Swing analysis device and program

Abstract

Provided are a swing analysis device and a program that can accurately obtain the movement and movement trajectory of the face surface of a club head using an existing ballistic measuring instrument or the like without installing a photographing device that photographs a swing from above. 【Solution means】A feature amount acquisition unit 21 that acquires a plurality of feature amounts at the time when the club head 41a of the golf club 41 and the golf ball 42 collide when the golf club 41 is swung, and based on the plurality of feature amounts acquired by the feature amount acquisition unit 21, an estimation unit 22 that estimates at least one of the temporal change in the angle of the face surface of the club head 41a and the movement trajectory of the club head 41a.

Description

Technical Field

[0001] The present invention relates to a swing analysis device and a program for analyzing the temporal changes and movement trajectories of the face surface of a club head during the impact of a golf club and the swing before and after the impact.

Background Art

[0002] Conventionally, various devices have been proposed for photographing a golfer's swing and providing the photographed information. By providing the photographed information of the swing to the golfer, it is possible to promote the improvement of golf and to promote the selection of a golf club suitable for the golfer.

[0003] As the above-described device, for example, Patent Document 1 proposes a device that photographs a golf club and a golf ball from above when the golf club is swung, detects the face surface of the club head included in each photographed image, and displays a line segment representing the plurality of detected face surfaces. By displaying a line segment representing the face surface in this way, it is possible to recognize the opening of the face surface during the impact and the swing before and after the impact.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] Here, as important elements in a golfer's swing, there are the movement of the face surface of the club head and the movement trajectory of the club head during the impact and the swings before and after the impact. By recognizing the movement of the face surface of the club head and the movement trajectory during the impact and the swings before and after the impact, the characteristics of the golfer's swing can be grasped, and a golf club can be selected according to the characteristics.

[0006] Even with the device described in Patent Document 1 mentioned above, it is possible to display a line segment representing the face surface of the club head or display the movement trajectory, but a photographing device for photographing the swing from above is required, the device becomes large, and it may be difficult to install the photographing device for photographing from above depending on the installation environment. Also, new construction work and costs are involved.

[0007] In view of the above problems, an object of the present invention is to provide a swing analysis device and a program capable of accurately obtaining the movement of the face surface of the club head and the movement trajectory using an existing ballistics measuring instrument or the like without installing a photographing device for photographing the swing from above.

Means for Solving the Problems

[0008] The swing analysis device of the present invention includes a feature amount acquisition unit that acquires a plurality of feature amounts at the time when the club head of the golf club collides with the golf ball when the golf club is swung, and an estimation unit that estimates at least one of the temporal change in the angle of the face surface of the club head and the movement trajectory of the club head based on the plurality of feature amounts acquired by the feature amount acquisition unit.

Effects of the Invention

[0009] According to the swing analysis device of the present invention, when a golf club is swung, a plurality of feature amounts at the time when the club head of the golf club collides with the golf ball are acquired, and based on the plurality of feature amounts, at least one of the temporal change in the angle of the face surface of the club head and the movement locus of the club head is estimated. Therefore, by acquiring the above-described feature amounts using an existing ballistic measuring instrument or the like, it is possible to accurately obtain the movement and movement locus of the face surface of the club head without installing a photographing device that photographs the swing from above.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Embodiments for Carrying Out the Invention

[0011] Hereinafter, a golf impact analysis system using an embodiment of the swing analysis device of the present invention will be described in detail with reference to the drawings. The golf impact analysis system of this embodiment is characterized by a method for estimating the temporal change in the angle of the face surface of the club head and the movement trajectory of the club head when the golf club is swung. First, the entire golf impact analysis system will be described. FIG. 1 is a schematic configuration diagram of the golf impact analysis system 1 of this embodiment.

[0012] As shown in FIG. 1, the golf impact analysis system 1 of this embodiment includes a trajectory measurement device 10, a control device 20, and a projection device 30. The trajectory measurement device 10 and the control device 20, and the control device 20 and the projection device 30 are communicably connected by wire or wirelessly, and are configured to be able to exchange various signals.

[0013] The trajectory measurement device 10 measures eight characteristic quantities of the golf ball 42, namely, the speed, launch direction, side spin, head speed of the club head 41a, club path, face-to-target, face-to-path, and impact point, at the time when the golfer 40 swings the golf club 41 and hits the golf ball 42 (impact time). In addition to the above eight characteristic quantities, the trajectory measurement device 10 can acquire about a dozen measurement items. However, in this embodiment, since they are not used in the estimation process of the face surface and trajectory information described later, the description thereof is omitted.

[0014] The speed of the golf ball 42 is the speed of the golf ball 42 measured immediately after the club head 41a impacts the golf ball 42. The launch direction is the direction in which the golf ball 42 flies out immediately after the club head 41a impacts the golf ball 42.

[0015] The side spin is the number of revolutions of the lateral rotation of the golf ball 42 immediately after the club head 41a impacts the golf ball 42.

[0016] The head speed of the club head 41a is the moving speed of the club head 41a at the time when the club head 41a impacts the golf ball 42.

[0017] The club path is the trajectory of the club head 41a immediately before the club head 41a and the golf ball 42 meet impact.

[0018] Face-to-target is the orientation of the face of the club head 41a with respect to the target line. The target line is a straight line intersecting the golf ball 42 as shown in FIG. 1, for example.

[0019] Face-to-path is the degree of opening and closing of the face of the club head 41a with respect to the club path.

[0020] The impact point is the position where the face of the club head 41a and the golf ball 42 contacted when the club head 41a and the golf ball 42 impacted.

[0021] As the trajectory measurement device 10, for example, GCQuad (manufactured by Foresight Sports) can be used. GCQuad can measure the above eight feature quantities by photographing the moment of impact between the club head of the golf club 41 and the golf ball with four cameras. However, the trajectory measurement device 10 is not limited to GCQuad, and other existing devices can be used.

[0022] Also, instead of measuring the above eight feature quantities with one trajectory measurement device 10, the eight feature quantities may be measured using a plurality of existing trajectory measurement devices 10. For example, a golf ball 42 trajectory measuring device using various technologies such as general-purpose infrared rays, X-band radar, K-band radar, lasers, stereo cameras, and machine vision, or a measuring device for performing three-dimensional analysis of the club head 41a can be used.

[0023] The control device 20 is composed of a computer or the like, and includes a semiconductor memory such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory), a storage such as a hard disk, and hardware such as a communication I / F. In the present embodiment, the control device 20 corresponds to the swing analysis device of the present invention.

[0024] The control device 20 includes a feature amount acquisition unit 21, an estimation unit 22, a control unit 23, a display unit 24, and an input unit 25.

[0025] In the semiconductor memory or the hard disk of the control device 20, an embodiment of the swing analysis program of the present invention is installed. Then, when this swing analysis program is executed by the CPU and the GPU, the above-described feature amount acquisition unit 21, estimation unit 22, and control unit 23 function. In the present embodiment, all the functions of the above-described respective units are executed by the swing analysis program, but the present invention is not limited to this, and some or all of the functions may be configured by hardware such as an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or other electric circuits.

[0026] Hereinafter, each unit of the control device 20 will be described in detail.

[0027] When the golf club 41 is swung, the feature amount acquisition unit 21 acquires a plurality of feature amounts at the time when the club head 41a of the golf club 41 collides with the golf ball 42 (at the time of impact). Specifically, the feature amount acquisition unit 21 of the present embodiment acquires the eight features measured by the above-described trajectory measurement device 10. When the eight feature amounts are acquired, the coordinates of the position of the tee (the installation position P shown in FIG. 1) among the ground space coordinates are also acquired.

[0028] Based on the plurality of feature quantities acquired by the feature quantity acquisition unit 21, the estimation unit 22 estimates the temporal change in the angle of the face surface of the club head 41a and the movement locus of the club head 41a. The angle of the face surface of the club head 41a is the angle of the face surface when the club head 41a during the swing is viewed from above, and the movement locus of the club head 41a is the movement locus when the club head 41a during the swing is viewed from above.

[0029] Specifically, the estimation unit 22 of the present embodiment has a learning model obtained by previously performing machine learning on the relationship between the above eight feature quantities acquired in advance and the temporal change in the angle of the face surface and the movement locus of the club head during the swing when the feature quantities were acquired.

[0030] The learning model is machine-learned using a method such as CNN (Convolutional Neural Network), but the machine learning method is not limited to this, and other known methods can be used.

[0031] Then, the estimation unit 22 inputs the eight feature quantities acquired by the feature quantity acquisition unit 21 to the above learning model, thereby estimating the temporal change in the angle of the face surface of the club head 41a and the movement locus of the club head 41a. At this time, the coordinates of the tee position (installation position P shown in FIG. 1) are also used.

[0032] That is, the estimation unit 22 uses, as explanatory variables, the eight feature quantities acquired in advance, and uses a learning model that has been previously machine-learned with the temporal change in the angle of the face surface and the movement locus of the club head when the eight feature quantities were acquired as target variables, to estimate the temporal change in the angle of the face surface and the movement locus of the club head during an arbitrary swing.

[0033] Regarding the eight feature quantities that are explanatory variables used when generating the learning model, they can be measured using a device similar to the above-described ballistic measurement device 10. Further, regarding the temporal change in the angle of the face surface of the club head and the movement trajectory of the club head, which are objective variables used when generating the learning model, for example, they can be obtained based on a captured image obtained by photographing the club head from above with a photographing device when a golf club is swung.

[0034] As a method for obtaining the temporal change in the angle of the face surface of the club head based on a captured image obtained by photographing the club head from above, for example, the toe and heel of the club head included in the captured image are detected, and a line segment connecting the detected toe and heel is obtained as a line segment representing the face surface. For example, by obtaining the line segments representing the face surface from a plurality of captured images photographed at a frame rate of 1 ms, the temporal change in the angle of the face surface can be obtained. When performing machine learning, the angle of the line segment representing the face surface described above is used as the objective variable.

[0035] As a method for detecting the toe and heel from the captured image, for example, marks may be attached to the club head and the marks may be detected, or the captured image annotated with the positions of the toe and heel of the club head may be detected using a learning model obtained by machine learning.

[0036] Also, in the above description, a line segment connecting the toe and heel of the club head is detected as a line segment representing the face surface. However, when the golf club is a driver, the boundary line between the face surface and the crown portion of the driver may be detected as a line segment representing the face surface. Also in this case, the boundary line may be detected using a learning model obtained by machine learning of a captured image annotated with the position of the boundary line between the face surface and the crown portion of the club head included in the captured image. As machine learning, semantic segmentation machine learning or key point machine learning considering a heat map can be used.

[0037] In addition, as a method for obtaining the movement trajectory of the club head based on a captured image of the club head taken from above, for example, as described above, the center coordinates of the line segments representing the face surfaces detected from each captured image captured at a frame rate of 1 ms are obtained, and by obtaining a line segment connecting the center coordinates obtained from each captured image, the movement trajectory of the club head can be detected. When performing machine learning, the center coordinates of the line segments representing the face surfaces are used as target variables.

[0038] Regarding the learning model, conversion to an onnx model or the like may be performed in order to speed up the processing. Also, various existing statistical methods can be used to remove outliers in the learning data.

[0039] Next, the control unit 23 controls the entire golf impact analysis system 1. Specifically, the control unit 23 communicates with the trajectory measurement device 10 via the communication I / F, acquires eight feature amounts measured by the trajectory measurement device 10, and inputs the eight feature amounts into the learning model, thereby estimating the temporal change of the face surface of the club head and the movement trajectory of the club head.

[0040] The control unit 23 generates line segments representing a plurality of face surfaces based on the angles of the plurality of face surfaces output from the learning model, and generates a line segment connecting the plurality of center coordinates based on the center coordinates of the plurality of face surfaces output from the learning model.

[0041] Then, the control unit 23 communicates with the projection device 30 via the communication I / F, and causes the projection device 30 to project the line segments representing the plurality of face surfaces described above in time series onto the floor surface, and generates a line segment connecting the center coordinates of the plurality of face surfaces and projects it onto the floor surface as the movement trajectory of the club head.

[0042] The display unit 24 includes a display device such as a liquid crystal display. Further, the input unit 25 includes an input device such as a mouse or a keyboard. Alternatively, the control device 20 may be configured by a tablet terminal, and the display unit 24 and the input unit 25 may be configured by a touch panel.

[0043] Further, the display unit 24 may be configured to display the line segments representing the plurality of face surfaces described above and the movement locus of the club head.

[0044] In the present embodiment, the control unit 23 communicates with the ballistic measurement device 10 via the communication I / F and acquires the eight feature amounts measured by the ballistic measurement device 10. However, the present invention is not limited to this, and the eight feature amounts measured by the ballistic measurement device 10 may be set and input by the user using the input unit 25.

[0045] Returning to FIG. 1, the projection device 30 is configured by a projector or a laser beam irradiation device, and projects and displays the line segment representing the face surface of the club head 41a and the movement locus of the club head on the floor surface under the control of the control unit 23 of the control device 20 as described above.

[0046] The projection device 30 is installed so as to be able to project and display the line segment representing the face surface of the club head 41a and the movement locus of the club head in a predetermined projection range including the vicinity of the installation position P of the golf ball 42 on the floor surface. The projection device 30 may be installed on a support member such as a stand, or may be installed on the ceiling.

[0047] Next, the processing flow of the golf impact analysis system 1 will be described with reference to the flowchart shown in FIG. 2.

[0048] First, each device of the golf impact analysis system 1 is activated (S10). Next, as shown in FIG. 1, a golfer stands near the installation position P of the golf ball 42, and installs the golf ball 42 at the installation position P.

[0049] Then, a measurement start instruction of the trajectory measurement device 10 is set and input by the user, and the measurement by the trajectory measurement device 10 is started (S12).

[0050] After the measurement by the trajectory measurement device 10 is started, the golfer swings the golf club 41 and hits the golf ball 42 (S14).

[0051] While the golfer swings the golf club 41, the measurement by the trajectory measurement device 10 is performed (S16), and the above-mentioned eight feature amounts are output to the control device 20. The trajectory measurement device 10 ends the measurement when a preset time has elapsed since the start of the measurement (S18).

[0052] The eight feature amounts output from the trajectory measurement device 10 are input to the control device 20 and acquired by the feature amount acquisition unit 21 (S20). Then, the estimation unit 22 inputs the eight feature amounts acquired by the feature amount acquisition unit 21 into the above-mentioned learning model to estimate the temporal change in the angle of the face surface of the club head 41a and the movement trajectory of the club head 41a (S22). Specifically, the estimation unit 22 estimates and outputs a plurality of face surface angles arranged in time series as the temporal change in the angle of the face surface, and estimates and outputs the center coordinates of a plurality of face surfaces as the movement trajectory of the club head.

[0053] Then, the control unit 23 generates a line segment representing a plurality of face surfaces arranged in time series based on the plurality of face surface angles output from the estimation unit 22, generates a line segment representing the movement trajectory of the club head by connecting the center coordinates of the plurality of face surfaces, and combines these to be projected and displayed on the floor surface by the projection device 30 (S24).

[0054] In the above embodiment, the temporal change in the face surface of the club head and the movement trajectory of the club head are estimated. However, at least one of these may be estimated.

[0055] In the above-described embodiment, eight feature amounts, namely, the speed of the golf ball, the hitting direction, the side spin, the head speed of the club head, the club path, the face-to-target, the face-to-path, and the hitting point, are used. However, the present invention is not limited to this, and other feature amounts may be added or replaced as long as they are the feature amounts measured at the time when the club head of the golf club collides with the golf ball. Further, the number of feature amounts is more preferably eight or more, but may be, for example, five or four. Note that the number of feature amounts used when machine learning the learning model may be changed in accordance with the number of feature amounts input to the learning model.

[0056] Next, an example of a projection image projected onto the floor surface according to the above-described embodiment will be described.

[0057] FIGS. 3A and 3B are diagrams showing an example of a projection image of a line segment F representing a plurality of face surfaces and a line segment representing a movement locus T of the club head 41a estimated by measuring eight feature amounts using GCQuad (manufactured by Foresight Sports) as the trajectory measurement device 10 as described above and inputting these to a learning model.

[0058] FIG. 3A is an example of a projection image when an iron is used as the golf club 41. Along with the line segment F representing the face surface and the movement locus T of the club head, a three-dimensional image IG of the iron club head and a three-dimensional image BG of the golf ball are projected. The three-dimensional image IG of the club head moves along the movement locus T of the club head toward the three-dimensional image BG of the golf ball. When it reaches the three-dimensional image BG of the golf ball, the three-dimensional image BG of the golf ball moves in the hitting direction. In FIG. 3A, the line segment F representing the face surface of the club head and the movement locus T are displayed up to the position of the three-dimensional image BG of the golf ball. However, even after the club head reaches the golf ball, the line segment F representing the face surface of the club head and the movement locus T may be continuously estimated, obtained, and displayed.

[0059] Figure 3B shows an example of a projected image when a driver is used as the golf club 41. Along with the line segment F representing the face surface and the movement trajectory T of the club head, a three-dimensional image DG of the club head of the driver and a three-dimensional image BG of the golf ball are projected. The three-dimensional image DG of the club head moves along the movement trajectory T of the club head toward the three-dimensional image BG of the golf ball. When it reaches the three-dimensional image BG of the golf ball, the three-dimensional image BG of the golf ball moves in the hitting direction. Also, in Fig. 3B, even after the club head reaches the golf ball, the line segment F representing the face surface of the club head and the movement trajectory T may be estimated and obtained continuously and displayed.

[0060] According to the experiments by the applicant, the line segment F representing the face surface of the club head and the line segment representing the movement trajectory T of the club head included in the projected images shown in Figs. 3A and 3B are very close to the line segment representing the face surface and the line segment representing the movement trajectory of the club head obtained from the captured image of the club head taken from above, and it has been found that the temporal changes in the angle of the face surface of the actual club head 41a and the movement trajectory can be expressed with high accuracy.

[0061] Next, Figs. 4A and 4B are diagrams showing examples of projected images of line segments F representing a plurality of face surfaces and line segments T representing the movement trajectory of the club head 41a estimated by measuring five feature quantities (ball speed, hitting direction, side spin, head speed, and club path) using GC3 (manufactured by Foresight Sports) as the ballistics measurement device 10 and inputting these into a learning model.

[0062] FIG. 4A is an example of a projected image when an iron is used as the golf club 41, and FIG. 4B is an example of a projected image when a driver is used as the golf club 41. The projection and movement of the three-dimensional image IG of the iron club head, the three-dimensional image DG of the driver club head, and the three-dimensional image BG of the golf ball are the same as the examples in FIGS. 3A and 3B. Also, in FIGS. 4A and 4B, after the club head reaches the golf ball, the line segment F representing the face surface of the club head and the movement locus T may be estimated, obtained, and displayed continuously.

[0063] According to the experiments by the applicant, compared with the examples of the projected images in FIGS. 3A and 3B using eight feature amounts, there were some differences in the position and angle of the line segment F representing the face surface and the accuracy of the curvature of the movement locus T of the line segment representing the movement locus T of the club head 41a included in the projected images shown in FIGS. 4A and 4B, but it was confirmed that these were not particularly problematic differences.

[0064] Next, FIGS. 5A and 5B are diagrams showing examples of projected images of line segments F representing a plurality of face surfaces and line segments representing the movement locus T of the club head, which are measured using four feature amounts (ball speed, launch direction, side spin, and head speed) using GCQuad (manufactured by Foresight Sports) or GC3 (manufactured by Foresight Sports) as the ball flight measurement device 10 and input into the learning model. Note that as the ball flight measurement device 10, a measuring instrument with fewer measurement items than GCQuad or GC3 may be used.

[0065] FIG. 5A is an example of a projected image when an iron is used as the golf club 41, and FIG. 5B is an example of a projected image when a driver is used as the golf club 41. Regarding the projection and movement of the three-dimensional image IG of the club head of the iron, the three-dimensional image DG of the club head of the driver, and the three-dimensional image BG of the golf ball, it is the same as the examples of FIGS. 3A and 3B. Also, in FIGS. 5A and 5B, even after the club head reaches the golf ball, a line segment F representing the face surface of the club head and a movement locus T may be estimated, obtained, and displayed continuously.

[0066] According to experiments by the applicant, the line segment F representing the face surface of the club head 41a included in the projected images shown in FIGS. 5A and 5B and the line segment representing the movement locus T of the club head, when compared with the examples of the projected images in FIGS. 3A and 3B using eight feature amounts and the examples of the projected images in FIGS. 4A and 4B using five feature amounts, there were somewhat differences in the position and angle of the line segment F representing the face surface and the accuracy of the linearity of the movement locus T, but it was confirmed that these were not particularly problematic differences.

[0067] As described above, according to experiments by the applicant, it was found that the more the number of feature amounts, the higher the accuracy of the line segment F representing the face surface and the movement locus T of the club head.

[0068] According to the golf impact analysis system of the above-described embodiment, when the golf club 41 is swung, a plurality of feature amounts at the time when the club head 41a of the golf club 41 collides with the golf ball 42 are acquired, and based on the plurality of feature amounts, at least one of the temporal change in the angle of the face surface of the club head 41a and the movement locus of the club head 41a is estimated. Therefore, by acquiring the above-described feature amounts using an existing ballistic measurement device 10 or the like, the movement of the face surface and the movement locus of the club head 41a can be obtained with high accuracy without installing a photographing device for photographing the swing from above.

[0069] In the golf impact analysis system of the above-described embodiment, by inputting a plurality of feature amounts to a learning model obtained by machine learning in advance, the temporal change in the angle of the face surface of the club head 41a and the movement locus of the club head 41a are estimated, so that these can be estimated with higher accuracy.

[0070] In the golf impact analysis system of the above-described embodiment, as the feature amounts, the speed of the golf ball, the hitting direction, the side spin, the head speed of the club head, the club path, the face to target, the face to path, and the hitting point are used. Therefore, the feature amounts can be measured by the existing trajectory measurement device 10, and the temporal change in the angle of the face surface and the movement locus of the club head 41a can be estimated with higher accuracy.

[0071] In the above-described embodiment, the eight feature amounts described above are used. However, when the closure rate can be measured by the trajectory measurement device 10, the closure rate may be further added. The closure rate is the degree (opening / closing rate) at which the face surface rotates when the club head moves toward the impact position, and is measured as a relative value with respect to the club path.

[0072] Since the closure rate is the change rate of the face surface of the club head measured by 3D 6dof (degree of freedom), it cannot be immediately used for the two-dimensional display of the face surface as shown in FIGS. 3 to 5. However, in the above-described embodiment, since it is used as an explanatory variable in the learning model, it can be used for the two-dimensional display of the face surface.

[0073] According to the experiments of the applicant, it has been found that when the closure rate is used, the temporal change in the angle of the face surface of the club head and the movement locus after impact can be obtained with higher accuracy.

[0074] Here, it has been found that one of the causes of a mishit shot caused by an impact where the impact point is away from the toe direction or the heel direction from the center of the face surface of the club head is that the face surface of the club head itself is overwhelmed by the gravity of the ball. In recent years, golf clubs have been designed to increase the MOI (moment of inertia) in order to provide resistance to this being overwhelmed, and even for a mishit shot (off-center shot) where the center of the face surface is missed, a more stable straight shot can now be hit.

[0075] As described above, by more accurately displaying the temporal change and movement trajectory of the angle of the face surface of the club head after impact, it is possible to display on the floor surface how much the club head has been overwhelmed after impact, which is very effective in comparing golf clubs from the perspective of seeing the effect of the MOI.

[0076] Also, in the golf impact analysis system of the above-described embodiment, as the temporal change in the angle of the face surface, a plurality of line segments representing the face surface are arranged in time series and projected and displayed, so that the temporal change in the angle of the face surface can be recognized more clearly.

[0077] Also, in the golf impact analysis system of the above-described embodiment, as the movement trajectory of the club head 41a, a line segment representing the movement trajectory of the club head 41a is projected and displayed, so that the movement trajectory of the club head 41a can be recognized more clearly.

[0078] Note that the present invention is not limited to the above-described embodiment, and components can be modified and embodied without departing from the gist thereof at the implementation stage. Also, various inventions can be formed by appropriately combining a plurality of components disclosed in the above-described embodiment. For example, all the components shown in the embodiment may be appropriately combined. Of course, various modifications and applications are possible within the scope not departing from the gist of the invention.

[0079] Regarding the present invention, the following additional remarks are disclosed.

[0080] (Supplementary Note 1) The swing analysis device of the present invention includes a feature quantity acquisition unit that acquires a plurality of feature quantities at the time when the club head of the golf club and the golf ball collide when the golf club is swung, and based on the plurality of feature quantities acquired by the feature quantity acquisition unit, an estimation unit that estimates at least one of the temporal change in the angle of the face surface of the club head and the movement locus of the club head.

[0081] (Supplementary Note 2) In the swing analysis device described in Supplementary Note 1, the estimation unit inputs the plurality of feature quantities acquired by the feature quantity acquisition unit to a learning model obtained by previously performing machine learning on the relationship between the plurality of feature quantities acquired in advance and at least one of the temporal change in the angle of the face surface of the club head and the movement locus of the club head in the swing when the feature quantities were acquired, thereby estimating at least one of the temporal change in the angle of the face surface of the club head and the movement locus of the club head.

[0082] (Supplementary Note 3) In the swing analysis device described in Supplementary Note 1 or 2, the plurality of feature quantities can include at least one of the speed of the golf ball, launch direction, side spin, head speed of the club head, club path, face-to-target, face-to-path, and impact point.

[0083] (Supplementary Note 4) In the swing analysis device described in any one of Supplementary Notes 1 to 3, the plurality of feature quantities can include all of the speed of the golf ball, launch direction, side spin, head speed of the club head, club path, face-to-target, face-to-path, impact point, and closure rate.

[0084] (Supplementary Note 5) In the swing analysis device according to any one of Supplementary Notes 1 to 4, it is possible to include a display control unit that displays at least one of the temporal change in the angle of the face surface of the club head and the movement trajectory of the club head estimated by the estimation unit.

[0085] (Supplementary Note 6) In the swing analysis device according to Supplementary Note 5, the display control unit can display a plurality of line segments representing the face surface in time series as the temporal change in the angle of the face surface.

[0086] (Supplementary Note 7) In the swing analysis device according to Supplementary Note 5 or 6, the display control unit can display a line segment representing the movement trajectory of the club head as the movement trajectory of the club head.

[0087] (Supplementary Note 8) The swing analysis program of the present invention causes a computer to execute a step of acquiring a plurality of feature amounts at the time when the club head of a golf club and a golf ball collide when the golf club is swung, and a step of estimating at least one of the temporal change in the angle of the face surface of the club head and the movement trajectory of the club head based on the plurality of feature amounts acquired by the feature amount acquisition unit.

Explanation of Signs

[0088] 1 Golf Impact Analysis System 10 Ballistic Measurement Device 20 Control Device 21 Feature Amount Acquisition Unit 22 Estimation Unit 23 Control Unit 24 Display Unit 25 Input Unit 30 Projection Device 40 Golfer 41 Golf Club 41a Club Head 42 Golf Ball F Line Segment Representing the Face Surface P Installation Position Movement trajectory of the T club head

Claims

1. a feature acquisition unit that acquires a plurality of feature values ​​at the time when a club head of the golf club collides with a golf ball when the golf club is swung, the feature values ​​being a speed of the golf ball, a launch direction, a side spin, a head speed of the club head, a club path, a face-to-target, a face-to-path, and a hitting point; and an estimation unit that estimates at least one of a change over time in the angle of the face of the club head as viewed from above before and after a position at which the club head impacts a golf ball and a movement trajectory of the club head as viewed from above before and after a position at which the club head impacts a golf ball, by inputting the multiple feature amounts acquired by the feature amount acquisition unit into a learning model obtained in advance by machine learning the relationship between the multiple feature amounts acquired in advance and at least one of a change over time in the angle of the face of the club head as viewed from above and a movement trajectory of the club head acquired based on an image of the club head photographed from above during a swing when the feature amounts were acquired.

2. A swing analysis device as described in claim 1, wherein the multiple features are measured by one or more trajectory measurement devices installed at a location other than the golf club.

3. A swing analysis device as described in claim 2, wherein the trajectory measurement device measures the multiple feature amounts by photographing the moment of impact between the club head and the golf ball with a camera.

4. 2. The swing analysis device according to claim 1, further comprising a display control unit that displays at least one of the change over time in the angle of the face surface of the club head estimated by the estimation unit and the movement trajectory of the club head.

5. 2. The swing analysis device according to claim 1, further comprising a display control unit that displays a plurality of line segments representing the face surface in a time series based on the change over time in the angle of the face surface of the club head estimated by the estimation unit.

6. The swing analysis device according to claim 5 , wherein the display control unit displays a line segment representing the movement trajectory of the club head as the movement trajectory of the club head.

7. acquiring a plurality of feature quantities, such as a speed of the golf ball, a launch direction, a side spin, a head speed of the club head, a club path, a face-to-target, a face-to-path, and a hitting point, at a time when a club head of the golf club collides with a golf ball when the golf club is swung; a step of estimating at least one of a change over time in the angle of the face of the club head as viewed from above before and after a position where the club head impacts a golf ball and a movement trajectory of the club head as viewed from above before and after a position where the club head impacts a golf ball by inputting a plurality of feature amounts acquired by the feature amount acquisition unit into a learning model obtained by previously performing machine learning on the relationship between the plurality of feature amounts acquired in advance and at least one of a change over time in the angle of the face of the club head as viewed from above and a movement trajectory of the club head as viewed from above before and after a position where the club head impacts a golf ball.

Images