This rule base contains intensity grading standards corresponding to different exercise goals, including at least fat loss, endurance training, and explosive power training. The intensity grading standard for each exercise goal is divided based on heart rate zones, jump rope frequency change rate, and energy consumption rate. After dynamically grading the current exercise intensity index according to the preset grading rule base, the method further includes: based on the current exercise intensity grading result, providing real-time exercise intensity level feedback to the user via a linked smart terminal, and adjusting subsequent exercise intensity suggestions according to a preset adaptive strategy. This adaptive strategy includes rules for dynamically optimizing the grading threshold based on the user's historical exercise data and current physical condition.

2. The method according to claim 1, characterized in that, The acquisition of three-dimensional motion data during the rope skipping process includes: Data is collected in real time by a motion sensor assembly installed on the smart jump rope. The motion sensor assembly includes at least a three-axis accelerometer, a three-axis angular velocity sensor, and a gyroscope installed on the corresponding handle of the smart jump rope. It is used to acquire translational acceleration data, rotational angular velocity data, and spatial attitude angle data of the smart jump rope in three-dimensional space. The motion sensor assembly also includes a tension sensor installed on the corresponding rope body of the smart jump rope to assist in collecting information on the change of rope tension when the smart jump rope is swung.

3. The method according to claim 1, characterized in that, The step of extracting features from three-dimensional motion data to obtain jump rope motion features includes: The original 3D motion data is preprocessed, and high-frequency noise and abnormal abrupt data are removed by a sliding window filtering algorithm to generate a smooth and continuous motion data sequence. The preprocessed sequence is subjected to time-domain feature analysis to extract the time period of a single rope skipping action, the spatial amplitude extreme value of the handle displacement, and the peak value of the angular velocity change. Frequency-domain feature analysis is performed to obtain the main frequency component and secondary frequency component of the rope skipping motion through Fourier transform. The rope skipping motion features are constructed by combining the time-domain and frequency-domain features.

4. The method according to claim 1, characterized in that, The step of obtaining the time interval and displacement trajectory repeatability of two adjacent actions in the target data segment includes: The continuous target data segment is time-series segmented, and the start time points of two adjacent effective actions are extracted to calculate the time interval. The normal rope skipping frequency range is dynamically adjusted according to the user's age and physical condition. The normal rope skipping frequency includes an interval formed by a preset ratio of fluctuation above and below the average frequency based on the user's historical rope skipping data statistics. The repeatability of the displacement trajectory is measured by calculating the Euclidean distance similarity of the spatial displacement trajectory of the handle in two adjacent movements. The preset threshold is set in advance according to the standard movement model of rope skipping.

5. The method according to claim 1, characterized in that, The method of cross-validating jump rope counting results by combining foot contact signals collected during the user's jump to correct abnormal jump rope counting results includes: By collecting foot contact signals through pressure sensors worn on the user's feet or pressure sensing devices integrated into the jump rope pedal, the periodic and intensity characteristics of the contact signals are extracted. The ground contact signal cycle is compared with the rope skipping action cycle in time synchronization. When the difference between the number of ground contact signals and the number of rope skipping counts in the same time window exceeds the preset count threshold, or when the ground contact signal intensity does not reach the minimum pressure threshold for landing, it is judged as an abnormal count and is corrected.

6. The method according to claim 1, characterized in that, The system combines basic information with real-time data on jump rope frequency, duration of exercise, cumulative number of jumps, and heart rate to calculate the current exercise intensity index, including: The theoretical maximum heart rate is calculated based on the user's age and height, and the heart rate reserve percentage is obtained by combining real-time heart rate data. The frequency change coefficient is calculated based on the difference between the jump rope frequency and the user's historical average frequency. The real-time energy consumption rate is estimated by combining the continuous exercise time and the cumulative number of jump ropes through a preset energy consumption model. The weighted fusion of heart rate reserve percentage, frequency change coefficient, and energy consumption rate generates a dimensionless current exercise intensity index.

7. The method according to claim 1, characterized in that, The dynamic classification of the current exercise intensity index is performed according to a preset classification rule base, wherein the classification rule base contains intensity classification standards corresponding to different exercise goals, including: For weight loss goals, the grading standard is divided into three levels: low intensity, medium intensity, and high intensity based on the percentage of heart rate reserve. The medium intensity level corresponds to 60%-75% of the maximum heart rate and requires the jump rope frequency to be stable at 80%-110% of the base frequency. For endurance training goals, the grading standard combines continuous exercise time and frequency variation coefficient, requiring high-intensity levels to maintain a frequency variation coefficient within ±5% and a duration of more than 40 minutes; For explosive power training, the grading standard is based on the ratio of the peak frequency of rope skipping to the baseline frequency. The high-intensity level requires the frequency surge to exceed 30% of the baseline frequency and be maintained for at least 10 seconds.

8. A system for correcting counting errors and dynamically grading exercise intensity in an intelligent jump rope, characterized in that, Applications in smart jump ropes; including: The data acquisition unit is used to acquire three-dimensional motion data during the rope skipping process. The three-dimensional motion data includes at least the spatial displacement information, rotation angle information and motion acceleration information corresponding to the smart rope skipping. The feature extraction unit is used to extract features from three-dimensional motion data to obtain rope skipping motion features. The rope skipping motion features include the periodic features, displacement amplitude features, and angular velocity change features of normal rope skipping movements. The rope skipping motion features are used to identify the target data segment corresponding to the valid rope skipping movement in the three-dimensional motion data. This includes: dynamically matching the real-time acquired three-dimensional motion data sequence with a preset normal rope skipping movement feature template. The feature template includes pre-trained periodic threshold ranges, displacement amplitude threshold ranges, and angular velocity change rate threshold ranges for different user body types. When the periodic feature of any segment of data in the three-dimensional motion data sequence falls within the corresponding threshold range, the displacement amplitude feature reaches the minimum displacement threshold for bottoming out and jumping, and the angular velocity change feature conforms to the direction change law of the handle swing, it is determined to be the target data segment corresponding to the valid rope skipping movement. An anomaly correction unit is used to obtain the time interval and displacement trajectory repeatability of two adjacent actions in the target data segment. When the time interval exceeds the normal rope skipping frequency range or the displacement trajectory repeatability is lower than a preset threshold, it is determined to be an invalid count and discarded. Combined with the foot contact signal collected during the user's jump, the rope skipping count result is cross-validated to correct the abnormal rope skipping count result. The dynamic grading unit is used to acquire the user's basic information, including age, weight, height, and preset exercise goals. Combining this basic information with real-time collected data on jump rope frequency, duration of exercise, cumulative jump rope count, and heart rate, it calculates the current exercise intensity index. The unit then dynamically grades the current exercise intensity index according to a preset grading rule base. This rule base contains intensity grading standards corresponding to different exercise goals, including at least fat loss, endurance training, and explosive power training. The intensity grading standard for each exercise goal is divided based on heart rate zones, jump rope frequency change rate, and energy consumption rate. After dynamically grading the current exercise intensity index according to the preset grading rule base, the unit further includes: providing the user with real-time exercise intensity levels via a linked smart terminal based on the current exercise intensity grading results; and adjusting subsequent exercise intensity suggestions according to a preset adaptive strategy. This adaptive strategy includes rules for dynamically optimizing the grading threshold based on the user's historical exercise data and current physical condition.

Images