Method for Processing Sports Data during Football Kicking and Training Evaluation Method for Football Specialties

Abstract

Embodiments of the present application provide a method for processing motion data during kicking and a training evaluation method for football special training. The method includes: based on n sets of motion data collected by sensors installed on a football, using a 3D engine to generate a spatial coordinate point model for each set of motion data; establishing a core data region according to core spatial points and other spatial points, using the mini-batch gradient descent method to determine an objective function, performing convolution calculation on the motion data and the objective function to determine whether a certain moment is a kicking action, and calculating the kicking moment, the number of kicks, the kicking direction and the kicking force accordingly. After importing them into the training evaluation system, the special score and the overall score of football training are obtained, and a comprehensive evaluation five-dimensional graph is generated. The present application can obtain the special data of football special training more accurately and quickly in real time, and can comprehensively evaluate various indicators in the football training process, assist relevant personnel in judging the situation of football training, and make improvements accordingly.

Description

Technical Field

[0001] Embodiments of the present application relate to the technical field of data processing, and in particular, to a method for processing motion data during kicking and a method for training evaluation in football. Background Art

[0002] With the rapid development of technologies related to portable sensors in recent years, the measurement of physical data indicators during exercise has become easier. Most of the common sensors for capturing motion on the market are worn on the human torso or arm to collect physical data indicators of the wearer, such as heart rate, blood oxygen concentration, number of steps, step length, exercise distance, speed, energy consumption, etc. However, more accurate data cannot be obtained specifically in special training. For example, in football, existing wearable sensors cannot accurately obtain data such as whether a person is kicking a ball, the force of the kicked ball, the direction of the kicked ball, and the number of times of kicking the ball. Summary of the Invention

[0003] To solve the above-mentioned technical problems, embodiments of the present application provide a method for processing motion data during kicking and a method for training evaluation in football.

[0004] In a first aspect, embodiments of the present application provide a method for processing motion data during kicking, the method including:

[0005] Obtain n groups of motion data collected by sensors in a football training device, each group of the motion data including quaternions; import the quaternions into a 3D engine, and generate a spatial coordinate point model of each group of the motion data based on the 3D engine;

[0006] Taking the line segment composed of core spatial points as the axis and the minimum average value of the distances between other spatial points and the core spatial points as the radius, establish a core data area, where the core spatial points are the spatial coordinate points that appear the most frequently among the spatial coordinate points of the n groups of motion data;

[0007] Based on the spatial coordinate points within the core data area, use the mini-batch gradient descent method to determine an objective function, where the objective function is a neural network linear layer with four different data volumes;

[0008] Perform convolution calculation on the motion data and the objective function, and determine whether a certain moment is a kicking action based on the calculation result.

[0009] In a possible implementation manner, after determining whether a certain moment is a kicking action based on the calculation result, it further includes:

[0010] If it is a kicking action, record a consecutive a spatial coordinate points after the moment when the kicking action is completed;

[0011] Calculate the spatial vectors formed by adjacent pairs of spatial coordinate points, and based on the average value of a - 1 spatial vectors, obtain the movement direction at the current moment.

[0012] In a possible implementation manner, after determining whether a certain moment is a kicking action based on the calculation result, it further includes:

[0013] If it is a kicking action, record the acceleration value collected by the sensor at the moment when the kicking action is completed, and multiply it by the mass of the sensor to obtain the force value at the kicking moment.

[0014] In a possible implementation manner, it further includes recording the number of kicks based on each kicking action.

[0015] In a possible implementation manner, the following formula is used to import the quaternion into the 3D engine: qV = q * v * q -1

[0016] Where q is the quaternion produced by the sensor, V is the value of the default forward vector in the 3D engine, v is the value of the default forward vector in the 3D engine converted into the value of a preset vector quaternion, and q -1 represents the inverse of the quaternion produced by the sensor, and * represents the cross - product operation.

[0017] In a possible implementation manner, before establishing the core data region with the line segment composed of the core spatial points as the axis and the minimum average value of the distances between other spatial points and the core spatial points as the radius, it further includes:

[0018] Eliminate the jitter data, where the jitter data are the spatial coordinate points whose spatial distances in the previous and next frames differ from the average distance by more than a preset value.

[0019] In a possible implementation manner, calculating the objective function using the mini - batch gradient descent method based on the spatial coordinate points in the core data region includes:

[0020] Obtain all the spatial coordinate points remaining in the core data region in the nth group of motion data;

[0021] Select 3 consecutive spatial coordinate points among them, calculate the spatial vectors and spatial accelerations between adjacent pairs of spatial coordinate points, and use them as a training set;

[0022] Import m non - repeating training sets and training results as inputs into the deep neural network in the pytorch framework for training and calculation in batches;

[0023] When the loss value no longer decreases, determine the objective function.

[0024] In a possible implementation manner, the objective function is:

[0025]

[0026] Among them, J(θ) represents the loss value, m represents the number of input training sets + training results, and y i represents the result of each input training set, and h θ is the objective function, and x i represents the specific value of each input training set.

[0027] In a second aspect, an embodiment of the present application provides a training evaluation method for football special training, including:

[0028] Obtain the motion data collected by the sensors in the football training device;

[0029] Convert the motion data into second-order data, and the second-order data includes the kicking moment, the number of kicks, the kicking direction, and the kicking force;

[0030] Import the second-order data into the training evaluation system to obtain the scores of force control, direction control, accuracy, reaction time, and fluency;

[0031] Calculate the total score based on the scores of force control, direction control, accuracy, reaction time, and fluency, and generate a comprehensive evaluation five-dimensional diagram.

[0032] In a possible implementation manner, the calculation of the scores of force control, direction control, accuracy, reaction time, and fluency includes:

[0033] The score of the force control is calculated based on the deviation between the force of each kick and the specified kicking force;

[0034] The score of the direction control is calculated based on the deviation between the direction of each kick and the specified kicking direction;

[0035] The score of the accuracy is calculated based on the number of effective kicks, the number of kicks that should be made, and the total number of kicks. The number of effective kicks is the number of kicks that meet the force control and direction control standards at the same time;

[0036] The score of the reaction time is calculated based on the deviation between the time of each kick and the specified kicking time;

[0037] The score of the fluency is calculated based on the ratio of the highest number of consecutive effective kicks to the number of kicks that should be made in this training.

[0038] In summary, the present application includes the following beneficial technical effects:

[0039] 1. Based on n sets of motion data collected by sensors installed on a football, a spatial coordinate point model for each set of motion data is generated using a 3D engine; a core data region is established based on core spatial points and other spatial points, the objective function is determined using the mini-batch gradient descent method, the motion data is convolved with the objective function to determine whether a kicking action occurs at a certain moment, and based on this, the kicking moment, the number of kicks, the kicking direction, and the kicking force are calculated, enabling more accurate and rapid real-time acquisition of the special data for football special training.

[0040] 2. After importing the kicking moment, the number of kicks, the kicking direction, and the kicking force into the training evaluation system, the special score and the overall score of the football training are obtained, and a comprehensive evaluation five-dimensional graph is generated, which can assist relevant personnel in judging the situation of football training and making improvements accordingly.

[0041] It should be understood that the content described in the invention content part is not intended to limit the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become easily understood through the following description. Brief Description of the Drawings

[0042] Combined with the drawings and referring to the following detailed description, the above and other features, advantages, and aspects of the embodiments of the present application will become more obvious. In the drawings, the same or similar reference numerals represent the same or similar elements.

[0043] Figure 1 Shows a flowchart of the method for processing motion data during kicking in an embodiment of the present application.

[0044] Figure 2 Shows a three-dimensional structural schematic diagram of a football training device in an embodiment of the present application.

[0045] Figure 3 Shows a three-dimensional structural schematic diagram of the ball rope of the football training device in an embodiment of the present application

[0046] Figure 4 Shows a schematic diagram of the football training device during actual training in an embodiment of the present application.

[0047] Figure 5 Shows a flowchart of the training evaluation method for football special items in an embodiment of the present application.

[0048] Figure 6 Shows a schematic diagram of the comprehensive evaluation five-dimensional graph in an embodiment of the present application.

[0049] Figure 7 Shows a structural diagram of an electronic device in an embodiment of the present application.

[0050] Among them, 100 is the ball body; 210 is the first rope segment; 220 is the second rope segment; 230 is the third rope segment; 240 is the engaging member; 250 is the holding portion; 260 is the first connecting member; 270 is the second connecting member. Detailed implementation manners

[0051] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application.

[0052] To facilitate the understanding of the embodiments of the present application, the application scenarios involved in the present application will be introduced first. It should be noted that the application scenarios described in the embodiments of the present application are scenarios where students, athletes and other relevant personnel conduct football training. This is only to more clearly illustrate the technical solutions in the embodiments of the present application and does not constitute a limitation on the technical solutions provided in the embodiments of the present application. The method for processing motion data during football kicking provided in the embodiments of the present application is also applicable to similar or analogous scenarios where other relevant personnel kick the ball in football-specific training evaluation methods.

[0053] Figure 1 The flowchart of the method for processing motion data during football kicking in the embodiments of the present application is shown. Refer to Figure 1 and the method includes the following steps:

[0054] Step 101, obtain n groups of motion data collected by sensors in a football training device, and each group of the motion data includes quaternions.

[0055] Among them, the sensor is a sensor capable of collecting basic motion data of a football during motion, and the quaternion is a simple hypercomplex number, which consists of a real number plus three imaginary units and is mainly used to represent rotation in a three-dimensional space.

[0056] In the embodiments of the present application, the sensors used are installed on a specific football training device, such as Figure 2 shown.

[0057] Among them, the football training device includes a ball body 100 and a ball rope for football training. The ball body 100 is detachably connected to the ball rope to realize the traction of the ball body 100.

[0058] Specifically, as shown in Figure 2 and Figure 3As shown in the figure, the ball rope for football training includes a first rope segment 210, a second rope segment 220, and a third rope segment 230 that are connected in sequence. A clamping member 240 is provided at one end of the first rope segment 210 away from the second rope segment 220 for connection with the ball body 100; the ball body 100 has a clamping portion that matches the clamping member 240. In this embodiment, the clamping portion and the clamping member 240 are preferably detachably connected, which is convenient for storage. When training is not required, the ball body 100 can be used alone as a training tool.

[0059] A holding portion 250 is provided at one end of the third rope segment 230 away from the second rope segment 220, which is convenient for the trainer to hold and lift by hand, preventing the ball rope from slipping during the kicking training process.

[0060] In the embodiment of the present application, the holding portion 250 is preferably a handle, and the third rope segment 230 is connected to the end of the handle, which is convenient for the palm to hold the handle tightly and does not interfere with the movement of the third rope segment 230 at the same time.

[0061] The second rope segment 220 is rotatably connected to the first rope segment 210, that is, the connection between the second rope segment 220 and the first rope segment 210 has a 360-degree rotation freedom. Specifically, a first connecting member 260 is provided between the first rope segment 210 and the second rope segment 220 to offset the rotation transmitted from the first rope segment 210. The first connecting member 260 includes a first connecting portion for connecting with the first rope segment 210 and a second connecting portion for connecting with the second rope segment 220; the first connecting portion and the second connecting portion are hinged, and during the training process, the rotation action point is located inside the first connecting member 260, effectively reducing the wear on the first rope segment 210 and the second rope segment 220.

[0062] The third rope segment 230 is rotatably connected to the second rope segment 220, that is, the connection between the second rope segment 220 and the third rope segment 230 has a 360-degree rotation freedom. Specifically, a second connecting member 270 is provided between the second rope segment 220 and the third rope segment 230 to offset the rotational influence brought by the inertia of the second rope segment 220, ensure the smoothness of the third rope segment 230, and further prevent the rotation of the ball body 100 from affecting the attitude of the third rope segment 230 itself during the training process.

[0063] Specifically, the second connecting member 270 includes a third connecting portion for connecting with the second rope segment 220 and a fourth connecting portion for connecting with the third rope segment 230; the third connecting portion and the fourth connecting portion are hinged, and during the training process, the rotation action point is located inside the second connecting member 270, effectively reducing the wear on the second rope segment 220 and the third rope segment 230.

[0064] Refer to Figure 3During the training, the trainee lifts the ball body 100 to a preset height by holding the handle to perform ball kicking training; the first rope segment 210, the third rope segment 230 and the second rope segment 220 are rotatably connected. When the ball body 100 rotates, the first rope segment 210 rotates along with the ball body 100. The second rope segment 220 is connected to the first rope segment 210 through rotation, and the second rope segment 220 is not driven to rotate synchronously. The third rope segment 230 is connected to the second rope segment 220 through rotation, and multiple turns of the third rope segment 230 are not caused. The independence of the rope segment where the grip 250 is located is ensured, and uninterrupted training can be achieved without stopping each time to wait for the connecting rope to rotate and reset, which greatly improves the effective training efficiency.

[0065] In the embodiment of the present application, the locking member 240 is preferably a rope buckle, which has a compact structure, flexible rotation, high connection strength, and meets the training intensity; the first connecting member 260 and the second connecting member 270 are preferably universal buckles, which have a lightweight structure, effectively reduce the impact on the training process, and are easy to replace after wear.

[0066] The first rope segment 210 is provided with at least two strands to improve the strength of the connection segment with the football and meet the training intensity.

[0067] The length of the first rope segment 210 is L1, and the length of the second rope segment 220 is L2, L1∈[3cm,5cm], L2∈[30cm,32cm]. In the present embodiment, the length of the second rope segment 220 is the distance between the first connecting member 260 and the second connecting member 270. If the distance between the two is too large, the third rope segment 230 itself will have a large deviation in posture due to the twisting of the rope when kicking the ball. If the distance between the two is too small, the shaking caused by kicking the ball will be transmitted to the third rope segment 230, thereby affecting the deflection of the third rope segment 230 on the Y-axis. The solution disclosed in the present application can not only solve the influence of the shaking caused by kicking the ball on the third rope segment 230, but also solve the influence of the rotation of the ball body 100 on the twisting of the third rope segment 230.

[0068] The sensor is arranged on the third rope segment 230 to obtain the kicking data in training in real time. In the present embodiment, the kicking data that can be collected by the sensor include quaternions, acceleration and angular velocity, so as to obtain the movement direction of the ball body 100 corresponding to each kick, the kicking force, the number of kicks and other training data, so as to provide a reference for the training effect.

[0069] The length of the third rope segment 230 is L3, where L3 > 5 cm, ensuring that during training, the third rope segment 230 is not affected by the rotation of the ball body 100 connected to the first rope segment 210, enabling accurate testing of the acquisition device; the distance between the acquisition device and the holding part 250 is L, where L ∈ [3 cm, 5 cm]. In this embodiment, the set position of the acquisition device not only completely cancels out the rotation of the ball body 100 through the provided first connecting piece 260 and second connecting piece, without causing deviation in the self-attitude of the third rope segment 230 due to the rotation of the ball body 100, but also ensures that effective kicking data can be collected during the kicking process. At the same time, it ensures the accurate position of the acquisition device in the longitudinal direction, without self-deflection, guaranteeing the accuracy of the collected kicking direction and improving the stability of the acquisition device during training, ensuring the effectiveness of the collected data.

[0070] Through the rotational connection between the second rope segment 220 and the first rope segment 210, and the rotational connection between the third rope segment 230 and the second rope segment 220, the torsion of the rope during movement can be effectively reduced, ensuring that the acquisition device provided on the third rope segment 230 can accurately and effectively collect and record kicking data.

[0071] In the embodiment of the present application, in the same test environment, n groups of basic motion data during movement are detected and recorded by the motion sensor chip in the sensor, and the basic motion data recorded by the motion sensor is sent to the control terminal through Bluetooth 4.0. The control terminal includes but is not limited to electronic devices such as mobile phones and computers, and each group of motion data is recorded based on the same time interval.

[0072] Step 102, import the quaternion into the 3D engine, and generate a spatial coordinate point model for each group of motion data based on the 3D engine.

[0073] Among them, the 3D engine is a set of algorithm implementations that abstract substances in reality into forms such as polygons or various curves, perform relevant calculations in the computer, and output the final image.

[0074] In the embodiment of the present application, all the collected quaternions are imported into the 3D engine through the coordinate point construction formula, and the coordinate point construction formula is:

[0075] qV = q * v * q -1

[0076] Among them, q is the quaternion output by the sensor, V is the value of the default forward vector in the 3D engine, v is the value of the default forward vector in the 3D engine converted into the value of the preset vector quaternion, q -1 represents the inverse of the quaternion output by the sensor, and * represents the cross product operation.

[0077] Further, a spatial coordinate point model of each set of motion data is generated in the 3D engine.

[0078] It should be noted that the generated spatial coordinate point information should correspond one by one to the motion trajectory of the sensor in the real world. However, since there will always be a certain data deviation in the sensor, it is still necessary to screen out the spatial coordinate points that conform to the motion of the sensor in the real world.

[0079] Further, jitter data is removed, and the jitter data is the spatial coordinate points whose spatial distance within one frame before and after differs from the average distance by more than a preset value.

[0080] Among them, the average distance is calculated based on the distances between every two spatial coordinate points of each set of motion data, and the preset value is a value exceeding 1.5 times the average distance.

[0081] Further, after removing the jitter data, a density region is established based on the remaining spatial coordinate points, and n density regions are obtained based on n sets of motion data.

[0082] Among them, to make the experimental results more universal and accurate, usually the value of n needs to be greater than 100.

[0083] Step 103: Taking the line segment composed of the core spatial points as the axis and the minimum average of the distances between other spatial points and the core spatial points as the radius, a core data region is established, and the core spatial points are the spatial coordinate points that appear the most frequently among the spatial coordinate points of the n sets of motion data.

[0084] Specifically, first, the spatial coordinate points of all density regions are combined and observed to find the spatial points that appear the most frequently, and these types of coordinate points are used as core spatial points; then, the distances between all other spatial points and the core spatial points are calculated to obtain the minimum average of the distances between other spatial points and the core spatial points. Finally, taking the line segment composed of the core spatial points as the axis and the minimum average of the distances between other spatial coordinate points and the core spatial points as the radius, a core data region is established.

[0085] Step 104: Based on the spatial coordinate points within the core data region, the objective function is calculated using the mini-batch gradient descent method, and the objective function is a neural network linear layer with four different data volumes.

[0086] Specifically, first, all the spatial coordinate points remaining in the core data region in the nth set of motion data are obtained.

[0087] Further, three consecutive spatial coordinate points are selected, and the spatial vector and spatial acceleration between adjacent two spatial coordinate points are calculated, and they are used as a training set.

[0088] Specifically, three spatial coordinate points that can form two consecutive frames are selected, the spatial vectors and spatial accelerations between these spatial coordinate points are calculated, and these data are used to form a 1*8 data matrix, as shown in the following formula:

[0089] [x1,y1,z1,a1,x2,y2,z2,a2]

[0090] Among them, x1, y1, z1 represent the first spatial vector between the first two spatial coordinate points in three-dimensional space, a1 represents the first acceleration of this spatial vector; x2, y2, z2 represent the second spatial vector between the last two spatial coordinate points in three-dimensional space, and a2 represents the second acceleration of this spatial vector.

[0091] A data matrix composed of these data is used as a training set.

[0092] Furthermore, n non-repeating training sets and training results are imported into the pytorch framework in batches for training calculation.

[0093] Among them, the training result is the known result information input manually, with 1 set as the kicking state and 0 as the non-kicking state.

[0094] Specifically, using the machine learning function in the pytorch framework, countless non-repeating training sets and training results are imported in batches, and training calculations are performed through forward propagation, loss calculation, backpropagation, and weight parameter update.

[0095] Furthermore, when the loss value no longer decreases, the objective function is obtained.

[0096] Among them, the objective function is calculated using the following formula:

[0097]

[0098] Among them, J(θ) represents the loss value, m represents the number of input training sets + training results, y i represents the result of each input training set, h θ is the objective function, and x i represents the specific value of each input training set.

[0099] The finally obtained objective function h θ is a neural network linear layer with four different data volumes, as shown below:

[0100] 1. The linear layer of the first neural network

[0101] 8*6 matrix

[0102] [[-2.0858465e-01 4.2256981e-01 -6.3773811e-02 -8.4404916e-012.9696169e-01 -8.0306008e-02][8.1829590e-01 -3.3348568e-02 -3.2874532e-026.9495338e-01 1.5423660e-01-2.0698278e+00]

[0103] [-3.0405903e-01 2.1241915e-01 -4.2769842e-02 -6.5027428e-017.3617077e-03 1.5932605e-02][-2.5770218e+00 -2.5256660e+00 4.8891115e+00 -1.1283780e+00 4.3884535e+00 1.3587926e-02]

[0104] [-3.3575095e-02 -6.0653096e-01 5.7584938e-02 7.0231223e-01 -3.3358401e-01 8.6650729e-02][-4.5327666e-01 -1.7633407e-01 8.0567531e-02 -1.8680903e-01 -3.7389347e-01 -4.6210583e-02][8.4835678e-01 -3.6101708e-013.7634932e-02 1.2239283e+00 -6.2904367e-03 2.2635721e-03][-3.4972413e+00 -5.9436792e-01 1.7405230e+00 -1.0003562e+00 1.1643212e+00 -4.1551292e-01]]

[0105] 1*6 matrix

[0106] [[3.3285499 4.40595-4.966682-4.3135886 -5.0203648 2.006038]]

[0107] 2. Linear layer of the second neural network

[0108] 6*3 matrix

[0109] [[-0.1202608 -0.18058163 -0.2191126]

[0110] [-0.11482702 -0.17583184 -0.21495515]

[0111] [0.66076404 0.99890953 -0.0674594]

[0112] [0.15407622 0.23211639 -0.20161746] [0.24602908 0.3725015 0.05278474]

[0114] [-0.47738948 -0.7288987 -0.23836522]]

[0115] 1*3 matrix

[0116] [[-2.8674054 -2.349118 -0.17485738]]

[0117] 3. Linear layer of the third - layer neural network

[0118] 3*2 matrix

[0119] [[-1.8445239 -0.43225044]

[0120] [1.2203918 -0.13168007]

[0121] [-0.3407658 -0.07664967]]

[0122] 1*2 matrix

[0123] [[-2.1394548 -0.49731734]]

[0124] 4. Linear layer of the fourth - layer neural network

[0125] 2*1 matrix [[3.4910445] [0.5836249]]

[0128] 1*1 matrix

[0129] [[-0.00029497]]. Step 105, perform convolution calculation on the motion data and the objective function, and determine whether a certain moment is a kicking action based on the calculation result.

[0130] Specifically, the calculation result is between 0 and 1. When the calculation result is closer to 1, it is determined that the kicking action is in progress. When the calculation result is closer to 0, it is determined that the kicking action is not in progress.

[0131] Further, after determining that it is a kicking action, record the number of kicks based on each kicking action.

[0132] Further, after determining that it is a kicking action, record the 5 consecutive spatial coordinate points after the completion of the kicking action; calculate the spatial vector between two adjacent spatial coordinate points, and take the average of 4 consecutive spatial vectors to obtain the movement direction at the current moment.

[0133] It should be noted that when recording the coordinate points, ensure that the distance between each coordinate point and the previous coordinate point is greater than 1 to prevent the sensor from having no movement.

[0134] Further, after determining that it is a kicking action, record the acceleration value collected by the sensor at the moment of completing the kicking action, and multiply it by the mass of the sensor to obtain the force value at the kicking moment.

[0135] According to the embodiments of the present disclosure, the following technical effects are achieved:

[0136] Based on n groups of motion data collected by the sensor, use a 3D engine to generate a spatial coordinate point model for each group of motion data; establish a core data area according to the core spatial points and other spatial points, use the mini-batch gradient descent method to calculate the objective function, perform convolution calculation on the motion data and the objective function, obtain the feedback of whether it is a kicking action at a certain moment, and calculate the kicking moment, the number of kicks, the kicking direction and the kicking force based on this, which is more accurate and fast for the real-time acquisition of the special data of football special training.

[0137] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0138] Based on the above introduction of the method embodiments for processing motion data during kicking, the present invention also provides an embodiment of a training evaluation method for football special training to further illustrate the solution of the present application.

[0139] Figure 5 Shows the flowchart of the training evaluation method for football special training in the embodiments of the present application. Refer to Figure 5 This method includes:

[0140] Step 501: Obtain the motion data collected by the sensors in the football training device.

[0141] Step 502: Convert the motion data into second-order data according to the above-mentioned processing method for motion data during kicking. The second-order data includes the kicking moment, the number of kicks, the kicking direction, and the kicking force.

[0142] Step 503: Import the second-order data into the training evaluation system to obtain the scores for power control, direction control, accuracy, reaction time, and fluency.

[0143] Step 504: Calculate the total score based on the scores for power control, direction control, accuracy, reaction time, and fluency, and generate a comprehensive evaluation five-dimensional graph.

[0144] Among them, the comprehensive evaluation five-dimensional graph is as Figure 6 shown, including the scores for power control, direction control, accuracy, reaction time, and fluency. The processing method for the total score is as follows:

[0145] Total score = Accuracy score * 70% + [(Power control score + Direction control score + Reaction time score + Fluency score) / 4] * 30%.

[0146] Furthermore, the score for power control is calculated based on the deviation of the force of each kick from the specified kicking force.

[0147] Specifically, in different training course tests, there are different specified kicking force standards. Record and calculate the deviation of the force of each kick from the specified kicking force for scoring.

[0148] Furthermore, the score for direction control is calculated based on the deviation of the direction of each kick from the specified kicking direction.

[0149] Specifically, in different training course tests, there are different specified kicking direction standards. Record and calculate the deviation of the direction of each kick from the specified kicking direction for scoring.

[0150] Furthermore, the score for accuracy is calculated based on the number of effective kicks, the number of kicks to be made, and the total number of kicks. The number of effective kicks is the number of kicks that meet both the power control and direction control standards.

[0151] Among them, the number of kicks specified throughout the test is recorded as the number of kicks to be made; the number of kicks completed in the test is recorded as the total number of kicks; the number of times that meet both the power control and direction control standards among the total number of kicks is recorded as the number of effective kicks.

[0152] Specifically, when the total number of kicks exceeds the required number of kicks, the score = (the number of valid kicks / the total number of kicks) * 100%; when the total number of kicks does not exceed the required number of kicks, the score = (the number of valid kicks / the required number of kicks) * 100%.

[0153] Further, the scoring of the reaction time is calculated based on the deviation between the time of each kick and the specified kick time.

[0154] Specifically, there are different specified kick time standards in different training course tests. Record and calculate the deviation between the time of each kick and the specified kick time for scoring.

[0155] Further, the scoring of fluency is calculated based on the ratio of the maximum number of consecutive valid kicks to the required number of kicks in this training.

[0156] Specifically, fluency = (the maximum number of consecutive valid kicks / the required number of kicks in this training) * 100%.

[0157] According to the embodiments of the present disclosure, the following technical effects are achieved:

[0158] After importing the kick time, the number of kicks, the kick direction, and the kick strength into the training evaluation system, a special score and a total score for football training are obtained, and a comprehensive evaluation five-dimensional diagram is generated, which can assist relevant personnel in judging the situation of football training and making improvements accordingly.

[0159] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application. In addition, the embodiments of the football special training evaluation method and the processing method of sports data during kicking provided above belong to the same concept, and the specific implementation process can be seen in the embodiments of the processing method of sports data during kicking, which will not be elaborated here.

[0160] Figure 7 The structure diagram of an electronic device according to an embodiment of the present application is shown. Refer to Figure 7 , the electronic device 700 includes a processor 701 and a memory 703. Among them, the processor 701 and the memory 703 are connected, such as through a bus 702. Optionally, the electronic device 700 may further include a transceiver 704. It should be noted that in actual applications, the transceiver 704 is not limited to one, and the structure of the electronic device 700 does not constitute a limitation to the embodiments of the present application.

[0161] The processor 701 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules, and circuits described in connection with the disclosure of this application. The processor 701 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0162] The bus 702 may include a path for transmitting information between the above components. The bus 702 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 702 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 only a thick line is shown in the figure, but it does not mean that there is only one bus or one type of bus.

[0163] The memory 703 may be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, or it may also be an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0164] The memory 703 is used to store the application program code for executing the solution of this application, and is controlled by the processor 701 to execute. The processor 701 is used to execute the application program code stored in the memory 703 to implement the positioning of the map.

[0165] Among them, the electronic device includes but is not limited to: mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. It should be noted that Figure 7 The electronic device shown is only an example and should not impose any limitations on the functions and usage scope of the embodiments of this application.

[0166] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on the computer, the processes or functions described in the embodiments of this application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server, or data center to another website, computer, server, or data center in a wired manner (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that the computer can access, or a data storage device such as a server, data center, etc. that includes one or more available media integrated. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a digital versatile disc (DVD)), or a semiconductor medium (such as a solid state disk (SSD)), etc. It is worth noting that the computer-readable storage medium mentioned in the embodiments of this application can be a non-volatile storage medium, in other words, it can be a non-transitory storage medium.

[0167] It should be understood that the "at least one" mentioned herein refers to one or more, and the "multiple" refers to two or more. In the description of the embodiments of the present application, unless otherwise specified, " / " means "or". For example, A / B may mean A or B. The "and / or" herein is merely a description of the relationship between associated objects, indicating that there can be three relationships. For example, A and / or B may mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, for the convenience of clearly describing the technical solutions of the embodiments of the present application, in the embodiments of the present application, terms such as "first" and "second" are used to distinguish the same items or similar items with basically the same functions and effects. Those skilled in the art can understand that the terms such as "first" and "second" do not limit the quantity and execution order, and the terms such as "first" and "second" do not necessarily limit differences.

[0168] The above are the exemplary embodiments provided by the present application, which are not intended to limit the embodiments of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the embodiments of the present application shall be included within the protection scope of the present application.

Claims

1. A method for processing sports data during kicking, characterized in that, Including: Obtain n sets of motion data collected by sensors in a football training device, where each set of the motion data includes quaternions; import the quaternions into a 3D engine, and generate a spatial coordinate point model for each set of the motion data based on the 3D engine; with the line segment formed by the core spatial points as the axis and the minimum average of the distances between other spatial points and the core spatial points as the radius, establish a core data region, and the core spatial points are the spatial coordinate points that appear the most frequently among the spatial coordinate points of the n sets of motion data. Based on the spatial coordinate points within the core data region, use the mini-batch gradient descent method to determine an objective function, and the objective function is a neural network linear layer with four different data volumes; perform a convolution calculation on the motion data and the objective function, and determine whether a certain moment is a kicking action based on the calculation result.

2. The method according to claim 1, characterized in that, After determining whether a certain moment is a kicking action based on the calculation result, it further includes: if it is a kicking action, record a consecutive a spatial coordinate points after the moment when the kicking action is completed; calculate the spatial vectors formed by adjacent two spatial coordinate points, and obtain the motion direction at the current moment based on the average value of a - 1 spatial vectors.

3. The method according to claim 2, characterized in that, After determining whether a certain moment is a kicking action based on the calculation result, it further includes: if it is a kicking action, record the acceleration value collected by the sensor at the moment when the kicking action is completed, and multiply it by the mass of the sensor to obtain the force value at the kicking moment.

4. The method according to claim 3, characterized in that, It further includes recording the number of kicks based on each kicking action.

5. The method according to claim 1, characterized in that, Import the quaternion into the 3D engine using the following formula: qV = q * v * q -1 where q is the quaternion output by the sensor, V is the value of the default forward vector in the 3D engine, v is the value of the default forward vector in the 3D engine converted to the value of the preset vector quaternion, q -1 represents the inverse of the quaternion output by the sensor, and * represents the cross product operation.

6. The method according to claim 1, characterized in that, Before establishing the core data region with the line segment formed by the core spatial points as the axis and the minimum average of the distances between other spatial points and the core spatial points as the radius, it further includes: removing jitter data, and the jitter data are spatial coordinate points whose spatial distances within one frame before and after differ from the average distance by more than a preset value.

7. The method according to claim 1, characterized in that, Using the mini-batch gradient descent method to calculate the objective function based on the spatial coordinate points within the core data region includes: obtaining all the spatial coordinate points remaining in the core data region in the nth set of motion data; selecting 3 consecutive spatial coordinate points among them, calculating the spatial vectors and spatial accelerations between adjacent two spatial coordinate points, and using them as a training set; importing m non-repeated training sets and training results as inputs into a deep neural network in the pytorch framework for training and calculation in batches; when the loss value no longer decreases, determine the objective function.

8. The method according to claim 7, characterized in that, The objective function is: Where J(θ) is the objective function, m represents the number of input training sets + training results, yi represents the result of each input training set, hθ represents the loss value, and xi represents the specific value of each input training set.

9. A training evaluation method for football speciality, characterized in that, Including: Obtain the motion data collected by sensors in a football training device; Convert the motion data into second-order data according to the method for processing motion data during kicking as described in claim 4, and the second-order data includes the kicking moment, the number of kicks, the kicking direction, and the kicking force; import the second-order data into a training evaluation system to obtain scores for force control, direction control, accuracy, reaction time, and fluency. Calculate the total score based on the scores of the force control, direction control, accuracy, reaction time, and fluency, and generate a comprehensive evaluation five-dimensional graph.

10. The method according to claim 9, characterized in that, The calculation of the scores of the force control, direction control, accuracy, reaction time, and fluency includes: the score of the force control is calculated based on the deviation of the force of each kick from the specified kicking force; the score of the direction control is calculated based on the deviation of the direction of each kick from the specified kicking direction; the score of the accuracy is calculated based on the number of effective kicks, the number of kicks that should be made, and the total number of kicks, and the number of effective kicks is the number of kicks that meet the standards of both force control and direction control at the same time; the score of the reaction time is calculated based on the deviation of the time of each kick from the specified kicking time; the score of the fluency is calculated based on the ratio of the highest number of consecutive effective kicks to the number of kicks that should be made in this training.

Images