A pole winding training assessment system and assessment data acquisition method thereof

Abstract

The present invention relates to the field of sports training assessment systems and methods, and in particular to a pole-circling training assessment system and an assessment data acquisition method thereof. The assessment system comprises an assessment range, a sign pole, a binocular stereoscopic vision device and a background processor. The sign poles are arranged vertically in the assessment range at intervals. The binocular stereoscopic vision device has a binocular camera, which is arranged around the assessment range to cover the assessment range from multiple visual angles. The stereoscopic vision device is connected to the background processor in a wired or wireless manner, and the binocular stereoscopic vision device transmits visual data to the background processor for calculation to obtain assessment data. The assessment system can replace manual intelligent assessment, and the assessment system has a simple structure. The assessment data acquisition method uses stereoscopic vision data to calculate and obtain assessment process data, and the data has high accuracy and reliability.

Description

Technical Field

[0001] The present invention relates to the field of sports training assessment systems and methods, and in particular to an S-shaped pole winding training assessment system and a method for acquiring assessment data thereof. Background Art

[0002] The S-shaped pole-circling assessment is a common dribbling training assessment item in some ball sports, and there are also such assessment items in some physical training. When school students are training in football or basketball, or when professional football or basketball training institutions are assessing, the assessment venue is usually very simple with multiple poles set up. The assessment is mainly conducted by coaches or teachers to manually determine whether the students' S-shaped pole-circling meets the training and assessment standards, that is, it is mainly a manual judgment method. At present, there are also some relatively simple settings to assist manual assessment, such as installing infrared sensors on the poles, or wearing or installing induction tags on the assessment objects, and assisting manual judgment through the induction of infrared sensors, etc. However, this method cannot accurately assess the exact position of the dribbling students and the football or basketball, and is prone to misjudgment or cheating by candidates. Summary of the invention

[0003] The purpose of the present invention is to provide a pole winding training assessment system which can replace manual intelligent assessment, has a simple assessment system structure, and has relatively reliable assessment process data.

[0004] Another object of the present invention is to provide a method for obtaining assessment data of the above-mentioned pole-circling training assessment system, which uses stereoscopic vision data to calculate and obtain assessment process data, and the data has high accuracy and reliability.

[0005] To achieve the above object, the technical solution of the present invention is:

[0006] A pole-circling training and assessment system, characterized in that it includes an assessment range, a marker pole, a binocular stereo vision device and a background processor, wherein the marker poles are arranged vertically in the assessment range at intervals, the binocular stereo vision device has a binocular camera, and multiple cameras are arranged around the assessment range to cover the assessment range from multiple visual angles, the stereo vision device is connected to the background processor via a wired or wireless method, and the binocular stereo vision device transmits visual data to the background processor for calculation to obtain assessment data.

[0007] The binocular stereo vision device includes a shell, the binocular camera arranged on the shell, and a battery and a circuit module arranged on the shell. The battery is connected to the circuit module for power supply. The circuit module includes an antenna, a wireless transmission unit and a control unit. The binocular camera is connected to the wireless transmission unit and the control unit respectively. The data captured by the binocular camera is transmitted to the background processor through the wireless transmission unit. The background processor controls the operation of the binocular camera through the control unit.

[0008] The binocular stereo vision devices are respectively arranged at four diagonal positions in the examination range.

[0009] The binocular stereoscopic vision device is mounted on a lifting and placing frame.

[0010] The S-shaped pole winding training and assessment system also includes a display screen for displaying assessment information.

[0011] The display screen is mounted on a lifting rack and arranged around the assessment range.

[0012] There are five sign poles arranged in a straight line in the middle area of ​​the assessment area.

[0013] The method for obtaining assessment data of the above-mentioned pole-circling training assessment system, wherein the method steps for obtaining process data during the assessment are as follows:

[0014] 1. Each binocular camera takes pictures to obtain the depth image data of the assessment process, which is used to reconstruct the three-dimensional scene of the assessment process;

[0015] Second, each binocular camera processes the depth image data to calculate point cloud data to generate panoramic three-dimensional point cloud data of the scene;

[0016] 3. Each binocular camera transmits the panoramic 3D point cloud data to the background processor;

[0017] Fourth, the background processor receives the panoramic 3D point cloud data from each binocular camera and removes invalid point cloud data to obtain real-time 3D point cloud data;

[0018] 5. The background processor obtains the database standard template point cloud data of the assessment object and performs point cloud registration processing with the real-time three-dimensional point cloud data to obtain the coordinate information data of the assessment object in the real-time three-dimensional point cloud data.

[0019] The conversion formula for processing the depth image data in step 2 to calculate the point cloud data is as follows:

[0020]

[0021] In the formula, f x , f y , cx , c y is the camera internal parameter, which is given by the camera manufacturer or self-calibration; u, v are the pixel coordinate values ​​in the camera coordinate system; X, Y, Z are the actual coordinate values ​​in the world coordinate system.

[0022] The invalid point cloud data in step 4 is eliminated by voxel filtering and / or radius filtering.

[0023] The point cloud registration processing method in step five includes the point cloud registration method of a 4-point consistency set, the point cloud registration method of a normal distribution transformation and / or the point cloud registration method of an iterative nearest point for registration.

[0024] The point cloud registration processing method of the iterative nearest point point cloud registration method is as follows: Assuming that Q i Represents the database standard template point cloud data, P i Represents real-time 3D point cloud data. The alignment of two point clouds is converted to minimize the objective function E of the following formula:

[0025]

[0026] In the formula, R is the rotation matrix, T is the translation matrix, n is the number of point clouds, and i is the sequence number of the point cloud;

[0027] The least square method is used to solve the above formula to obtain R and T, thereby obtaining the coordinate information data of the assessment object in the real-time three-dimensional point cloud data.

[0028] By adopting the above technical solution, the beneficial effects of the present invention are as follows: the S-shaped pole-circling training assessment system of the above structure is based on binocular stereo vision technology, and multiple binocular cameras in different orientations can simultaneously shoot the assessment range area, and can construct the three-dimensional structure in the venue in real time. The multi-view recording is more accurate, and the real-time and accurate position of the assessment judgment object can be recorded very intuitively and accurately. There is no need to install any sensor on the sign pole, and there is no need to wear or install a sensing tag on the assessment judgment object. It can also effectively prevent candidates from cheating and effectively reduce the possibility of misjudgment. The binocular camera can also be used for video recording, which is convenient for later playback and provides testable and traceable data for the assessment process. The assessment data acquisition method of the present invention uses stereo vision data to calculate the assessment process data, and the data has high accuracy and reliability, thereby achieving the purpose of the present invention to replace manual intelligent assessment, and the assessment system structure is simple, and the assessment process data is relatively reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a structural schematic diagram of a pole winding training and assessment system according to the present invention;

[0030] Figure 2It is a structural schematic diagram of a binocular stereoscopic vision device in a pole-circling training and assessment system according to the present invention;

[0031] Figure 3 The invention discloses a flow structure diagram of an assessment data acquisition method of a pole winding training assessment system.

[0032] In the figure:

[0033] Assessment scope area 1; Sign post 2;

[0034] Binocular stereo vision device 3; binocular camera 31; housing 32; battery 33;

[0035] Circuit module 34; antenna 341; display screen 4; background processor 5. DETAILED DESCRIPTION

[0036] In order to further explain the technical solution of the present invention, the present invention is described in detail below through specific embodiments.

[0037] The invention discloses a pole winding training and assessment system, such as Figure 1 As shown, it includes an assessment range area 1, a sign post 2, a binocular stereoscopic vision device 3, a display screen 4 and a background processor 5.

[0038] The assessment range area 1 is an assessment site with an effective assessment range area. As shown in the figure, it is a rectangular area. A plurality of marker poles 2 are arranged vertically at intervals in the assessment range area 1. They can also be arranged specifically according to the content of the assessment items. As shown in the figure, there are 5 marker poles arranged in a straight line along the length direction of the assessment range area 1 in the middle area of ​​the assessment range area 1, which are used for the assessment of S-shaped pole winding training.

[0039] like Figure 2As shown, the binocular stereo vision device 3 has a binocular camera 31, so as to realize the function of binocular stereo vision shooting of the present invention, that is, for three-dimensional panoramic reconstruction of the assessment site, for three-dimensional panoramic reconstruction of the assessment process during the assessment, generating panoramic three-dimensional point cloud data of the site, recording, etc., and it can be connected to the background processor 5 by wired or wireless means to realize data transmission. In this embodiment, the binocular stereo vision device 3 also includes a housing 32, a battery 33 and a circuit module 34 arranged on the housing 32, the binocular camera 31 is arranged on the housing 32, the battery 33 is connected to the circuit module 34 for power supply, and the circuit module 34 can be set according to the corresponding circuit structure according to the system function requirements. For example, in this embodiment, the circuit module 34 includes an antenna 341, a wireless transmission unit (such as a WiFi module) and a control unit (not visible in the figure), the binocular camera 31 is respectively connected to the wireless transmission unit and the control unit, and the data shot by the binocular camera 31 is transmitted to the background processor 5 through the wireless transmission unit, and the background processor 5 can control the shooting work of the binocular camera 31 through the control unit. The binocular stereo vision device 3 is arranged in multiple units around the assessment range 1 to cover the assessment range from multiple visual angles. As shown in the figure, the binocular stereo vision device 3 is arranged in the four diagonal directions of the rectangle of the assessment range 1 respectively. The direction arrangement is reasonable, and the viewing angle can cover the entire binocular view, that is, it can ensure that three-dimensional point cloud data from different angles can be obtained, and the accuracy of the system can be determined.

[0040] After receiving the data from the binocular stereo vision device 3, the background processor 5 calculates and obtains the assessment data (a brief description is given here, and the assessment data acquisition method is described in detail later).

[0041] The display screen 4 is used to display assessment information, such as real-time display of the training results or assessment scores of the assessment subject, and can be placed around the assessment range 1 for easy viewing by the assessment subject. In this embodiment, the binocular stereoscopic vision device 3 and the display screen 4 are mounted on a lifting rack 5, which is convenient for adaptively adjusting the position, height, angle, etc. according to the needs of assessment, and is also convenient for sorting and storage operations.

[0042] The above-mentioned structure is simple in site layout structure for a pole-circling training assessment system, and only requires 4 binocular stereo vision devices 3. The assessment operation is also simple and convenient. Its use effect is as described in the beneficial effect part above, and will not be repeated here. The assessment data acquisition method of a pole-circling training assessment system of the present invention is described in detail below.

[0043] A method for obtaining assessment data of a pole-circling training assessment system, the method steps for obtaining process data during the assessment, such as Figure 3 As shown, including the following:

[0044] 1. Each binocular camera takes pictures to obtain the depth image data of the assessment process, which is used to reconstruct the three-dimensional scene of the assessment process;

[0045] Second, each binocular camera processes the depth image data to calculate the point cloud data to generate panoramic three-dimensional point cloud data of the scene; in this embodiment, the conversion formula for processing the depth image data to calculate the point cloud data is as follows:

[0046]

[0047] In the formula, f x , f y , c x , c y is the camera internal parameter, which is given by the camera manufacturer or self-calibration; u, v are the pixel coordinate values ​​in the camera coordinate system; X, Y, Z are the actual coordinate values ​​in the world coordinate system.

[0048] 3. Each binocular camera transmits the panoramic 3D point cloud data to the background processor;

[0049] 4. The background processor receives the panoramic three-dimensional point cloud data from each binocular camera and performs elimination processing on invalid point cloud data to obtain real-time three-dimensional point cloud data; the elimination processing of invalid point cloud data here can be performed by obtaining the calculated point cloud data of the assessment site through the binocular stereo vision device before the assessment system is used for assessment to calculate the range of numerical parameters required for elimination processing, and then using these range of numerical parameters for calculation processing in the elimination processing of invalid point cloud data to achieve the elimination of invalid point cloud data; the specific elimination processing of invalid point cloud data can be processed by voxel filtering and / or radius filtering, which are well-known technologies in the field and will not be described in detail here. They can be obtained on the Internet or related books, forums, industry journals, etc.

[0050] 5. The background processor obtains the database standard template point cloud data of the assessment object and performs point cloud registration processing with the real-time three-dimensional point cloud data to obtain the coordinate information data of the assessment object in the real-time three-dimensional point cloud data; here, the database standard template point cloud data of the assessment object is a pre-prepared template data database, which is called out for registration when used.

[0051] The above-mentioned point cloud registration processing method can be registered by using a 4-point consistency set (4PCS) point cloud registration method, a normal distribution transform (NDT) point cloud registration method and / or an iterative closest point (ICP) point cloud registration method; as shown in the figure of the present embodiment, the 4-point consistency set (4PCS) point cloud registration method and the normal distribution transform (NDT) point cloud registration method are first used to perform preliminary point cloud registration, that is, a rough registration is performed first, and then a secondary registration is performed by using the iterative closest point (ICP) point cloud registration method, that is, a fine registration is performed again;

[0052] The point cloud registration processing method of the iterative nearest point point cloud registration method is as follows: Assuming that Q i Represents the database standard template point cloud data, P i Represents real-time 3D point cloud data. The alignment of two point clouds is converted to minimize the objective function E of the following formula:

[0053]

[0054] In the formula, R is the rotation matrix, T is the translation matrix, n is the number of point clouds, and i is the sequence number of the point cloud;

[0055] Among them, R and T are the rotation parameters and translation parameters between the database standard template point cloud data and the real-time three-dimensional point cloud data, so that the two point set data satisfy the optimal match under the measurement criterion. The least squares method is used to solve the above formula to obtain R and T, thereby obtaining the coordinate information data of the assessment object in the real-time three-dimensional point cloud data.

[0056] The above-mentioned 4-point consistency set (4PCS) point cloud registration method, normal distribution transform (NDT) point cloud registration method and iterative closest point (ICP) point cloud registration method are well-known technologies in the art and will not be described in detail here. They can be obtained on the Internet or related books, forums, industry journals, etc.

[0057] After the assessment data (i.e., the coordinate information data of the assessment object) is obtained by the above method, the assessment result can be obtained by further intelligent comparison through background processing, or the assessment result can be obtained by the examiner through the coordinate information data of the assessment object, or other methods. This result-deriving part is not the main protection object of this case, and will not be disclosed in detail here. The non-disclosure of this part does not affect the clear understanding of the above technical solution. Through the above assessment system structure and assessment data acquisition method, the assessment system structure is simple, the assessment data is highly accurate, and the assessment process data is highly reliable, thereby achieving the purpose and effect of the present invention.

[0058] The above embodiments and drawings do not limit the product form and style of the present invention. Any appropriate changes or modifications made by ordinary technicians in the relevant technical field should be deemed to be within the patent scope of the present invention.

Claims

1. A method for acquiring assessment data of a pole-circling training assessment system, characterized in that: The pole-circling training assessment system includes an assessment range, a sign pole, a binocular stereoscopic vision device and a background processor. The sign poles are arranged vertically in the assessment range at intervals. The binocular stereoscopic vision device has a binocular camera, which is arranged around the assessment range to cover the assessment range from multiple visual angles. The stereoscopic vision device is connected to the background processor by wire or wirelessly, and the binocular stereoscopic vision device transmits visual data to the background processor for calculation to obtain assessment data. The method steps for obtaining process data during the assessment are as follows:

1. Each binocular camera takes pictures to obtain the depth image data of the assessment process, which is used to reconstruct the three-dimensional scene of the assessment process; Second, each binocular camera processes the depth image data to calculate point cloud data to generate panoramic three-dimensional point cloud data of the scene; 3. Each binocular camera transmits the panoramic 3D point cloud data to the background processor; Fourth, the background processor receives the panoramic 3D point cloud data from each binocular camera and removes invalid point cloud data to obtain real-time 3D point cloud data; 5. The background processor obtains the database standard template point cloud data of the assessment object and performs point cloud registration processing with the real-time three-dimensional point cloud data to obtain the coordinate information data of the assessment object in the real-time three-dimensional point cloud data; The conversion formula for processing the depth image data in step 2 to calculate the point cloud data is as follows: In the formula, , , , It is the camera internal parameter, which is given by the camera manufacturer or self-calibration; , is the pixel coordinate value in the camera coordinate system; , , , is the actual coordinate value in the world coordinate system.

2. The method for acquiring assessment data of a pole-circling training assessment system according to claim 1, characterized in that: The invalid point cloud data in step 4 is eliminated by voxel filtering and / or radius filtering.

3. The method for acquiring assessment data of a pole-circling training assessment system according to any one of claims 1 to 2, characterized in that: The point cloud registration processing method in step five includes the point cloud registration method of a 4-point consistency set, the point cloud registration method of a normal distribution transformation and / or the point cloud registration method of an iterative nearest point for registration.

4. The method for acquiring assessment data of a pole-circling training assessment system according to claim 3, characterized in that: The point cloud registration processing method of the iterative closest point point cloud registration method is as follows: Represents the database standard template point cloud data, The real-time 3D point cloud data is shown. The alignment of two point clouds is converted into the objective function of the following formula Minimum: , where is the rotation matrix, is the translation matrix, is the number of point clouds, is the sequence number of the point cloud; the least squares method is used to solve the above formula, and we get and , thereby obtaining the coordinate information data of the assessment object in the real-time three-dimensional point cloud data.

5. The method for acquiring assessment data of a pole-circling training assessment system according to claim 1, characterized in that: The binocular stereo vision device includes a shell, the binocular camera arranged on the shell, and a battery and a circuit module arranged on the shell. The battery is connected to the circuit module for power supply. The circuit module includes a wireless transmission unit and a control unit. The binocular camera is connected to the wireless transmission unit and the control unit respectively. The data captured by the binocular camera is transmitted to the background processor through the wireless transmission unit. The background processor controls the operation of the binocular camera through the control unit.

6. The method for acquiring assessment data of a pole-circling training assessment system according to claim 1, characterized in that: The binocular stereo vision devices are respectively arranged at four diagonal positions in the examination range.

7. The method for acquiring assessment data of a pole-circling training assessment system according to claim 1, characterized in that: It also includes a display screen for displaying assessment information.

8. The method for acquiring assessment data of a pole-circling training assessment system according to claim 7, characterized in that: The binocular stereoscopic vision device and the display screen are respectively mounted on a lifting and placing frame.

Images