Curling motion state evaluation method and system based on three-dimensional image dynamic analysis

Abstract

The invention discloses a curling motion state evaluation method and system based on three-dimensional image dynamic analysis, and belongs to the technical field of image analysis. Obtaining curling stone throwing point data, and predicting a curling stone static point according to the curling stone prediction model; acquiring an actual static point of the curling stone, and analyzing the deviation condition of the actual static point and the predicted static point; according to the method, the static point of the curling stone is predicted by training the neural network, and then the positioning of the abnormal point of the curling stone is analyzed according to the deviation condition of the curling stone, so that the abnormal curling stone can be found timely and accurately, and the accuracy of the abnormal positioning of the curling stone is improved. Therefore, the influence of abnormal curling on competition or training is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of image analysis, and in particular relates to a curling motion state evaluation method and system based on three-dimensional image dynamic analysis. Background Art

[0002] Curling, also known as curling or curling, is a throwing competition on ice performed by teams. Using curling stones with abnormalities will affect the level of competition, fairness of competition, and the training efficiency of athletes. Therefore, it is extremely important to detect whether curling stones have abnormalities.

[0003] Before a game or training session, the curling stone will be inspected, usually by inspecting its appearance or using a scanner to conduct a comprehensive inspection of the curling stone. The former is prone to missing or ignoring abnormalities, while the latter has low detection efficiency. Therefore, there is a lack of predicting the still point of the curling stone first, and then analyzing the abnormal point of the curling stone according to the deviation of the curling stone, so that the abnormal coordinates of the curling stone cannot be accurately located, resulting in low detection efficiency;

[0004] In order to solve the above problems, the present invention proposes a curling motion state evaluation method and system based on three-dimensional image dynamic analysis. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a curling motion state evaluation method and system based on three-dimensional image dynamic analysis. The present invention collects curling throwing point data; obtains curling throwing point data, and predicts the static point of curling according to a curling prediction model; obtains the actual static point of curling, and analyzes the deviation between the actual static point and the predicted static point; obtains throwing point data and deviation data, and obtains the location of curling abnormality according to a curling abnormality positioning model. The present invention predicts the static point of curling by training a neural network, and then analyzes the location of the curling abnormality point according to the deviation of the curling, which is conducive to timely and rapid discovery of abnormal curling to reduce the impact of abnormal curling on competition or training.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The curling motion state evaluation method based on three-dimensional image dynamic analysis includes the following specific steps:

[0008] Step 1: Collect curling throwing point data;

[0009] Step 2: Obtain the curling throwing point data, and predict the curling still point according to the curling prediction model;

[0010] Step 3, obtaining the actual still point of the curling stone, and analyzing the deviation between the actual still point and the predicted still point;

[0011] Step 4: Obtain throwing point data and deviation data, and obtain the location of curling anomaly according to the curling anomaly location model.

[0012] Preferably, the step 1 includes the following specific steps:

[0013] Step 11: Collecting curling throwing point data, wherein the throwing point data includes rotation speed and initial speed;

[0014] Step 1 and 2: Use a camera to record the curling throwing process, automatically identify the feature points on the curling stone through an image recognition algorithm, and establish a three-dimensional motion trajectory of the curling stone based on the feature point tracking;

[0015] Step 13: Set the coordinates of the curling stone leaving the hand before throwing as Mark Point 1, mark the moment when the curling stone leaves the hand, obtain the coordinates of the curling stone on the three-dimensional motion trajectory at the next set moment, set it as Mark Point 2, and calculate the linear velocity of Mark Point 1 according to the distance and time interval between Mark Point 1 and Mark Point 2, where the linear velocity calculation formula is: Wherein, d is the distance between the marking point 1 and the marking point 2, and t is the time interval between the release moment and the next set moment; the distance from the marking point 1 to the center point of the curling stone is obtained to calculate the angular velocity of the marking point 1, where the calculation formula of the angular velocity is: Where r is the distance from the mark point 1 to the center of the curling stone; the rotation speed of the curling stone at the moment of leaving the hand is obtained according to the angular velocity;

[0016] Step 14: Calculate the initial velocity of the curling stone according to the displacement and time interval from the marked point 1 to the marked point 2, wherein the calculation formula of the initial velocity is: Among them, Δx is the displacement from marker point 1 to marker point 2.

[0017] Preferably, the step 2 comprises the following specific steps:

[0018] Step 21, obtain several groups of historical throwing data of curling, wherein the historical throwing data include historical hand-off coordinates, historical rotation speed, historical initial speed and historical actual stationary point, construct a curling motion neural network model whose input is the hand-off coordinates, rotation speed and initial speed of the curling and whose output is the stationary point of the curling, divide the historical throwing data of the curling into 80% parameter training set and 20% parameter test set, input 80% parameter training set into the curling motion neural network model for training to obtain the curling motion initial neural network model, input 20% parameter test set into the curling motion initial neural network model for testing, and output the optimal curling motion initial neural network model with the highest accuracy in judging the stationary point of the curling as the curling motion neural network model;

[0019] Step 22: The output strategy formula of a specific neuron in the curling sport neural network model is:

[0020]

[0021] in, is the stationary point abscissa output of the n+1 layer N-term neurons, is the stationary point ordinate output of the n+1 layer N-item neuron, σ() is the Sigmoid activation function, is the connection weight between the nth layer neuron k and the n+1th layer N neurons, K is the number of neurons in the nth layer, is the hand-off horizontal coordinate input of neuron k in the nth layer, is the hand-off ordinate input of neuron k in the nth layer, is the speed input of neuron k in the nth layer, is the initial velocity input of neuron k in the nth layer, is the bias of the linear relationship between the nth layer neuron k and the n+1th layer N-item neuron;

[0022] Step 2: Input the curling stone's hand-off coordinates, rotation speed and initial speed into the curling motion neural network model, and output the predicted stationary point coordinates.

[0023] Preferably, the step three includes the following specific steps:

[0024] Step 31: Obtain the coordinates of the actual stationary point and the predicted stationary point of the curling stone;

[0025] Step 32: Calculate the offset displacement of the curling stone according to the offset displacement calculation formula, the offset displacement calculation formula is: Among them, x sj is the horizontal coordinate of the actual stationary point, x yc Predict the horizontal coordinate of the stationary point, y sj is the ordinate of the actual stationary point, y yc Predict the ordinate of the stationary point;

[0026] Step 3: Calculate the deviation angle of the curling stone according to the deviation angle calculation formula, the deviation angle calculation formula is: Among them, arctan is the inverse tangent function.

[0027] Preferably, the step 4 includes the following specific steps:

[0028] Step 41. Obtain several groups of historical offset data, historical rotation speed and historical initial speed of curling, wherein the historical offset data includes historical offset displacement and historical offset angle, construct a curling abnormality positioning neural network model whose input is the rotation speed, initial speed, offset displacement and offset angle of curling and whose output is the curling abnormality positioning point, divide the historical offset data, historical rotation speed and historical initial speed of curling into 80% parameter training set and 20% parameter test set, input 80% parameter training set into the curling abnormality positioning neural network model for training to obtain the curling abnormality positioning initial neural network model, input 20% parameter test set into the curling abnormality positioning initial neural network model for testing, and output the optimal curling abnormality positioning initial neural network model with the highest accuracy in locating the curling abnormality point as the curling abnormality positioning neural network model;

[0029] Step 42: The output strategy formula of the specific neuron in the curling abnormality positioning neural network model is:

[0030]

[0031] in, is the horizontal coordinate output of the curling abnormal point of the i+1 layer I-item neuron, is the ordinate output of the curling anomaly point of the i+1 layer I-term neuron, is the connection weight between the i-th layer neuron p and the i+1-th layer I-item neuron, P is the number of neurons in the i-th layer, is the offset displacement input of the i-th layer neuron p, is the offset angle input of the i-th layer neuron p, is the speed input of the neuron p in the i-th layer, V i p is the initial velocity input of the neuron p in the i-th layer, is the bias of the linear relationship between the neuron p in the i-th layer and the neuron I in the i+1 layer;

[0032] Step 43: Input the offset displacement, offset angle, rotation speed and initial speed of the curling stone into the curling stone abnormality positioning neural network model, and output the coordinates of the curling stone abnormal point.

[0033] A curling motion state evaluation system based on three-dimensional image dynamic analysis is used to implement a curling motion state evaluation method based on three-dimensional image dynamic analysis, including a data acquisition module, an image analysis module, a coordinate prediction module, a model building module, a deviation analysis module, an abnormality positioning module and a control module;

[0034] Preferably, the data acquisition module is used to collect data on the throwing point of the curling stone, the actual stationary point of the curling stone, and deviation data;

[0035] The image analysis module is used to analyze the curling throwing process recorded by the camera, pre-process the image, identify the characteristic points of the curling, and obtain the movement trajectory of the curling;

[0036] The model building module is used to obtain several groups of historical data of curling, divide the historical data of curling into 80% parameter training sets and 20% parameter test sets, input 80% parameter training sets into the neural network model for training, and input 20% parameter test sets into the neural network model for testing, thereby building a neural network model;

[0037] The coordinate prediction module is used to obtain the curling throwing point data and import it into the curling motion neural network model to predict the static point of the curling stone;

[0038] The deviation analysis module is used to obtain the actual static point of the curling stone and analyze the deviation displacement and deviation angle between the actual static point and the predicted static point;

[0039] The anomaly location module is used to obtain the throwing point data and deviation data and import them into the curling anomaly location neural network model to obtain the location of the curling anomaly;

[0040] Preferably, the control module is used to control the operation of the data acquisition module, the image analysis module, the coordinate prediction module, the model building module, the deviation analysis module and the anomaly positioning module.

[0041] An electronic device includes: a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes the above-mentioned curling motion state evaluation method based on three-dimensional image dynamic analysis by calling the computer program stored in the memory.

[0042] A computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the curling motion state evaluation method based on three-dimensional image dynamic analysis.

[0043] The beneficial effects of the present invention are as follows: the present invention collects curling throwing point data; obtains curling throwing point data, and predicts the stationary point of curling according to a curling prediction model; obtains the actual stationary point of curling, and analyzes the deviation between the actual stationary point and the predicted stationary point; obtains throwing point data and deviation data, and obtains the location of curling abnormalities according to a curling abnormality positioning model. The present invention predicts the stationary point of curling by training a neural network, and then analyzes the location of the curling abnormality point according to the deviation of the curling, which is conducive to timely and rapid discovery of abnormal curling to reduce the impact of abnormal curling on competition or training. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 It is a schematic flow chart of a curling motion state evaluation method based on three-dimensional image dynamic analysis of the present invention;

[0046] Figure 2 It is a flow chart of step 1 of the curling motion state evaluation method based on three-dimensional image dynamic analysis of the present invention;

[0047] Figure 3 A simulated diagram of a curling competition venue of the present invention;

[0048] Figure 4 A schematic diagram showing the marking of the center point of the curling stone of the present invention;

[0049] Figure 5 It is a schematic diagram of the overall framework of the curling motion state evaluation system based on three-dimensional image dynamic analysis of the present invention. DETAILED DESCRIPTION

[0050] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the present invention will be more comprehensive and complete and fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0051] Example 1

[0052] See also Figure 1-Figure 4 The present invention provides an embodiment: a curling motion state evaluation method based on three-dimensional image dynamic analysis, which includes the following specific steps:

[0053] Step 1: Collect curling throwing point data;

[0054] In this embodiment, step one includes the following specific steps:

[0055] Step 11: Collecting curling throwing point data, the throwing point data including rotation speed and initial speed;

[0056] Step 1 and 2: Use a camera to record the curling throwing process, automatically identify the feature points on the curling stone through an image recognition algorithm, and establish a three-dimensional motion trajectory of the curling stone based on the feature point tracking;

[0057] It should be noted that the video data recorded by the high-speed camera is imported using video processing software. Due to the camera sensor or environmental factors, the video data may contain noise. A filter is first used to remove the noise and improve the image quality. Then the curling stone and the ice surface are separated from the image (this can be achieved by setting a threshold, using a color segmentation or background subtraction algorithm). The image features are enhanced by adjusting the contrast and brightness to make the feature points more obvious. An image recognition algorithm is used to automatically identify the feature points on the curling stone (such as edge detection, corner point detection or recognition of a specific color / shape). A tracking algorithm (such as a Kalman filter, an optical flow method, etc.) is used to track the feature points in consecutive frames. The positions of the tracked feature points in each frame are connected into a trajectory to obtain the motion trajectory of the feature points on the two-dimensional image plane.

[0058] Use a calibration board (such as a chessboard) to calibrate the camera, obtain the camera's internal parameters (focal length, principal point, etc.) and external parameters (position and direction) through the calibration process, and convert the two-dimensional image coordinates into accurate three-dimensional space coordinates according to the calibration results; compare images from different camera perspectives, determine the positions of feature points in three-dimensional space, and use matching algorithms (such as block matching, feature matching, or deep learning) to find the corresponding feature points from different perspectives; calculate the three-dimensional coordinates of the feature points using the triangulation principle based on the camera's calibration parameters and stereo matching results; connect the calculated three-dimensional coordinates of the feature points into a trajectory to reconstruct the motion trajectory of the curling stone, and use a smoothing algorithm (such as spline interpolation) to optimize the representation of the trajectory and reduce noise and errors that may occur during data acquisition.

[0059] Step 13: Set the coordinates of the curling stone leaving the hand before throwing as Mark Point 1, mark the moment when the curling stone leaves the hand, obtain the coordinates of the curling stone on the three-dimensional motion trajectory at the next set moment, set it as Mark Point 2, and calculate the linear velocity of Mark Point 1 according to the distance and time interval between Mark Point 1 and Mark Point 2, where the linear velocity calculation formula is: Wherein, d is the distance between the marking point 1 and the marking point 2, and t is the time interval between the release moment and the next set moment; the distance from the marking point 1 to the center point of the curling stone is obtained to calculate the angular velocity of the marking point 1, where the calculation formula of the angular velocity is: Where r is the distance from the mark point 1 to the center of the curling stone; the rotation speed of the curling stone at the moment of leaving the hand is obtained according to the angular velocity;

[0060] It should be noted that multiple synchronized high-speed cameras are used to record the process of throwing and sliding the curling stone, ensuring that the camera covers the entire path of the curling stone and can capture the rotation of the curling stone from different angles; the collected video data is preprocessed, including denoising, background removal, etc.; an image recognition algorithm is used to identify and track feature points on the curling stone, and the two-dimensional image coordinates are converted into three-dimensional space coordinates using the camera's calibration information. By tracking the marked points, the three-dimensional motion trajectory of the curling stone is established; the center point of the curling stone is marked as the center of rotation. The center point of the curling stone is obtained by obtaining a top view of the curling stone, drawing a circumscribed square of the curling stone, and obtaining the intersection of the two diagonals of the circumscribed square as the center point of the curling stone, as shown in the following figure. Figure 4 As shown;

[0061] Example: The camera tracks the curling stone's movement of 0.2 meters from the hand coordinates within 1 second. The distance from the hand coordinates to the rotation center is 0.1 meters. Then the linear velocity v = 0.2 meters per second, the angular velocity ω = 0.2 meters per second / 0.1 meters = 2 radians per second, converted to rotation speed, RPM = 2 / (2π)*60≈19.1.

[0062] Step 14: Calculate the initial velocity of the curling stone according to the displacement and time interval from the marked point 1 to the marked point 2, wherein the calculation formula of the initial velocity is: Among them, Δx is the displacement from marker point 1 to marker point 2.

[0063] It should be noted that the coordinates of the curling stone leaving the hand before throwing are determined, the moment when the curling stone leaves the hand is marked, that is, the moment when the curling stone starts to slide, and a short motion trajectory after the curling stone leaves the hand is selected for analysis (the acceleration of the curling stone can be considered constant in this trajectory, or the friction and air resistance of the curling stone are small and can be ignored);

[0064] Example: From the video analysis, we know that the curling stone moves 1 meter in the first 0.5 seconds after leaving the hand, so the initial speed is: v0 = 1 meter / 0.5 seconds = 2 meters / second;

[0065] The initial velocity can also be expressed by the formula Get, where s is the distance the curling stone moves, m is the weight of the curling stone, and f is the friction force on the curling stone (the friction force and air resistance on the curling stone in this trajectory are small and can be ignored);

[0066] The weight of the curling stone affects the sliding distance. The heavier the curling stone, the farther it slides. Normally, the weight of the curling stone (including the handle and the bolt) is 19.96 kg. The curling stone is made of natural Scottish granite without mica, with a circumference of about 91.44 cm and a height (from the bottom to the top) of 11.43 cm. The standard diameter of the curling stone is 29 cm and the thickness is 11.5 cm.

[0067] The weight of the stone can be obtained by measuring it directly using an accurate scale (such as a digital or mechanical scale), or by finding its weight on the bottom or label of the stone (the manufacturer will indicate the weight of the stone to ensure that it meets competition standards);

[0068] It should be noted that if Figure 3 As shown, the coordinate system of this embodiment is established by taking the intersection of the front throw line and the center line of the curling competition venue as the coordinate origin, taking the front throw line as the horizontal coordinate, and taking the center line as the vertical coordinate to construct the coordinate system.

[0069] Step 2: Obtain the curling throwing point data, and predict the curling still point according to the curling prediction model;

[0070] In this embodiment, step 2 includes the following specific steps:

[0071] Step 21. Obtain several groups of historical throwing data of curling stones, the historical throwing data including historical hand-off coordinates, historical rotation speeds, historical initial speeds and historical actual stationary points, construct a curling motion neural network model whose input is the hand-off coordinates, rotation speed and initial speed of the curling stone and whose output is the stationary point of the curling stone, divide the historical throwing data of the curling stone into an 80% parameter training set and a 20% parameter test set, input the 80% parameter training set into the curling motion neural network model for training to obtain the curling motion initial neural network model, input the 20% parameter test set into the curling motion initial neural network model for testing, and output the optimal curling motion initial neural network model with the highest accuracy in judging the stationary point of the curling stone as the curling motion neural network model;

[0072] Step 2: The output strategy formula of a specific neuron in the curling neural network model is:

[0073]

[0074] in, is the stationary point abscissa output of the n+1 layer N-term neurons, is the stationary point ordinate output of the n+1 layer N-item neuron, σ() is the Sigmoid activation function, is the connection weight between the nth layer neuron k and the n+1th layer N neurons, K is the number of neurons in the nth layer, is the hand-off horizontal coordinate input of neuron k in the nth layer, is the hand-off ordinate input of neuron k in the nth layer, is the speed input of neuron k in the nth layer, is the initial velocity input of neuron k in the nth layer, is the bias of the linear relationship between the nth layer neuron k and the n+1th layer N-item neuron;

[0075] Step 2: Input the curling stone's hand-off coordinates, rotation speed and initial speed into the curling motion neural network model, and output the predicted stationary point coordinates.

[0076] Step 3, obtaining the actual still point of the curling stone, and analyzing the deviation between the actual still point and the predicted still point;

[0077] In this embodiment, step three includes the following specific steps:

[0078] Step 31: Obtain the coordinates of the actual stationary point and the predicted stationary point of the curling stone;

[0079] Step 32: Calculate the offset displacement of the curling stone according to the offset displacement calculation formula. The offset displacement calculation formula is: Among them, x sj is the horizontal coordinate of the actual stationary point, x yc Predict the horizontal coordinate of the stationary point, y sj is the ordinate of the actual stationary point, y yc Predict the ordinate of the stationary point;

[0080] Step 3: Calculate the deviation angle of the curling stone according to the deviation angle calculation formula. The deviation angle calculation formula is: Among them, arctan is the inverse tangent function.

[0081] Step 4: Obtain throwing point data and deviation data, and obtain the location of curling abnormality according to the curling abnormality location model;

[0082] In this embodiment, step 4 includes the following specific steps:

[0083] Step 41. Obtain several groups of historical offset data, historical rotation speed and historical initial speed of curling, the historical offset data including historical offset displacement and historical offset angle, construct a curling abnormality positioning neural network model whose input is the rotation speed, initial speed, offset displacement and offset angle of curling and whose output is the abnormal positioning point of curling, divide the historical offset data, historical rotation speed and historical initial speed of curling into 80% parameter training set and 20% parameter test set, input 80% parameter training set into the curling abnormality positioning neural network model for training to obtain the curling abnormality positioning initial neural network model, input 20% parameter test set into the curling abnormality positioning initial neural network model for testing, output the optimal curling abnormality positioning initial neural network model with the highest accuracy in locating the abnormal point of curling as the curling abnormality positioning neural network model;

[0084] Step 42: The output strategy formula of a specific neuron in the curling anomaly location neural network model is:

[0085]

[0086] in, is the horizontal coordinate output of the curling abnormal point of the i+1 layer I-item neuron, is the ordinate output of the curling anomaly point of the i+1 layer I-term neuron, is the connection weight between the i-th layer neuron p and the i+1-th layer I-item neuron, P is the number of neurons in the i-th layer, is the offset displacement input of the i-th layer neuron p, is the offset angle input of the neuron p in the i-th layer, is the speed input of the neuron p in the i-th layer, V i p is the initial velocity input of the neuron p in the i-th layer, is the bias of the linear relationship between the neuron p in the i-th layer and the neuron I in the i+1 layer;

[0087] Step 43: Input the offset displacement, offset angle, rotation speed and initial speed of the curling stone into the curling stone abnormality positioning neural network model, and output the coordinates of the curling stone abnormal point.

[0088] Example 2

[0089] See also Figure 5 , a curling motion state evaluation system based on three-dimensional image dynamic analysis, used to implement a curling motion state evaluation method based on three-dimensional image dynamic analysis, including a data acquisition module, an image analysis module, a coordinate prediction module, a model building module, a deviation analysis module, an abnormality positioning module and a control module;

[0090] In this embodiment, the data collection module is used to collect data on the throwing point of the curling stone, the actual stationary point of the curling stone, and deviation data;

[0091] The image analysis module is used to analyze the curling throwing process recorded by the camera, pre-process the image, identify the characteristic points of the curling, and obtain the movement trajectory of the curling;

[0092] The model building module is used to obtain several groups of historical data of curling, divide the historical data of curling into 80% parameter training sets and 20% parameter test sets, input 80% parameter training sets into the neural network model for training, and input 20% parameter test sets into the neural network model for testing, thereby building a neural network model;

[0093] The coordinate prediction module is used to obtain the data of the curling throwing point and import it into the curling motion neural network model to predict the static point of the curling stone;

[0094] The deviation analysis module is used to obtain the actual static point of the curling stone and analyze the deviation displacement and deviation angle between the actual static point and the predicted static point;

[0095] The anomaly location module is used to obtain the throwing point data and deviation data and import them into the curling anomaly location neural network model to obtain the location of the curling anomaly;

[0096] In this embodiment, the control module is used to control the operation of the data acquisition module, the image analysis module, the coordinate prediction module, the model building module, the deviation analysis module and the anomaly positioning module.

[0097] Example 3

[0098] This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes the above-mentioned curling motion state evaluation method based on three-dimensional image dynamic analysis by calling the computer program stored in the memory.

[0099] The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) and one or more memories, wherein the memory stores at least one computer program, and the computer program is loaded and executed by the processor to implement the curling motion state evaluation method based on three-dimensional image dynamic analysis provided by the above method embodiment. The electronic device may also include other components for realizing the functions of the device, for example, the electronic device may also have components such as a wired or wireless network interface and an input and output interface to input and output data. This embodiment will not be described in detail here.

[0100] Example 4

[0101] This embodiment provides a computer-readable storage medium storing instructions, which, when a computer program is executed on a computer device, enables the computer device to execute the above-mentioned curling motion state evaluation method based on three-dimensional image dynamic analysis.

[0102] For example, the computer readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, etc.

[0103] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0104] In the description of this specification, the descriptions with reference to the terms "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example, and the specific features, structures, materials or characteristics described may be combined in an appropriate manner in any one or more embodiments or examples.

Claims

1. A curling motion state evaluation method based on three-dimensional image dynamic analysis, characterized in that: The specific steps include: Step 1: Collect curling throwing point data; Step 2: Obtain the curling throwing point data, and predict the curling still point according to the curling prediction model; Step 3, obtaining the actual still point of the curling stone, and analyzing the deviation between the actual still point and the predicted still point; Step 4: Obtain throwing point data and deviation data, and obtain the location of curling anomaly according to the curling anomaly location model.

2. The curling motion state evaluation method based on three-dimensional image dynamic analysis according to claim 1, characterized in that: The step 2 includes the following specific steps: Step 21, obtain several groups of historical throwing data of curling, wherein the historical throwing data include historical hand-off coordinates, historical rotation speed, historical initial speed and historical actual stationary point, construct a curling motion neural network model whose input is the hand-off coordinates, rotation speed and initial speed of the curling and whose output is the stationary point of the curling, divide the historical throwing data of the curling into 80% parameter training set and 20% parameter test set, input 80% parameter training set into the curling motion neural network model for training to obtain the curling motion initial neural network model, input 20% parameter test set into the curling motion initial neural network model for testing, and output the optimal curling motion initial neural network model with the highest accuracy in judging the stationary point of the curling as the curling motion neural network model; Step 22: The output strategy formula of a specific neuron in the curling sport neural network model is: in, is the stationary point abscissa output of the n+1 layer N-term neurons, is the stationary point ordinate output of the n+1 layer N-item neuron, σ() is the Sigmoid activation function, is the connection weight between the nth layer neuron k and the n+1th layer N neurons, K is the number of neurons in the nth layer, is the hand-off horizontal coordinate input of neuron k in the nth layer, is the hand-off ordinate input of neuron k in the nth layer, is the speed input of neuron k in the nth layer, is the initial velocity input of neuron k in the nth layer, is the bias of the linear relationship between the nth layer neuron k and the n+1th layer N-term neurons; Step 2: Input the curling stone's hand-off coordinates, rotation speed and initial speed into the curling motion neural network model, and output the predicted stationary point coordinates.

3. The curling motion state evaluation method based on three-dimensional image dynamic analysis according to claim 2, characterized in that: The step three includes the following specific steps: Step 31: Obtain the coordinates of the actual stationary point and the predicted stationary point of the curling stone; Step 32: Calculate the offset displacement of the curling stone according to the offset displacement calculation formula, the offset displacement calculation formula is: Among them, x sj is the horizontal coordinate of the actual stationary point, x yc Predict the horizontal coordinate of the stationary point, y sj is the ordinate of the actual stationary point, y yc Predict the ordinate of the stationary point; Step 3: Calculate the deviation angle of the curling stone according to the deviation angle calculation formula, the deviation angle calculation formula is: Among them, arctan is the inverse tangent function.

4. The curling motion state evaluation method based on three-dimensional image dynamic analysis according to claim 3, characterized in that: The step 4 includes the following specific steps: Step 41. Obtain several groups of historical offset data, historical rotation speed and historical initial speed of curling, wherein the historical offset data includes historical offset displacement and historical offset angle, construct a curling abnormality positioning neural network model whose input is the rotation speed, initial speed, offset displacement and offset angle of curling and whose output is the curling abnormality positioning point, divide the historical offset data, historical rotation speed and historical initial speed of curling into 80% parameter training set and 20% parameter test set, input 80% parameter training set into the curling abnormality positioning neural network model for training to obtain the curling abnormality positioning initial neural network model, input 20% parameter test set into the curling abnormality positioning initial neural network model for testing, and output the optimal curling abnormality positioning initial neural network model with the highest accuracy in locating the curling abnormality point as the curling abnormality positioning neural network model; Step 42: The output strategy formula of the specific neuron in the curling abnormality positioning neural network model is: in, is the horizontal coordinate output of the curling abnormal point of the i+1 layer I-item neuron, is the ordinate output of the curling anomaly point of the i+1 layer I-term neuron, is the connection weight between the i-th layer neuron p and the i+1-th layer I-item neuron, P is the number of neurons in the i-th layer, is the offset displacement input of the i-th layer neuron p, is the offset angle input of the neuron p in the i-th layer, is the speed input of the neuron p in the i-th layer, V i p is the initial velocity input of the neuron p in the i-th layer, is the bias of the linear relationship between the neuron p in the i-th layer and the neuron I in the i+1 layer; Step 43: Input the offset displacement, offset angle, rotation speed and initial speed of the curling stone into the curling stone abnormality positioning neural network model, and output the coordinates of the curling stone abnormal point.

5. The curling motion state evaluation method based on three-dimensional image dynamic analysis according to claim 4, characterized in that: The step 1 includes the following specific steps: Step 11: Collecting curling throwing point data, wherein the throwing point data includes rotation speed and initial speed; Step 1 and 2: Use a camera to record the curling throwing process, automatically identify the feature points on the curling stone through an image recognition algorithm, and establish a three-dimensional motion trajectory of the curling stone based on the feature point tracking; Step 13: Set the coordinates of the curling stone leaving the hand before throwing as Mark Point 1, mark the moment when the curling stone leaves the hand, obtain the coordinates of the curling stone on the three-dimensional motion trajectory at the next set moment, set it as Mark Point 2, and calculate the linear velocity of Mark Point 1 according to the distance and time interval between Mark Point 1 and Mark Point 2, where the linear velocity calculation formula is: Wherein, d is the distance between the marking point 1 and the marking point 2, and t is the time interval between the release moment and the next set moment; the distance from the marking point 1 to the center point of the curling stone is obtained to calculate the angular velocity of the marking point 1, where the calculation formula of the angular velocity is: Where r is the distance from the mark point 1 to the center of the curling stone; the rotation speed of the curling stone at the moment of leaving the hand is obtained according to the angular velocity; Step 14: Calculate the initial velocity of the curling stone according to the displacement and time interval from the marked point 1 to the marked point 2, wherein the calculation formula of the initial velocity is: Among them, Δx is the displacement from marker point 1 to marker point 2.

6. A curling motion state evaluation system based on three-dimensional image dynamic analysis, used to implement the curling motion state evaluation method based on three-dimensional image dynamic analysis as claimed in any one of claims 1 to 5, characterized in that: It includes a data acquisition module, an image analysis module, a coordinate prediction module, a model building module, a deviation analysis module, an abnormality positioning module and a control module; The data acquisition module is used to collect data on the throwing point of the curling stone, the actual stationary point of the curling stone, and deviation data; The image analysis module is used to analyze the curling throwing process recorded by the camera, pre-process the image, identify the characteristic points of the curling, and obtain the movement trajectory of the curling; The model building module is used to obtain several groups of historical data of curling, divide the historical data of curling into 80% parameter training sets and 20% parameter test sets, input 80% parameter training sets into the neural network model for training, and input 20% parameter test sets into the neural network model for testing, thereby building a neural network model; The coordinate prediction module is used to obtain the curling throwing point data and import it into the curling motion neural network model to predict the static point of the curling stone; The deviation analysis module is used to obtain the actual static point of the curling stone and analyze the deviation displacement and deviation angle between the actual static point and the predicted static point; The anomaly location module is used to obtain the throwing point data and deviation data and import them into the curling anomaly location neural network model to obtain the location of the curling anomaly; The control module is used to control the operation of the data acquisition module, the image analysis module, the coordinate prediction module, the model building module, the deviation analysis module and the abnormality positioning module.

7. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores a computer program that can be called by the processor, and the processor executes the curling motion state evaluation method based on three-dimensional image dynamic analysis as described in any one of claims 1 to 5 by calling the computer program stored in the memory.

8. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are executed on a computer, the computer is caused to execute the curling motion state evaluation method based on three-dimensional image dynamic analysis as described in any one of claims 1 to 5.

Images