CN116637342A - Method for processing movement data during kicking and training evaluation method special for football - Google Patents
Method for processing movement data during kicking and training evaluation method special for football Download PDFInfo
- Publication number
- CN116637342A CN116637342A CN202310605770.8A CN202310605770A CN116637342A CN 116637342 A CN116637342 A CN 116637342A CN 202310605770 A CN202310605770 A CN 202310605770A CN 116637342 A CN116637342 A CN 116637342A
- Authority
- CN
- China
- Prior art keywords
- kicking
- training
- space
- football
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012549 training Methods 0.000 title claims abstract description 101
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000011156 evaluation Methods 0.000 title claims abstract description 28
- 238000012545 processing Methods 0.000 title claims description 12
- 230000009471 action Effects 0.000 claims abstract description 27
- 238000004364 calculation method Methods 0.000 claims abstract description 26
- 238000011478 gradient descent method Methods 0.000 claims abstract description 8
- 239000013598 vector Substances 0.000 claims description 25
- 230000001133 acceleration Effects 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 10
- 230000004044 response Effects 0.000 claims description 8
- 230000035484 reaction time Effects 0.000 claims description 5
- 230000008569 process Effects 0.000 abstract description 12
- 238000003672 processing method Methods 0.000 abstract description 5
- 230000006872 improvement Effects 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 23
- 239000011159 matrix material Substances 0.000 description 10
- 238000012360 testing method Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 238000005299 abrasion Methods 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B24/00—Electric or electronic controls for exercising apparatus of preceding groups; Controlling or monitoring of exercises, sportive games, training or athletic performances
- A63B24/0075—Means for generating exercise programs or schemes, e.g. computerized virtual trainer, e.g. using expert databases
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B69/00—Training appliances or apparatus for special sports
- A63B69/002—Training appliances or apparatus for special sports for football
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B71/00—Games or sports accessories not covered in groups A63B1/00 - A63B69/00
- A63B71/06—Indicating or scoring devices for games or players, or for other sports activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B2214/00—Training methods
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B2220/00—Measuring of physical parameters relating to sporting activity
- A63B2220/17—Counting, e.g. counting periodical movements, revolutions or cycles, or including further data processing to determine distances or speed
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B2220/00—Measuring of physical parameters relating to sporting activity
- A63B2220/50—Force related parameters
- A63B2220/51—Force
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B2243/00—Specific ball sports not provided for in A63B2102/00 - A63B2102/38
- A63B2243/0025—Football
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Physical Education & Sports Medicine (AREA)
- Mathematical Physics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Databases & Information Systems (AREA)
- Computer Graphics (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The embodiment of the application provides a processing method of motion data during kicking and a training evaluation method special for football, wherein the method comprises the following steps: generating a space coordinate point model of each group of motion data by using a 3D engine based on n groups of motion data acquired by a sensor arranged on the football; and establishing a core data area according to the core space points and other space points, determining an objective function by using a small-batch gradient descent method, carrying out convolution calculation on motion data and the objective function, determining whether a certain moment is a kicking action, calculating the kicking moment, the kicking frequency, the kicking direction and the kicking force according to the convolution calculation, and after the calculation is imported into a training evaluation system, obtaining a special score and a general evaluation of football training, and generating a comprehensive evaluation five-dimensional graph. The application can acquire the special data of football special training more accurately and rapidly in real time, comprehensively evaluate various indexes in the football training process, assist relevant personnel in judging football training conditions and make improvements according to the indexes.
Description
Technical Field
The embodiment of the application relates to the technical field of data processing, in particular to a method for processing movement data during kicking and a training evaluation method special for football.
Background
With the rapid development of portable sensor related technologies in recent years, measurement of body data index during exercise becomes easier. At present, most of common sensors for capturing motion in the market are worn on the trunk or arms of a human body, and collect body data indexes of the wearer, such as heart rate, blood oxygen concentration, step number, step length, movement distance, speed, energy consumption and the like. However, more accurate data cannot be obtained in special training, such as whether a ball is kicked, the force of the kicking, the direction of the kicking, the number of times of the kicking, and the like, cannot be accurately obtained by the existing wearing sensor in football.
Disclosure of Invention
In order to solve the above-mentioned technical problems, the embodiment of the application provides a method for processing motion data during kicking and a training evaluation method special for football.
In a first aspect, an embodiment of the present application provides a method for processing motion data during kicking, including:
acquiring n groups of motion data acquired by a sensor in a football training device, wherein each group of motion data comprises quaternions; importing the quaternion into a 3D engine, and generating a space coordinate point model of each group of motion data based on the 3D engine;
taking a line segment formed by core space points as an axis, taking the minimum average of distances between other space points and the core space points as a radius, and establishing a core data area, wherein the core space points are the space coordinate points with the largest repeated occurrence times in the space coordinate points of the n groups of motion data;
determining an objective function by using a small-batch gradient descent method based on the space coordinate points in the core data area, wherein the objective function is a neural network linear layer with four different data volumes;
and carrying out convolution calculation on the motion data and the objective function, and determining whether a certain moment is a kicking action or not based on a calculation result.
In one possible implementation manner, the determining whether a moment is a kicking action based on the calculation result further includes:
if the kicking action is performed, recording continuous a space coordinate points after the kicking action is completed;
and calculating a space vector formed by two adjacent space coordinate points, and obtaining the motion direction of the current moment based on the average value of the a-1 space vectors.
In one possible implementation manner, the determining whether a moment is a kicking action based on the calculation result further includes:
if the kicking action is performed, recording an acceleration value acquired by a sensor at the moment of finishing the kicking action, and multiplying the acceleration value by the mass of the sensor to obtain a force value at the moment of kicking.
In one possible implementation, the method further comprises recording a number of kicks based on each kicking action.
In one possible implementation, the quaternion is imported into the 3D engine using the following formula: qv=q×v×q -1
Wherein q is a quaternion produced by the sensor, V is a value of a default positive front vector of the 3D engine, V is a value of a preset vector quaternion converted from the value of the default positive front vector of the 3D engine, and q -1 Representing the inverse of the quaternion of the sensor output, represents a cross-product operation.
In one possible implementation manner, before the establishing the core data area, the method further includes taking a line segment formed by the core space points as an axis, taking a minimum average of distances between other space points and the core space points as a radius:
and eliminating jitter data, wherein the jitter data are space coordinate points of which the difference between the space distance and the average distance in each frame exceeds a preset value.
In one possible implementation manner, the calculating the objective function based on the spatial coordinate points in the core data area by using a small batch gradient descent method includes:
acquiring all space coordinate points which remain in a core data area in the nth group of motion data;
selecting 3 continuous space coordinate points, calculating space vectors and space accelerations between two adjacent space coordinate points, and taking the space vectors and the space accelerations as a training set;
m non-repeated training sets and training results are used as input to be led into a deep neural network in a pyrach frame for training calculation in a multiple-time mode;
when the loss value no longer drops, an objective function is determined.
In one possible implementation, the objective function is:
wherein J (θ) represents a loss value, m represents the number of input training sets+training results, y i Representing the results of the training set of each input, h θ As an objective function, x i Representing a specific value for each training set entered.
In a second aspect, an embodiment of the present application provides a training evaluation method for football special, including:
acquiring motion data acquired by a sensor in a football training device;
converting the motion data into second-order data, wherein the second-order data comprises kicking time, kicking frequency, kicking direction and kicking force;
the second-order data are imported into a training evaluation system to obtain scores of strength control, direction control, accuracy, response time and fluency;
and calculating total score based on the scores of the strength control, the direction control, the accuracy, the response time and the fluency, and generating a comprehensive evaluation five-dimensional graph.
In one possible implementation, the scoring calculation for strength control, direction control, accuracy, reaction time, and fluency includes:
the score of the power control is calculated based on the deviation of the force of each kicking and the specified kicking force;
the score of the direction control is calculated based on the deviation of the direction of each kicking and the specified kicking direction;
the scoring of the accuracy is calculated based on the effective kicking number, the number of the kicks to be kicked and the total kicking number, wherein the effective kicking number is the kicking number meeting the power control and direction control standards at the same time;
the score of the reaction time is calculated based on the deviation of the time of each kicking and the specified kicking time;
the fluency score is calculated based on the ratio of the highest continuous effective kicking number to the number of kicks that should be played in the present training.
In summary, the application has the following beneficial technical effects:
1. generating a space coordinate point model of each group of motion data by using a 3D engine based on n groups of motion data acquired by a sensor arranged on the football; establishing a core data area according to the core space points and other space points, determining an objective function by using a small batch gradient descent method, carrying out convolution calculation on motion data and the objective function, determining whether a certain moment is a kicking action, and calculating the kicking moment, the kicking frequency, the kicking direction and the kicking force according to the convolution calculation, so that the real-time acquisition of special data of special football training is more accurate and rapid;
2. after the ball kicking moment, the ball kicking frequency, the ball kicking direction and the ball kicking force are led into a training evaluation system, special scores and general scores of football training are obtained, a comprehensive evaluation five-dimensional chart is generated, and relevant personnel can be assisted to judge the condition of football training, and improvement is made accordingly.
It should be understood that the description in this summary is not intended to limit the critical or essential features of the embodiments of the application, nor is it intended to limit the scope of the application. Other features of the present application will become apparent from the description that follows.
Drawings
The above and other features, advantages and aspects of embodiments of the present application will become more apparent by reference to the following detailed description when taken in conjunction with the accompanying drawings. In the drawings, the same or similar reference numerals denote the same or similar elements.
Fig. 1 is a flowchart showing a processing method of motion data at the time of kicking a ball according to an embodiment of the present application.
Fig. 2 is a schematic perspective view of a football training device according to an embodiment of the present application.
Fig. 3 is a schematic perspective view showing a ball string of the soccer training device according to an embodiment of the present application
Fig. 4 shows a schematic diagram of a football training device according to an embodiment of the application during actual training.
Fig. 5 shows a flowchart of a training evaluation method for football specialty according to an embodiment of the present application.
FIG. 6 shows a schematic diagram of a comprehensive evaluation five-dimensional graph of an embodiment of the present application.
Fig. 7 shows a block diagram of an electronic device according to an embodiment of the present application.
100 parts of a ball body; 210. a first rope portion; 220. a second rope portion; 230. a third rope portion; 240. a clamping piece; 250. a grip portion; 260. a first connector; 270. and a second connecting piece.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of 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.
In order to facilitate understanding of the embodiments of the present application, first, an application scenario related to the present application will be described. It should be noted that, the application scenario described in the embodiment of the present application is a scenario in which related personnel such as students and athletes perform kicking training, which is only for more clearly describing the technical solution of the embodiment of the present application, and does not constitute a limitation on the technical solution provided by the embodiment of the present application. The processing method of the motion data during kicking provided by the embodiment of the application is also applicable to similar or analogous scenes of kicking by other related personnel as the training evaluation method of football special.
Fig. 1 shows a flowchart of a method for processing motion data at the time of kicking a ball according to an embodiment of the present application, referring to fig. 1, the method includes the steps of:
step 101, acquiring n groups of motion data acquired by a sensor in a football training device, wherein each group of motion data comprises quaternions.
The sensor is a sensor capable of collecting basic motion data of football during motion, and the quaternion is a simple supercomplex and is composed of a real number and three imaginary number units and is mainly used for representing rotation in a three-dimensional space.
In an embodiment of the present application, the sensor used is mounted on a specific football training device, as shown in fig. 2.
The football training device comprises a football body 100 and a football rope for football training, wherein the football body 100 is detachably connected with the football rope, and traction of the football body 100 is achieved.
Specifically, as shown in fig. 2 and 3, the ball rope for football training includes a first rope portion 210, a second rope portion 220 and a third rope portion 230 connected in sequence, and one end of the first rope portion 210, which is far away from the second rope portion 220, is provided with a clamping member 240 to be connected with the ball body 100; the ball body 100 has a clamping portion matched with the clamping piece 240; in this embodiment, the engaging portion and the engaging member 240 are preferably detachably connected, so that the ball body 100 can be used as a training tool when training is not needed.
The third rope portion 230 is provided with a grip portion 250 at one end far away from the second rope portion 220, so that a training person can hold the ball to lift the ball, and the ball is prevented from sliding down during the kicking training process.
In the embodiment of the present application, the gripping portion 250 is preferably a handle, and the third rope portion 230 is connected to an end of the handle, so that the palm can grip the handle without interfering with the movement of the third rope portion 230.
The second rope portion 220 is rotatably connected with the first rope portion 210, that is, the connection parts of the second rope portion 220 and the first rope portion 210 have 360 degrees of freedom of rotation; specifically, a first connector 260 is provided between the first rope portion 210 and the second rope portion 220 to counteract rotation transmitted from the first rope portion 210. The first connector 260 comprises a first connection part for connection with the first rope portion 210 and a second connection part for connection with the second rope portion 220; the first connecting portion is hinged to the second connecting portion, so that the rotation action part is located in the first connecting member 260 during the training process, and abrasion to the first rope portion 210 and the second rope portion 220 is effectively reduced.
The third rope portion 230 is rotatably connected with the second rope portion 220, i.e. the connection of the second rope portion 220 with the third rope portion 230 has a 360 degree rotational degree of freedom. Specifically, a second connector 270 is disposed between the second rope portion 220 and the third rope portion 230 to offset the rotation effect caused by the inertia of the second rope portion 220, ensure the stability of the third rope portion 230, and further prevent the rotation of the ball body 100 from affecting the posture of the third rope portion 230 during the training process.
In particular, the second connector 270 comprises a third connection for connection with the second rope portion 220 and a fourth connection for connection with the third rope portion 230; the third connecting part is hinged with the fourth connecting part, so that the rotation action part is positioned in the second connecting part 270 in the training process, and the abrasion to the second rope part 220 and the third rope part 230 is effectively reduced.
Referring to fig. 3, in the training process, a training person lifts the ball body 100 to a preset height by holding the handle to perform kicking training; the rotatable connection of the first rope portion 210, the third rope portion 230 and the second rope portion 220 is arranged, when the ball body 100 rotates, the first rope portion 210 is connected with the first rope portion 210 in a rotating way along with the rotation of the ball body 100, the second rope portion 220 is not driven to be screwed synchronously, the third rope portion 230 is connected with the second rope portion 220 in a rotating way, meanwhile, multiple circles of screwing of the third rope portion 230 is not caused, the independence of the rope portion where the holding portion 250 is located is guaranteed, uninterrupted training can be achieved, the rotation and reset of connecting ropes such as stopping each time are not needed, and the effective training efficiency is greatly improved.
In the embodiment of the application, the clamping piece 240 is preferably a rope buckle, has small structure, flexible rotation and high connection strength, and meets the training strength; the first connecting piece 260 and the second connecting piece 270 are all preferably universal buckles, so that the structure is light, the influence on the training process is effectively reduced, and the replacement is convenient after the wear.
The first rope portion 210 is provided in at least two strands, so that strength of a connecting portion with the soccer ball is improved, and training strength is satisfied.
The length of the first rope portion 210 is L1, the length of the second rope portion 220 is L2, L1 e [3cm,5cm ], L2 e [30cm,32cm ], in this embodiment, the length of the second rope portion 220 is the distance between the first connecting piece 260 and the second connecting piece 270, and too large a distance between the two can cause larger deviation of the posture of the third rope portion 230 due to twisting of the rope when kicking, and too small a distance between the two can be transmitted to the third rope portion 230 due to shaking caused by kicking so as to influence the deflection of the third rope portion 230 on the Y axis.
The third rope portion 230 is provided with the sensor to acquire kicking data in real time during training; in this embodiment, the kicking data that the acquisition sensor can acquire includes quaternion, acceleration and angular velocity, and then obtains training data such as the motion direction of the ball body 100, the force of kicking the ball, the number of kicking the ball that corresponds to each time, and provides a reference for the training effect.
The length of the third rope portion 230 is L3, and L3 is more than 5cm, so that the third rope portion 230 is not influenced by the rotation of the ball body 100 connected with the first rope portion 210 in the training process, and the accurate test of the acquisition device is realized; the distance between collection system and the portion 250 of gripping is L, L epsilon [3cm,5cm ], in this embodiment this collection system's setting position, both through the first connecting piece 260 that sets up, the rotation of ball body 100 is offset completely to the second connecting piece, can not produce the deviation to the self gesture of third rope segment 230 because of the rotation of ball body 100, can guarantee again to gather effectual kicking data at kicking in-process, guarantee collection system in longitudinal position accuracy simultaneously, do not produce the deflection of self, guarantee the accuracy of kicking direction of collection, improve collection system's stability in training process, guarantee the validity of collection data.
Through the rotary connection of the second rope portion 220 and the first rope portion 210, and the rotary connection of the third rope portion 230 and the second rope portion 220, the twisting of the rope in the moving process can be effectively reduced, and the acquisition device arranged on the third rope portion 230 can accurately and effectively collect and record kicking data.
In the embodiment of the application, n groups of basic motion data recorded by a motion sensor are detected by a motion sensor chip in the sensor under the same test environment, and the basic motion data recorded by the motion sensor are sent to a control terminal through Bluetooth 4.0, wherein the control terminal comprises but is not limited to electronic equipment such as a mobile phone, a computer and the like, and each group of motion data is recorded in a dividing way based on the same time interval.
Step 102, importing the quaternion into a 3D engine, and generating a space coordinate point model of each set of motion data based on the 3D engine.
The 3D engine is a collection of algorithm implementations that abstract real substances into expressions such as polygons or various curves, and perform correlation calculations in a computer to output a final image.
In the embodiment of the application, all collected quaternions are imported into a 3D engine through a coordinate point construction formula, wherein the coordinate point construction formula is as follows:
qV＝q*v*q -1
wherein q is a quaternion produced by the sensor, V is a value of a default positive front vector of the 3D engine, V is a value of a preset vector quaternion converted from the value of the default positive front vector of the 3D engine, and q -1 Representing the inverse of the quaternion of the sensor output, represents a cross-product operation.
Further, a spatial coordinate point model for each set of motion data is generated in the 3D engine.
It should be noted that, the generated spatial coordinate point information is in one-to-one correspondence with the motion trail of the sensor in the real world, but because the sensor always has a certain data deviation, the spatial coordinate points conforming to the motion of the sensor in the real world still need to be screened.
Further, eliminating jitter data, wherein the jitter data is a space coordinate point of which the difference between the space distance and the average distance in each frame exceeds a preset value.
The average distance is calculated based on the distance between every two space coordinate points of each group of motion data, and the preset value is a value exceeding 1.5 average distances.
Further, after eliminating the shake data, establishing a concentration area according to the reserved space coordinate points, and obtaining the concentration area for n times based on n groups of motion data.
In order to make the experimental result have more universality and accuracy, the value of n is usually greater than 100.
And 103, taking a line segment formed by core space points as an axis, taking the minimum average of the distances between other space points and the core space points as a radius, and establishing a core data area, wherein the core space points are the space coordinate points with the largest repeated occurrence times in the space coordinate points of the n groups of motion data.
Specifically, firstly, combining and observing the space coordinate points of all the concentration areas, finding out the space point with the largest repeated occurrence number, and taking the coordinate points as core space points; and calculating the distances from all other space points to the core space point to obtain the minimum average of the distances from other space points to the core space point, and finally establishing a core data area by taking a line segment formed by the core space points as an axis and the minimum average of the distances from other space coordinate points to the core space point as a radius.
And 104, calculating an objective function based on the space coordinate points in the core data area by using a small-batch gradient descent method, wherein the objective function is a neural network linear layer with four different data volumes.
Specifically, all spatial coordinate points in the n-th set of motion data which remain in the core data area are firstly acquired.
Further, 3 continuous space coordinate points are selected, and space vectors and space accelerations between two adjacent space coordinate points are calculated and used as a training set.
Specifically, 3 spatial coordinate points capable of forming two continuous frames are selected, the spatial vector and the spatial acceleration between the spatial coordinate points are calculated, and the data matrix 1*8 is formed by using the data, as shown in the following formula:
[x1,y1,z1,a1，x2,y2,z2,a2]
wherein x1, y1, z1 represent a first space vector between the first two space coordinate points in the three-dimensional space, and a1 represents a first acceleration of the space vector; x2, y2, z2 represent a second spatial vector between the last two spatial coordinate points in three-dimensional space, and a2 represents a second acceleration of this spatial vector.
A data matrix of these data is used as a training set.
Further, n non-repeated training sets and training results are used as input to be led into the pyrach framework for training calculation.
The training result is known result information input manually, 1 is set to be in a kicking state, and 0 is set to be in a non-kicking state.
Specifically, a machine learning function in a pyrach framework is used for importing innumerable non-repeated training sets and training results in a plurality of times, and training calculation is performed through forward propagation, calculation loss, backward propagation and updating weight parameters.
Further, when the loss value is no longer decreasing, an objective function is obtained.
Wherein the objective function is calculated using the following formula:
wherein J (θ) represents a loss value, m represents the number of input training sets+training results, y i Representing the results of the training set of each input, h θ As an objective function, x i Representing a specific value for each training set entered.
The final objective function h θ Is a linear layer of neural networks of four different data volumes, as follows:
1. linear layer of first layer neural network
8*6 matrix
[[-2.0858465e-01 4.2256981e-01 -6.3773811e-02 -8.4404916e-01 2.9696169e-01 -8.0306008e-02][8.1829590e-01 -3.3348568e-02 -3.2874532e-02 6.9495338e-01 1.5423660e-01-2.0698278e+00]
[-3.0405903e-01 2.1241915e-01 -4.2769842e-02 -6.5027428e-01 7.3617077e-03 1.5932605e-02][-2.5770218e+00 -2.5256660e+00 4.8891115e+00 -1.1283780e+00 4.3884535e+00 1.3587926e-02]
[-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-01 3.7634932e-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]]
1*6 matrix
[[3.3285499 4.40595-4.966682-4.3135886 -5.0203648 2.006038]]
2. Linear layer of a second layer neural network
6*3 matrix
[[-0.1202608-0.18058163-0.2191126]
[-0.11482702-0.17583184-0.21495515]
[0.66076404 0.99890953-0.0674594]
[0.15407622 0.23211639-0.20161746]
[0.24602908 0.3725015 0.05278474]
[-0.47738948-0.7288987-0.23836522]]
1*3 matrix
[[-2.8674054-2.349118-0.17485738]]
3. Linear layer of third layer neural network
3*2 matrix
[[-1.8445239-0.43225044]
[1.2203918-0.13168007]
[-0.3407658-0.07664967]]
1*2 matrix
[[-2.1394548-0.49731734]]
4. Linear layer of fourth layer neural network
2*1 matrix
[[3.4910445]
[0.5836249]]
1*1 matrix
[[ -0.00029497]]. And 105, carrying out convolution calculation on the motion data and the objective function, and determining whether a certain moment is a kicking action based on a calculation result.
Specifically, the calculation result is between 0 and 1, and when the calculation result is closer to 1, it is determined that the kicking operation is being performed, and when the calculation result is closer to 0, it is determined that the kicking operation is not being performed.
Further, after the kicking action is determined, the number of kicks is recorded based on each kicking action.
Further, after the kicking action is determined, recording continuous 5 space coordinate points after the kicking action moment is completed; and calculating the space vector between two adjacent space coordinate points, and taking the average value of 4 continuous space vectors to obtain the motion direction of the current moment.
It should be noted that, when recording the coordinate points, it is ensured that the distance between each coordinate point and the next coordinate point is greater than 1 to prevent the sensor from having any movement.
Further, after the kicking action is determined, the acceleration value acquired by the sensor at the moment of finishing the kicking action is recorded, and the acceleration value is multiplied by the mass of the sensor to obtain the force value at the moment of kicking.
According to the embodiment of the disclosure, the following technical effects are achieved:
based on n groups of motion data acquired by the sensor, generating a space coordinate point model of each group of motion data by using a 3D engine; and establishing a core data area according to the core space points and other space points, calculating an objective function by using a small-batch gradient descent method, carrying out convolution calculation on motion data and the objective function to obtain feedback of whether a certain moment is a kicking action, and calculating the kicking moment, the kicking frequency, the kicking direction and the kicking force according to the feedback, so that the real-time acquisition of special data of football special training is more accurate and rapid.
It should be noted that, for simplicity of description, the foregoing method embodiments are all described as a series of acts, but it should be understood by those skilled in the art that the present application is not limited by the order of acts described, as some steps may be performed in other orders or concurrently in accordance with the present application. Further, those skilled in the art will also appreciate that the embodiments described in the specification are alternative embodiments, and that the acts and modules referred to are not necessarily required for the present application.
Based on the above description of the embodiment of the method for processing the motion data during kicking, the application also provides an embodiment of a training evaluation method special for football, so as to further explain the scheme of the application.
Fig. 5 shows a flowchart of a training evaluation method for football special, referring to fig. 5, which includes:
step 501, acquiring motion data acquired by a sensor in a football training device;
step 502, converting the motion data into second-order data according to the processing method of the motion data during kicking, wherein the second-order data comprises kicking time, kicking frequency, kicking direction and kicking force;
step 503, importing the second-order data into a training evaluation system to obtain scores of strength control, direction control, accuracy, response time and fluency;
step 504, calculating total score based on the scores of the strength control, the direction control, the accuracy, the response time and the fluency, and generating a comprehensive evaluation five-dimensional graph.
The five-dimensional graph for comprehensive evaluation is shown in fig. 6, and comprises the scores of strength control, direction control, accuracy, response time and fluency, and the processing method of the total score is specifically as follows:
total score = accuracy score 70% + [ (strength control score + direction control score + response time score + fluency score)/4 ] × 30%.
Further, the score of the power control is calculated based on the deviation of the force of each kick from the prescribed kicking force.
Specifically, different specified kicking force standards exist in different training course tests, and deviation between the kicking force of each time and the specified kicking force is recorded and calculated for scoring.
Further, the score of the directional control is calculated based on the deviation of the direction of each kick from the prescribed kicking direction.
Specifically, different specified kicking direction standards exist in different training course tests, and deviation between each kicking direction and the specified kicking direction is recorded and calculated to score.
Further, the scoring of the accuracy is calculated based on the effective number of kicks, the number of kicks to be played, the total number of kicks, the effective number of kicks being the number of kicks that meet both the power control and directional control criteria.
Wherein, the number of kicks specified in the whole test process is recorded as the number of kicks to be played; the number of kicks completed in the test is recorded as the total number of kicks; and judging the times of meeting the power control and direction control standards in the total kicking number, and recording the times as the effective kicking number.
Specifically, when the total number of kicks exceeds the number of kicks to be played, the fraction= (effective number of kicks/total number of kicks) ×100%; when the total number of kicks does not exceed the number of kicks, the fraction= (effective kicks/number of kicks) ×100%.
Further, the score of the reaction time is calculated based on the deviation of the time of each kick from the prescribed kick time.
Specifically, different specified kicking time standards exist in different training course tests, and deviation between each kicking time and the specified kicking time is recorded and calculated to score.
Further, the fluency score is calculated based on the ratio of the highest number of consecutive valid kicks to the number of kicks that should be played in the present training.
Specifically, fluency= (maximum number of continuous effective kicks/number of kicks to be played in this training) ×100%.
According to the embodiment of the disclosure, the following technical effects are achieved:
after the ball kicking moment, the ball kicking frequency, the ball kicking direction and the ball kicking force are led into a training evaluation system, special scores and general scores of football training are obtained, a comprehensive evaluation five-dimensional chart is generated, and relevant personnel can be assisted to judge the condition of football training, and improvement is made accordingly.
It should be noted that, for simplicity of description, the foregoing method embodiments are all described as a series of acts, but it should be understood by those skilled in the art that the present application is not limited by the order of acts described, as some steps may be performed in other orders or concurrently in accordance with the present application. Further, those skilled in the art will also appreciate that the embodiments described in the specification are alternative embodiments, and that the acts and modules referred to are not necessarily required for the present application. In addition, the training evaluation method for football special and the embodiment of the method for processing the motion data during kicking are the same conception, and detailed implementation procedures of the training evaluation method are shown in the embodiment of the method for processing the motion data during kicking, and are not repeated here.
Fig. 7 shows a block diagram of an electronic device according to an embodiment of the present application. Referring to fig. 7, an electronic device 700 includes a processor 701 and a memory 703. The processor 701 is coupled to a memory 703, such as via a bus 702. Optionally, the electronic device 700 may also include a transceiver 704. It should be noted that, in practical applications, the transceiver 704 is not limited to one, and the structure of the electronic device 700 is not limited to the embodiment of the present application.
The processor 701 may be a CPU (Central Processing Unit ), general purpose processor, DSP (Digital Signal Processor, data signal processor), ASIC (Application Specific Integrated Circuit ), FPGA (Field Programmable Gate Array, field programmable gate array) or other programmable logic device, transistor logic device, hardware components, or any combination thereof. Which may implement or perform the various exemplary logic blocks, modules and circuits described in connection with this disclosure. The processor 701 may also be a combination that performs computing functions, such as including one or more microprocessors, a combination of a DSP and a microprocessor, or the like.
Bus 702 may include a path to transfer information between the components. Bus 702 may be a PCI (Peripheral Component Interconnect, peripheral component interconnect Standard) bus or an EISA (Extended Industry Standard Architecture ) bus, or the like. Bus 702 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in fig. 7, but not only one bus or one type of bus.
The Memory 703 may be, but is not limited to, ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, RAM (Random Access Memory ) or other type of dynamic storage device that can store information and instructions, EEPROM (Electrically Erasable Programmable Read Only Memory ), CD-ROM (Compact Disc Read Only Memory, compact disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.
The memory 703 is used for storing application program codes for executing the present application and is controlled by the processor 701 for execution. The processor 701 is configured to execute application code stored in the memory 703 to enable positioning of the map.
Among them, electronic devices include, but are not limited to: mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and the like, and stationary terminals such as digital TVs, desktop computers, and the like. It should be noted that the electronic device shown in fig. 7 is only an example, and should not impose any limitation on the functions and application scope of the embodiments of the present application.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may 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 a computer, the processes or functions described in accordance with embodiments of the present application are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may 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 may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, data subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital versatile disk (digital versatile disc, DVD)), or a semiconductor medium (e.g., solid State Disk (SSD)), etc. It is noted that the computer readable storage medium mentioned in the embodiments of the present application may be a non-volatile storage medium, in other words, may be a non-transitory storage medium.
It should be understood that references herein to "at least one" mean one or more, and "a plurality" means two or more. In the description of the embodiments of the present application, unless otherwise indicated, "/" means or, for example, a/B may represent a or B; "and/or" herein is merely an association relationship describing an association object, and means that three relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist together, and B exists alone. In addition, in order to facilitate the clear description of the technical solution of the embodiments of the present application, in the embodiments of the present application, the words "first", "second", etc. are used to distinguish the same item or similar items having substantially the same function and effect. It will be appreciated by those of skill in the art that the words "first," "second," and the like do not limit the amount and order of execution, and that the words "first," "second," and the like do not necessarily differ.
The above-mentioned exemplary embodiments of the present application are not intended to limit the embodiments of the present application, and any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the embodiments of the present application should be included in the protection scope of the present application.
Claims (10)
1. A method for processing motion data during kicking, comprising:
acquiring n groups of motion data acquired by a sensor in a football training device, wherein each group of motion data comprises quaternions;
importing the quaternion into a 3D engine, and generating a space coordinate point model of each group of motion data based on the 3D engine;
taking a line segment formed by core space points as an axis, taking the minimum average of distances between other space points and the core space points as a radius, and establishing a core data area, wherein the core space points are the space coordinate points with the largest repeated occurrence times in the space coordinate points of the n groups of motion data;
determining an objective function by using a small-batch gradient descent method based on the space coordinate points in the core data area, wherein the objective function is a neural network linear layer with four different data volumes;
and carrying out convolution calculation on the motion data and the objective function, and determining whether a certain moment is a kicking action or not based on a calculation result.
2. The method of claim 1, wherein determining whether a moment is after a kicking action based on the calculation result further comprises:
if the kicking action is performed, recording continuous a space coordinate points after the kicking action is completed;
and calculating a space vector formed by two adjacent space coordinate points, and obtaining the motion direction of the current moment based on the average value of the a-1 space vectors.
3. The method of claim 1, wherein determining whether a moment is after a kicking action based on the calculation result further comprises:
if the kicking action is performed, recording an acceleration value acquired by a sensor at the moment of finishing the kicking action, and multiplying the acceleration value by the mass of the sensor to obtain a force value at the moment of kicking.
4. The method of claim 1, further comprising recording a number of kicks based on each kicking event.
5. The method of claim 1, wherein the quaternion is imported into the 3D engine using the formula:
qV＝q*v*q -1
wherein q is a quaternion produced by the sensor, V is a value of a default positive front vector of the 3D engine, V is a value of a preset vector quaternion converted from the value of the default positive front vector of the 3D engine, and q -1 Representing the inverse of the quaternion of the sensor output, represents a cross-product operation.
6. The method of claim 1, wherein the line segment formed by the core space points is used as an axis, the minimum average distance between other space points and the core space points is used as a radius, and before the core data area is built, the method further comprises:
and eliminating jitter data, wherein the jitter data are space coordinate points of which the difference between the space distance and the average distance in each frame exceeds a preset value.
7. The method of claim 1, wherein calculating the objective function using a small batch gradient descent method based on the spatial coordinate points within the core data region comprises:
acquiring all space coordinate points which remain in a core data area in the nth group of motion data;
selecting 3 continuous space coordinate points, calculating space vectors and space accelerations between two adjacent space coordinate points, and taking the space vectors and the space accelerations as a training set;
m non-repeated training sets and training results are used as input to be led into a deep neural network in a pyrach frame for training calculation in a multiple-time mode;
when the loss value no longer drops, an objective function is determined.
8. The method of claim 7, wherein the objective function is:
wherein J (θ) represents a loss value, m represents the number of input training sets+training results, y i Representing the results of the training set of each input, h θ As an objective function, x i Representing a specific value for each training set entered.
9. The training evaluation method for the football special is characterized by comprising the following steps of:
acquiring motion data acquired by a sensor in a football training device;
converting the motion data into second-order data, wherein the second-order data comprises kicking time, kicking frequency, kicking direction and kicking force;
the second-order data are imported into a training evaluation system to obtain scores of strength control, direction control, accuracy, response time and fluency;
and calculating total score based on the scores of the strength control, the direction control, the accuracy, the response time and the fluency, and generating a comprehensive evaluation five-dimensional graph.
10. The method of claim 9, wherein the scoring of force control, direction control, accuracy, reaction time, and fluency comprises:
the score of the power control is calculated based on the deviation of the force of each kicking and the specified kicking force;
the score of the direction control is calculated based on the deviation of the direction of each kicking and the specified kicking direction;
the scoring of the accuracy is calculated based on the effective kicking number, the number of the kicks to be kicked and the total kicking number, wherein the effective kicking number is the kicking number meeting the power control and direction control standards at the same time;
the score of the reaction time is calculated based on the deviation of the time of each kicking and the specified kicking time;
the fluency score is calculated based on the ratio of the highest continuous effective kicking number to the number of kicks that should be played in the present training.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310605770.8A CN116637342A (en) | 2023-05-25 | 2023-05-25 | Method for processing movement data during kicking and training evaluation method special for football |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310605770.8A CN116637342A (en) | 2023-05-25 | 2023-05-25 | Method for processing movement data during kicking and training evaluation method special for football |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116637342A true CN116637342A (en) | 2023-08-25 |
Family
ID=87614757
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310605770.8A Pending CN116637342A (en) | 2023-05-25 | 2023-05-25 | Method for processing movement data during kicking and training evaluation method special for football |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116637342A (en) |
-
2023
- 2023-05-25 CN CN202310605770.8A patent/CN116637342A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Volleyball skill assessment using a single wearable micro inertial measurement unit at wrist | |
| Blank et al. | Sensor-based stroke detection and stroke type classification in table tennis | |
| AU2017331639B2 (en) | A system and method to analyze and improve sports performance using monitoring devices | |
| Velloso et al. | Qualitative activity recognition of weight lifting exercises | |
| US11210834B1 (en) | Article of clothing facilitating capture of motions | |
| US11551396B2 (en) | Techniques for establishing biomechanical model through motion capture | |
| Yang et al. | TennisMaster: An IMU-based online serve performance evaluation system | |
| Malawski | Depth versus inertial sensors in real-time sports analysis: A case study on fencing | |
| CN106730723A (en) | A kind of soldier based on Intelligent worn device pang ball training method and system | |
| Taha et al. | Preliminary investigation of an innovative digital motion analysis device for badminton athlete performance evaluation | |
| Chen et al. | Development of an upper limb rehabilitation system using inertial movement units and kinect device | |
| Taghavi et al. | Tennis stroke detection using inertial data of a smartwatch | |
| Rigozzi et al. | Application of wearable technologies for player motion analysis in racket sports: A systematic review | |
| Šeketa et al. | Real-time evaluation of repetitive physical exercise using orientation estimation from inertial and magnetic sensors | |
| Shan et al. | Investigation of sensor-based quantitative model for badminton skill analysis and assessment | |
| CN116637342A (en) | Method for processing movement data during kicking and training evaluation method special for football | |
| Cunha et al. | Real-time data movements acquisition of taekwondo athletes: first insights | |
| Viyanon et al. | Swingpong: analysis and suggestion based on motion data from mobile sensors for table tennis strokes using decision tree | |
| CN114247081A (en) | Method and system for determining motion energy consumption of intelligent stone lock | |
| Rusydi et al. | Pattern recognition of overhead forehand and backhand in badminton based on the sign of local euler angle | |
| CN108392805A (en) | A kind of golf swing action-analysing method and intelligent terminal | |
| Wang et al. | Analysis of movement effectiveness in badminton strokes with accelerometers | |
| Kurnia et al. | Analysis of the smash and dropshot movement patterns in badminton sports using Pearson correlation | |
| WO2015146155A1 (en) | Swing data compression method, swing data compression device, swing analysis device, and swing data compression program | |
| WO2023182726A1 (en) | Electronic device and method for segmentation of movement repetitions and extraction of performance metrics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |