Application of wearable devices based on deep learning algorithm in rope skipping data monitoring

At present, wearable devices have some problems, such as poor adaptability to human motion behavior, and the recognition accuracy required for different wearers cannot be achieved. Based on the principle of deep learning algorithm, this paper realizes the development of intelligent rope skipping movement data monitoring system. Through the universal human body analysis model, the attention mechanism is introduced and embedded into the decoding network. The data set of rope skipping is classified by multiple labels, and the convolution of spatial graph is constructed, which is extended to the time series dynamics of moving human skeleton data. Aiming at the problem of complex information data in the process of moving human body recognition, we use pose estimation to calculate the key points of moving human body, extract the dynamic structure information of human skeleton sequence. Due to the problems of line of sight occlusion in the process of moving human target tracking, a target tracking algorithm based on multi domain convolution neural network is adopted to improve the feature extraction ability of the algorithm by segmenting the target to be tracked and identifying the area around the target. The data set of rope skipping is collected by wearable sensors, and the difference in the numerical range may be large, so the data is normalized. Finally, through the loss function, the fitting effect of neural network can be evaluated, and the gradient optimization model parameters can be calculated, and coping with different data changes. Through the final system performance test, it is verified that the accuracy rate of the system designed in this paper is above 90%, which can effectively monitor the data of skipping rope and be used in the actual operation of skipping rope.


Introduction
Certain exercise intensity is the key standard of scientific fitness, which plays an important role in formulating appropriate exercise programs and monitoring real-time exercise status. Only by choosing the appropriate exercise intensity can we effectively improve various functions of the body, and then maintain the functions of various systems of the body, which can not only strengthen the body, but also maintain the body. Therefore, how to arrange the intensity of exercise in the research field related to exercise intensity has always been a research topic of great concern. As an aerobic fitness exercise, rope skipping can effectively exercise most of the muscles of the human body, enhance muscle strength, improve cardiopulmonary function, exercise physical coordination, etc. at the same time, rope skipping can also play a certain decompression effect and relieve mood. Because rope skipping is simple and easy to learn, it has low requirements for venues and equipment, and has good exercise effect. It has become a national fitness project. However, the current research on rope skipping has a relatively single point of interest, mostly focusing on its impact on the body and the health effects brought by rope skipping to people. The grading of rope skipping intensity is not very mature. As a result, most fitness people who are interested in rope skipping are difficult to grasp the appropriate exercise intensity when they are doing rope skipping, so they cannot scientifically control all aspects of factors during rope skipping and properly evaluate their fitness status.
With the progress of micro electronic devices and sensing technology and the rapid development of deep learning algorithm technology, deep learning has become the focus of artificial intelligence (Ting et al. 2019). Wearable devices based on this algorithm have become a new research object, which can instantly monitor human action recognition and physical condition from all aspects (Franchino et al. 2018). The related research mainly focuses on optimizing the network layout structure, improving the training efficiency and enhancing the expansion ability. Through the comprehensive utilization of underlying data technology and deep learning technology, deep learning can directly select the most recognizable features from metadata in the most appropriate and effective way based on the database (Geluvaraj et al. 2019). Compared with human behavior monitoring based on computer vision recognition, this wearable device is not limited by most scenes and environments, and has less resource consumption and lower cost, which is more conducive to popularization and application (Oudah et al. 2020). The intelligent rope skipping evaluation system designed in this paper is based on the deep learning algorithm, applies wireless transmission and connectivity technology, intelligent body sensors, etc. in the rope skipping equipment, designs the hardware and software system of the intelligent rope skipper point-to-point, avoids some shortcomings of wireless transmission, and improves its comprehensive efficiency (Esteva et al. 2021). The stability and program expansibility of the internal system of the intelligent rope skipper device can make a motion plan in practice, Scientific rope skipping fitness provides a certain reference.

Related work
It is mentioned in the literature that rope skipping, as an early sports event, is widespread in the lives of people in many countries, and there are various forms of existence. Many countries regard rope skipping as a professional event of National Games (Ha et al. 2015). In the world, rope skipping can be called a real national sport. For the fitness effect of skipping rope, the literature believes that skipping rope can achieve the effect of body shaping, and it is very safe (Tse et al. 2017). Compared with other physical confrontation sports, its injury probability is very small, which is called the most perfect way of sports, so it often appears in physical education teaching. According to the literature, rope skipping is a sport with both competition and cooperation, which is very suitable for competition and speed competition (Dong et al. 2021). Carrying out rope skipping in school after school can enrich the content of school life, cultivate students' spirit of unity and cooperation, and improve their physical quality. The literature believes that rope skipping is a coordinated activity of the whole body, which is effective for the development of all aspects of the body and is very conducive to the cultivation of students' physical quality. The literature mentioned the participation of various muscles in rope skipping, and found that many muscles involved in the exercise are related to breathing, of which the most important muscle part is located in the chest and back. The development of these muscles can improve the respiratory rate (Dobrowolski et al. 2020). Therefore, long-term adherence to rope skipping can not only make people stronger, but also be conducive to the comprehensive health care of the respiratory system and cardiovascular system. Literature studies have confirmed that long-term rope skipping can prevent a variety of diseases, not limited to physical diseases, but also have a certain adjustment effect on mental and psychological abnormalities. The literature has carried out experiments on the relationship between rope skipping and mental health, and found that rope skipping can make people's emotional state change positively, and has a certain effect on people's mental health care (Kirthika et al. 2019). At the same time, the literature mentioned that rope skipping has low requirements for venues and equipment, and has low requirements for objective factors such as time and weather, so it is very easy to start a series of activities in the school environment.
In recent years, there have been many studies on human behavior analysis based on gravity centrifugal force sensing technology. From data acquisition, preprocessing to data subdivision design, there are different studies at each level, but at the same time of development, sensing technology still faces many key technical checkpoints and various challenges (Wu et al. 2009). The literature introduces that deep learning is a part of machine algorithm learning, which mainly uses the method of feature learning to identify the feature expression of the original sensor data. It is mentioned in the literature that the deep learning of machine is to transmit the electrical signal sent by the lower unit to the higher logical unit through the algorithm inside the computer, so as to simulate the symbolic structure of neural connection between neurons in the human brain (Zhu et al. 2017). Then, from the hierarchical description of multiple transformation stages, the characteristics are summarized to get the true communication of the data. The literature believes that compared with the shallow learning of computers, deeper learning can enable computers to fit more complex types of required functions, so as to grasp more abstract concepts and images. And it can establish a complex model with fewer parameters (Davoudi et al. 2019). Relevant experiments show that a function can be expressed in S-layer structure, and it requires exponential quantitative parameters to express the original function in S-1 layer structure. The literature emphasizes that the deep learning of computer is a generation model, which can automatically learn the complex data layer and content of sensor data, and even learn highdimensional data without obvious signs. The literature focuses on the advantages of deep learning algorithm, that is, it can learn all the features required for correlation from the sensor data without obvious signs (Fayyad et al. 2020). For the changes independent of the required elements, only invariants are learned, while for the learning of high-dimensional sensor data, the calculation method of nonlinear dimensionality reduction is used. In this paper, the wearable sensor based on deep learning adopts the human behavior recognition scheme, and introduces the general process of deep learning method for wearable human behavior recognition from all aspects (Gupta 2021). The basic data of the collected sensor is used as the original bottom data, and then the original data is preprocessed, and the moving target subject is extracted. Finally, the sample generalization and classification expression of rope skipping behavior are obtained.
3 Research on motion data processing technology based on deep learning algorithm

Analysis of moving human body based on joint multi label classification
Aiming at the problem of feature selection of rope skipping motion data, this paper introduces the attention mechanism, and carries out the feature selection of motion data by enhancing the characteristics of specific channels according to different images in the training process. At the same time, in order to improve the analysis effect of moving image data, the method of data label classification is introduced for data training. The training process is shown in Fig. 1. The data recognition network used in this paper includes encoder and decoder. When performing moving image recognition, the image is first sent to the encoder network. After identifying, segmenting, and processing the image data, we placed the results into the multi label classification branch, displayed the vector results of the multi label classification of the data, and output the results to multiple convolutional layers in the decoder network for vector calculation and feature map fusion. Finally, we obtained the corrected results of moving human feature recognition.
The size of the output feature image space is affected by many factors, including the size of the input feature image, convolution calculation step size and other factors. When convolution calculation is carried out, the conversion function U Convert the input feature map, C represents the number of channels of the output feature map, H represents the height of the feature map, w represents the width of the feature map, and the computational complexity of the conversion function is expressed as: According to the convolution decomposition method, a convolution kernel can be divided into two asymmetric convolution kernels, and the computational complexity is: From the perspective of space, the convolution filter is constructed in the neighborhood range of 1 step around each node, and a convolution kernel of a size of K 9 K is set, and C represents the number of channels. To extend the convolution calculation to the data sequence calculation of the moving human skeleton, it is necessary to redefine the neighborhood range of the convolution of the sampling graph, and define the sampling function as: Each data set is numbered, and the node neighborhood is mapped to the label of the data subset. The function of assigning weight is: The center of gravity of the moving human skeleton is obtained by calculating all the key points of the human body and selecting the coordinate mean value. The three subsets of the key point v divide the neighborhood of the center of gravity of the human body into three parts, namely, the key point, the neighborhood point closer to the center of gravity around the key point, and the point farther from the center of gravity around the key point, that is: By defining the sampling function and the function of assigning weights, we finally get: Through data multi label classification, the spatial graph convolution is constructed and extended to the time series dynamics of moving human skeleton data, so that the neighborhood calculation method on the time series of key points is as follows: C represents the kernel size in the dynamic time series, and extends the label mapping function to the spatio-temporal dynamics. The calculation method is: Finally, assuming that n represents the number of overall samples, in the dynamic behavior recognition of moving human body:

Moving human target tracking algorithm
In the process of moving human body monitoring, the target tracking algorithm can mark and analyze the characteristics of one or more moving individuals recognized by the camera in a certain period of time, so that the moving targets that need to be studied can be recognized and tracked in time in the subsequent shooting process. The single target tracking process is shown in Fig. 2.
In the process of single target tracking, first of all, we need to identify the characteristics of the first frame target by initializing the tracker, clearly mark the position and size of the initial target, describe its motion model, predict the target area of the current frame according to the characteristics of the first frame target, extract the underlying features within the selected range, calculate the confidence score, and select the candidate box with the highest score as the prediction recognition target, And fine tune the candidate box to adapt to the small changes of the tracking target.
Model and process the characteristics of the target by training a correlation filter, namely: Then perform Fourier transform operation on formula: Fourier transform operation can improve the calculation efficiency of the system and reduce the overall amount of calculation. Formula (14) after Fourier transform operation can be abbreviated as: where G represents the response diagram, F represents the image, H* represents the filter, and the solution of H* is: In the actual recognition process, the recognition effect is affected by many factors such as the transformation path of the tracked target, the occlusion of the field of vision, the change of light and shade, etc. Therefore, in order to improve the overall tracking and recognition accuracy, a multi template reference strategy is introduced: The value of each element in filter H is calculated as follows: In order to ensure the recognition efficiency, the template is updated online by:

Motion data preprocessing algorithm
The loss function of data series prediction task is calculated as follows: The loss function can evaluate the fitting effect of neural network and calculate the parameters of gradient optimization model.
The final total loss function is: In the process of rope skipping data collection, the self built data set is collected through wearable sensors, and the original data content is more, and the difference in the numerical range may be large. If it is directly input into the model without data preprocessing, the overall convergence speed and recognition efficiency will be affected.
Through normalization, all data will be scaled to the range of -1 to 1 dimensions, as shown in Fig. 3.
The normalized motion data as the input model can improve the overall convergence speed, and can also reduce the impact of different channel numerical scales on network parameters. At the same time, considering that in the recognition process, the motion data may have some similarity at a single time point, which will affect the prediction effect of the overall model, this paper selects the data in a certain time period for input.
At the same time, considering that the action categories recognized in the process of rope skipping are relatively simple, the main index used is the recognition accuracy, that is, the proportion of the number of samples with correct classification in the total number of samples. The specific calculation method is: The calculation formulas are respectively:  Application of wearable devices based on deep learning algorithm in rope skipping data monit… 6803 In the process of detecting the algorithm used in this paper, it is found that the model can predict most of the motion data well and complete periodic judgment except for a few data outliers.
4 Design of intelligent rope skipping evaluation system based on wearable devices 4.1 Structure design of intelligent rope skipping evaluation system According to the requirements of rope skipping data monitoring, this paper designs an intelligent rope skipping evaluation system based on wearable devices. It's hardware structure is shown in Fig. 4. It can be seen from Fig. 4 that in the intelligent rope skipping evaluation system, the computer, as the server of the whole system, can control the operation of the system software. During the operation of the system, 50 rope skippers can be accommodated at the same time to connect with the computer through the wireless communication network. In the process of rope skipping movement data acquisition, the server computer sends out instructions first, and the intelligent rope skipper technology system starts to operate. The rope skipping drives the rotating shaft. The sensor receives the relevant data of the rope skipping rotating shaft and the moving human body and transmits it to the microcontroller, and then transmits it to the general server computer through the wireless communication module. At the same time, the number of rope skipping is displayed on the LCD screen of the intelligent rope skipper in real time. During the operation of the whole system, the power module continuously provides power for each functional module. The administrator can check the data collected by each rope skipper in time through the general server computer, and give practical guidance to the individual athletes.

Analysis of speed characteristics of different rope skipping movements
This paper adopts the method of sports anatomy to analyze the axial movement of human body. The x-axis represents the sagittal axis, and the human body can be divided into two planes in the fore-and-aft direction according to the direction of the sagittal axis; The y-axis represents the frontal axis, which is perpendicular to the sagittal axis, and can divide the human body into left and right sides; The vertical axis passing through the plane is the vertical axis, represented by Z, which divides the human body into up and down directions. The acting point of the resultant force of the gravity on the whole moving human body is the total center of gravity of the human body. The rope skipping action studied in this paper is mainly in the standing position, that is, the arm turning the rope mainly swings in the horizontal plane, and the landing position of the moving human body's footsteps is basically unchanged. On the whole, the change trend of the center of gravity speed of the moving human body is obvious on the Y axis, and the change trend of the center of gravity speed of the moving human body on the Y axis is small on the X axis and Z axis. Therefore, this paper focuses on the change trend of the center of gravity speed of the moving human body on the Y axis. Firstly, the subjects were divided into high-level and low-level groups according to rope skipping skills, practice duration and physical factors, and their exercise data were monitored to obtain the change trend of body center of the two groups of athletes when practicing rope skipping spiral rotation and rope laying. The results are shown in Fig. 5.
On the whole, after T1 moment, the center of gravity of all moving human bodies began to move to the left. The change range of the high-level test group was smaller than that of the low-level test group, and the change trend was relatively small. The overall trunk activity was relatively stable, and there was no excessive torsion of the trunk angle because of the need to increase the arm swing range. Because novices usually use too much trunk displacement to increase the swing arm range in the process of rope skipping to prevent rope skipping from contacting the body and affecting the overall movement, relatively speaking, the y-axis speed of the low-level test group changes greatly, which is a common mistake for novices of rope skipping. In the process of rope skipping, we should increase the swing range of rope skipping through the rotation of wrist joint, so as to avoid the situation that rope skipping hits the body or is trampled by feet, rather than relying on the displacement or angle torsion of the body, so as to make rope skipping more stable and efficient.

Results of physical fitness characteristics under different rope skipping speeds
In the process of rope skipping data monitoring, the subjects' heart rate and fitness RPE are indicators to judge whether their exercise intensity is appropriate, and their center rate is the actual physiological indicators of exercise subjects, which can be detected by sensors in the process of rope skipping; RPE is the subjective psychological index of exercise subjects, which indicates the self-perceived exercise intensity evaluation of exercise subjects. It is a subjective standard used by exercise individuals to evaluate and measure exercise intensity. Both of them can reflect the appropriateness of exercise intensity. Carry out statistics on the correlation between rope skipping speed and heart rate and RPE of sports subjects of different sexes, and test the Pearson correlation between heart rate and RPE index and rope skipping speed when the sports speed is 60, 80, 100, 140 and 180, respectively. The results are shown in Table 1: It can be seen from the data in Table 2 that when the rope skipping speed is 60, 80, 100, 140 times/min, the heart rate and RPE Pearson of male and female exercise subjects have a certain correlation, and the significance is 0.00, which proves that under these exercise speeds, the correlation between the heart rate and RPE of subjects is significant. When the rope skipping speed is 180 times/min, the Pearson correlation between male heart rate and RPE is 0.15, with a significance of 0.52, and the Pearson correlation between female heart rate and RPE is 0.25, with a significance of 0.26, indicating that the correlation between subjects' heart rate and RPE is not significant at this time.  At the same time, Pearson correlation analysis was carried out on rope skipping speed, heart rate and RPE to prove the relationship between rope skipping speed and exercise intensity. The results are shown in Table 2: From the data in Tables 1 and 2, it can be seen that no matter male or female subjects, their rope skipping speed is positively correlated with heart rate and RPE, and the correlation is significant. In addition, when the rope skipping speed was 180 times/min, the correlation between male and female subjects' heart rate and RPE was not significant. It is proved that different individuals have certain differences in the acceptance of different rope skipping speed, and the changes of heart rate and RPE of exercise subjects are relatively unstable after rope skipping speed exceeds 140 times/min, which provides a certain reference for the subsequent analysis of the relationship between rope skipping speed and exercise intensity.

Accuracy analysis of rope skipping counting system module
In the data monitoring process of rope skipping, the size of the collected data, the type of data and the change of threshold may have a certain impact on the overall recognition accuracy of the system, so this paper determines the most appropriate data value by adjusting parameters. In order to improve the overall data calculation speed, the ratio of the final input data sequence and threshold is converted into an integer. Figure 6 shows the change of the counting accuracy of the rope skipper with the number of samples. Figure 7 shows the specific situation that the counting accuracy of rope skipper changes with the data sampling threshold.  Fig. 7 Variation of counting accuracy with threshold value It can be seen from the data in Figs. 6 and 7 that in order to improve the operation efficiency, the number of data points used in the statistical process is an integer power of 2. Among them, the number of data points is k = 64, the sampling interval is i = 50 ms, t = 1.6, the counting accuracy is as high as 95%, and the overall data recognition accuracy is 90% or more, which proves the availability of this system in rope skipping data monitoring.
5 Research on fitness program of rope skipping under the background of intelligence

Analysis of the relationship between rope skipping speed and exercise intensity
The research of this paper shows that for low-level subjects who do not jump rope frequently, their average heartbeat rate is the lowest when the rope skipping rate is 60 times/ min, and the average heartbeat rate is the highest when the rope skipping rate is 140 times/min, showing a slow upward trend. It can be seen that for these people who do not have rope skipping habits, the rope skipping rate of 60-120 times/min is more appropriate. The reason for the slowdown of heart rate in the later stage may be that when the rate reaches the peak, it is close to the maximum rope skipping intensity that ordinary humans can bear, so the rate of 140 times/min is close to reaching the maximum heart rate. When the rope skipping speed of high-level subjects is 60, 80, 100, 140 times/min, the average heartbeat rate is almost lower than that of low-level subjects with the same rate. Among them, the average heart rate of high-level subjects was the lowest at 60 beats/min and the highest at 160 beats/min. Therefore, we can know that for high-level subjects who have the habit of skipping rope, the skipping speed of 60-140 times/min is more appropriate, and its maximum bearing strength is about 160 times/min. Whether for people who are used to rope skipping or not, the average heart rate of women is generally higher than that of men. Therefore, with the same rope skipping speed, women's exercise intensity when doing rope skipping is higher than that of men, and the effect of rope skipping exercise should be more obvious in theory.
No matter men or women, when the rope skipping speed is 60-140 times/min, the average heart rate and RPE value of people who are not used to rope skipping are much higher than those who often jump rope. In other words, there is a significant difference between the heartbeat rate and RPE value of people who do not often jump rope and those who often jump rope at various jump rope rates. It can be seen that for people who do not often jump rope and people who often jump rope, the relative intensity of different rope skipping speeds is not the same. For two people at the same rope skipping speed, the actual exercise intensity obtained by people who often jump rope is lower than those who do not jump rope frequently. The reason may be that people who often jump rope have mastered various movements with higher efficiency after long-term continuous practice, and their rope skipping skills are more skilled. For example, the jumping height is lower than those who do not often jump rope, and the range of rope swinging is smaller, This saves unnecessary physical strength; Or people who often habitually jump rope have better cardiopulmonary function. During the practice of rope skipping, cardiopulmonary function is gradually enhanced, muscle strength is also improved, and the perception of physical fitness consumed by sports is low at the same intensity, which unconsciously increases muscle and physical endurance. However, it cannot be ruled out that it is the result of the joint action of two factors.
Through the experiment, we can draw a conclusion that the comprehensive quality level of the body is related to the proficiency of rope skipping skills, and the proficiency of rope skipping is extremely related to the individual's cardiopulmonary function and the degree of leg development exercise, but its relationship with vital capacity and upper limb strength is not very obvious. Therefore, we can speculate that people who often jump rope can effectively improve the endurance and explosive power of lower limbs, but the improvement of vital capacity and upper limb strength is not obvious.

Fitness value analysis of rope skipping
With the continuous development of society and the further improvement of people's living standards, physical exercise has become a part of people's daily life. Although there are many sports, most sports need specific venues and professional equipment, which is difficult to meet the huge exercise groups. As an economical exercise method, rope skipping is basically not affected by external objective factors such as production site, season, time, etc., and the equipment is also relatively easy to obtain, low cost and convenient to use. In addition, rope skipping is simple and easy to learn, both young and old. Therefore, as a national fitness sport, rope skipping is more universal than other sports, so it is a sport with high sports value. Nowadays, the form of rope skipping has evolved from the original simple jumping with two feet to a variety of fancy jumping methods, such as cross jumping and multi person jumping. It is both entertaining and ornamental, and integrates fun and the purpose of exercising. Rope skipping is also often used as a competition event in major activities and competitions at home and abroad.
Skipping rope is good for your health. Rope skipping is a sport that can mobilize the activities of the whole human body, which can make people's actions more agile, effectively promote human blood circulation, is very conducive to the development of physical coordination and balance ability, and can improve the comprehensive physical quality of rope skippers while entertaining. Long term adherence to rope skipping can also improve immunity and prevent a variety of cardiovascular diseases. Skipping rope can also improve heart function, because during skipping rope, high-speed breathing can make the blood in the body get more oxygen, make the cardiovascular system stronger, and reduce the incidence of cardiovascular disease and heart disease. In addition, skipping rope also has the effect of strengthening brain and intelligence, which can effectively prevent Alzheimer's disease. During rope skipping, the rim movement of the rope handle can effectively massage the palm, activate the peripheral nerves, and enhance the vitality of muscle cells and nerve cells. Skipping rope also has a good weight-loss effect. Relevant studies have shown that skipping rope consumes twice as much calories as jogging. In the process of rope skipping, the buttocks, legs, arms, back and other large muscle groups can be exercised, and the muscles and bones of the whole body are in a relatively moving state, so it can relieve nerves and tight muscles, and effectively reduce the symptoms of muscle soreness.

Research on rope skipping fitness program strategy
The sports fitness guidance program consists of three parts, mainly including exercise intensity, exercise time and exercise frequency. Specifically, we take the heart rate as the main reference benchmark, adjust the given exercise intensity through the main body's fatigue, and establish a personalized and multi-directional fitness guidance program. Exercise frequency has a great impact on the effect of exercise. For most adults, aerobic exercise for 3-5 days a week is more appropriate, and the frequency changes according to the change of exercise intensity. When the exercise frequency of an athlete exceeds 3 days/week, due to the improvement of his muscle and cardiopulmonary endurance, it will not be conducive to the continuous improvement of the exercise effect. When the exercise frequency exceeds 5 days/week, there will be a suspended platform, which is what we call the exercise platform period. For rope skipping, the exercise plan is: moderate intensity rope skipping five days a week, or high-intensity rope skipping at least three days a week, or a combination of the two. Only by reaching this skipping frequency can we achieve better results. According to the test of people's heart rate change and subjective fatigue during rope skipping, for people who do not often jump rope, whether men or women, it belongs to medium-intensity sports when the rope skipping frequency is 60-100 times/min, and it belongs to high-intensity sports when the rope skipping speed is 100-140 times/min. The exercise and fitness time of rope skipping should be 30-60 min of medium-intensity rope skipping, or 20-60 min of high-intensity rope skipping every day, or a combination of both. Combined with the above exercise time and intensity, it can be concluded that for people who do not often jump rope, the total amount of rope skipping with medium intensity is at least about 9000 times a week, and the total amount of rope skipping with high intensity is at least about 5000 times.

Conclusion
Rope skipping can not only effectively exercise the human body and strengthen the body, but also play a certain decompression effect and soothe the mood. Moreover, rope skipping has been one of the national fitness programs because it requires less space and technology, which is conducive to collective activities in schools. Because most people are concerned about the health effects of rope skipping, they lack a certain understanding of its exercise intensity and training methods. With the development of various micro electronic devices and sensing technologies, sports wearable devices have gradually become a hot spot, which can realize the real-time monitoring of individual sports and physical conditions. Based on the deep learning algorithm and wearable sensor, this paper preprocesses the original data of rope skipping, extracts the characteristics of the moving target, and realizes the design of intelligent rope skipping data monitoring system. Finally, combined with the actual situation of rope skipping and the fitness value of rope skipping, this paper puts forward a fitness guidance scheme of rope skipping under the intelligent background, mainly including exercise intensity, exercise time and exercise frequency. In practice, it can provide some reference for the formulation of rope skipping program and scientific rope skipping fitness.
Funding The authors have not disclosed any funding.
Data availability Data will be made available on request.

Declarations
Conflict of interest The authors declare that they have no conflict of interests.
Ethical approval This article does not contain any studies with human participants performed by any of the authors.