JP2017113584A - Measuring method of golf club head - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a measuring method capable of highly precisely and easily measuring an impact state of a head.SOLUTION: A head measuring method of the present invention includes the steps (a), (b) and (c): (a) preparing a golf club having a head having a plurality of markers provided thereon; (b) photographing the head by a camera disposed at a position, at which a backward distance from a center point of a ball is equal to or greater than 0, to obtain a head image near an impact; and (c) analyzing the head image to calculate a position and attitude of the head near the impact. Preferably, three or more combinations of the markers exist where the interval between the markers is equal to or greater than a head vertical width.SELECTED DRAWING: Figure 4

Description

  The present invention relates to a head measuring method during a golf swing.

  The position or posture of the head during the swing is useful information. This information can be used for swing analysis, club fitting, and the like. In particular, the state of the head near the impact is important because it is closely related to the hitting result.

  Japanese Patent Application Laid-Open No. 2004-24488 discloses an impact state measurement method using a camera set at a predetermined height ahead in the flying ball direction from the ball set position.

  Japanese Patent Laying-Open No. 2007-167549 discloses a golf club head behavior analysis device including at least two camera devices.

  Japanese Patent Application Laid-Open No. 2004-61483 discloses a plurality of two-dimensional images obtained by photographing a rotating curved body having a plurality of marks on the surface a plurality of times, and a plurality of marks on the surface in the same manner as the curved body. A measurement method for obtaining the amount of rotation and the direction of the rotation axis of a curved surface using a virtual curved surface is disclosed.

JP 2004-24488 A JP 2007-167549 A JP 2004-61483 A

  In the prior art, the head at impact was shot from the front of the ball. The face surface at the impact can be photographed by photographing from the front. This face surface image is effective in measuring the position and orientation of the head. The image from the front is effective to know the positional relationship between the ball and the face in impact.

  In the image from the front, a part of the head can be hidden by the ball. Therefore, the marker attached to the head may be hidden by the ball. In this case, the measurement accuracy may deteriorate. Moreover, it is preferable that measurement can be performed more simply.

  An object of the present invention is to provide a measurement method capable of easily and accurately measuring the impact state of a head.

The head measuring method according to the present invention includes the following steps (a), (b) and (c).
(A) A step of preparing a golf club having a plurality of markers on the head.
(B) A step of capturing the head with a camera disposed at a position where the rear distance from the ball center point is 0 or more to obtain a head image near the impact.
(C) analyzing the head image and calculating the position and orientation of the head near the impact.

  When the ball is not used, such as swinging, the range near the impact and the camera position can be determined based on the assumed ball position.

  The three-dimensional distance between the markers is the marker interval. Preferably, there are three or more combinations of markers such that the marker interval is equal to or greater than the head vertical width.

  In the head image, the area of the marker formation region defined by connecting the plurality of markers with straight lines is Sm. In the head image, the area of the entire head area defined by the outer contour of the head is Sh. At this time, Preferably, Sm / Sh is 0.25 or more.

  Preferably, the number N of the markers is 3 or more and 20 or less.

  In the head image, when the centroid Gh of the outer contour line of the head is determined and the head image is divided into four by the straight lines L1 and L2 orthogonal to each other with the centroid Gh as an intersection, Of the four sections, at least one of the markers is arranged in each of the two sections at diagonal positions.

  Preferably, in the step (c), the position and orientation of the head in the vicinity of the impact are calculated by analyzing the head image by one camera.

  Preferably, three or more of the markers are arranged outside the face surface.

  The impact state of the head can be measured with high accuracy and simplicity.

FIG. 1 is a perspective view showing an example of a measuring apparatus used in the measuring method of the present invention. FIG. 2 is a front view showing the camera position in FIG. FIG. 3 is a perspective view showing another example of the measuring device used in the measuring method of the present invention. FIG. 4 is a front view showing the camera position in FIG. FIG. 5 is a plan view showing the camera position in FIG. FIG. 6 is a front view showing the head vertical width of the wood-type head. FIG. 7 is a front view showing the iron head vertical width. FIG. 8 shows an example of a head image. FIG. 8 shows a marker formation region. FIG. 9 shows an example of a head image. FIG. 9 shows the entire head area. FIG. 10 shows an example of a head image. FIG. 10 is an image obtained by reproducing the head state at impact with a jig. FIG. 11 shows the head (wood type head) used in FIG. FIG. 12 shows another example of the head image. FIG. 12 is an image obtained by reproducing the head state at impact using a jig. FIG. 13 shows the head (iron type head) used in FIG. FIG. 14 is a diagram for explaining the DLT method.

  Hereinafter, the present invention will be described in detail based on preferred embodiments with appropriate reference to the drawings.

  FIG. 1 shows an example of a measuring apparatus that can be used for the measurement of the present invention. The measuring device 10 includes a base 12, a support column 14, a floor board 16, a tee 18, a head camera CM1, a ball camera 22, a trigger sensor 23 (23a, 23b), a first sensor 24 (24a, 24b), and a second sensor 26 ( 26a, 26b), strobe 28 (28a, 28b), strobe 29 (29a, 29b), control device 30 and information processing device 32. The head camera CM1 is also simply referred to as a camera below.

  In FIG. 1, the golf club 34 and the golf ball 36 are shown together with the measuring device 10. The golf club 34 includes a head 34a and a shaft 34b. In FIG. 1, the address posture of the right-handed golf player P is indicated by a two-dot chain line. The golf ball 36 is launched to the left of the player in this address posture.

  The support column 14 and the floor board 16 are fixed to the base 12. The support column 14 extends upward from the base 12. A tee 18 is positioned and attached to the floor board 16. The camera CM1 is fixed on the upper side of the player P. The ball camera 22 is located in front of the tee 18 and is attached to the side surface of the floor board 16. The camera CM1 is arranged so that the head near the impact can be photographed. Although not shown, the camera CM1 is fixed to a ceiling, for example. The ball camera 22 is arranged so that the golf ball 36 immediately after hitting can be photographed.

  The trigger sensor 23 includes a light emitter 23a and a light receiver 23b. The light emitter 23 a is disposed on one side surface of the floor plate 16, and the light receiver 23 b is disposed on the other side surface of the floor plate 16. The first sensor 24 includes a light emitter 24a and a light receiver 24b. The light emitter 24 a is disposed on one side surface of the floor plate 16, and the light receiver 24 b is disposed on the other side surface of the floor plate 16. The second sensor 26 includes a light emitter 26a and a light receiver 26b. The light emitter 26 a is disposed on one side surface of the floor plate 16, and the light receiver 24 b is disposed on the other side surface of the floor plate 16. The first sensor 24 and the second sensor are positioned so that the head 34a or the shaft 34b of the golf club 34 to be swung down crosses between the light emitter 24a and the light receiver 24b and between the light emitter 26a and the light receiver 26b. 26 is arranged.

  The strobe 28 (28a, 28b) is attached to the support column 14. The control device 30 is attached to the base 12.

  The control device 30 is connected to the camera CM 1, the ball camera 22, the trigger sensor 23, the first sensor 24, the second sensor 26, the strobe 28, the strobe 29, and the information processing device 32. The control device 30 can receive a detection signal of the head 34 a or the shaft 34 b from the trigger sensor 23. Based on the detection signal from the trigger sensor 23, the control device 30 can transmit a shooting start signal to the camera CM1. Further, the control device 30 can transmit a shooting start signal to the ball camera 22. The control device 30 can receive image signals taken from the camera CM1 and the ball camera 22. The controller 30 may receive detection signals of the head 34a or the shaft 34b from the sensors 24 and 26. The control device 30 can transmit a light emission start signal to the strobes 28 and 29.

  Based on the shooting start signal, the shutter of the camera CM1 opens for a predetermined time (for example, 1/30 second). While the shutter is open, the head 34a or the shaft 34b crosses the sensor 24 and the sensor 26. When the sensor 24 is shut off, the strobe 28a emits light, and when the sensor 26 is shut off, the strobe 28b emits light. Therefore, the strobe 28a and the strobe 28b sequentially emit light while the shutter is open. As a result, the head when the sensor 24 is shut off and the head when the sensor 26 is shut off are copied onto one image.

  In the embodiment of FIG. 1, the head 34a is a wood-type head. The head may not be a wood type, and examples include an iron type head, a utility type head, a hybrid type head, and a putter type head.

  Although not shown, the information processing apparatus 32 includes a monitor as an output unit, an interface board as a data input unit, a memory, a CPU, and a hard disk. The information processing device 32 may include a keyboard and a mouse. A general-purpose computer may be used as the information processing apparatus 32 as it is.

  The hard disk stores a program. The memory is rewritable, and constitutes a storage area and a work area for programs called from the hard disk and various data. The CPU can read a program stored in the hard disk. The CPU can expand the program in the work area of the memory. The CPU can execute various processes according to the program. For example, this program can calculate the position and orientation of the head based on the head image.

  Head image data can be input to the interface board. Head image data, ball image data, and synchronization data of the two image data may be input. These input data are output to the CPU. The CPU performs various processes. By this processing, the posture and position of the head 34a can be calculated. Further, a club behavior value and a ball behavior value may be calculated. Of these calculated values, predetermined data is output to the monitor. In addition, predetermined data is stored in the hard disk.

  In the present application, the left-right direction of the player P in this address posture is the front-rear direction. The direction in which the ball flies is forward.

  For convenience of explanation, the X axis, the Y axis, and the Z axis are defined in the present application (see FIGS. 1 and 2). These X axis, Y axis, and Z axis are orthogonal coordinate systems. The X axis is parallel to the ground and parallel to a straight line connecting the ball 36 before hitting and the target direction. The Z axis is the vertical direction. The Y axis is perpendicular to the X axis and perpendicular to the Z axis. The front-rear direction in the present application is the direction of the X axis.

  The measuring device 10 can capture a head image near the impact. The vicinity of impact includes impact. The impact is a state where the head 34a and the ball 36 are in contact with each other. In the vicinity of the impact, the shortest distance between the head 34a and the center of the ball 36 is preferably within 30 cm, and more preferably within 20 cm. Preferably, the vicinity of the impact is before impact and before impact.

  A marker 38 is provided on the head 34a. The marker 38 is, for example, a tape. The marker 38 may be identified in the head image. A plurality of markers 38 are provided on the head 34a. In the embodiment of FIG. 1, four markers 38 are provided. In the present embodiment, three or more markers 38 are provided outside the face surface. In the present embodiment, all the markers 38 are provided outside the face surface. In the present embodiment, all the markers 38 are provided on the crown of the head 34a. A golf ball 36 is set on the tee 18. With this golf club 34, the golf player P addresses.

  In an example of a measurement method using the measurement device 10, first, the golf player P starts swinging with the golf club 34. In the process from the downswing to the impact, the first sensor 24 detects the golf club 34. A detection signal of the first sensor 24 is output to the control device 30. The control device 30 outputs a light emission start signal to the strobe 28a at time T1 after receiving the detection signal. In response to this signal, the flash 28a emits light. The control device 30 outputs a shooting start signal to the camera CM1 at time T2 after receiving the detection signal.

  Next, the second sensor 26 detects the golf club 34. A detection signal of the second sensor 26 is output to the control device 30. The control device 30 outputs a light emission start signal to the strobe 28b at time T3 after receiving the detection signal. In response to this signal, the flash 28b emits light. The control device 30 outputs a shooting start signal to the camera CM1 at time T4 after receiving the detection signal.

  As will be described later, in this embodiment, the posture and position of the head can be calculated from only one head image. In this case, the flash may be emitted once. When calculating the head trajectory, the number of flashes of the strobe is preferably plural from the viewpoint of obtaining a plurality of head images.

  The camera CM1 images the head 34a in the vicinity of the impact. As will be described later, in this embodiment, the position and orientation of the head 34a can be calculated from one image.

  The control device 30 outputs a light emission signal to the strobe 29a at time T5. The control device 30 outputs a shooting start signal to the ball camera 22 at time T6. The control device 30 outputs a light emission signal to the strobe 29b at time T7, and outputs a shooting start signal to the ball camera 22 at time T8.

  The control device 30 outputs the head image data to the information processing device 32. The control device 30 may output time data, head image data, and ball image data to the information processing device 32.

The information processing device 32 calculates the posture and position of the club from the head image data. Examples of data in the vicinity of the impact that can be calculated (hereinafter also referred to as impact data) are given below.
[Example of impact data]
・ Head position relative to the ball ・ Fitting point on the face surface ・ Lie angle ・ Loft angle ・ Face angle ・ Head track (blow angle, approach angle, etc.)

The head measurement method of this embodiment includes the following steps (a), (b) and (c).
(A) A step of preparing a golf club 34 in which a plurality of markers 38 are provided on the head 34a.
(B) A step of capturing the head with the camera CM1 arranged at a position where the rear distance from the ball center point is 0 or more to obtain a head image in the vicinity of the impact.
(C) A step of analyzing the head image and calculating the position and orientation of the head 34a in the vicinity of the impact.

The preferred embodiment includes the following steps St1 to St6.
Step St1: A marker 38 is provided on the head 34a of the golf club 34. Step St2: The camera CM1 is calibrated to determine camera constants.
Step St3: The golf player P performs a swing and photographs the head 34a near the impact one or more times to obtain one or more head images. When obtaining the head trajectory, two or more head images are obtained by photographing two or more times at different times.
Step St4: Pointing each of the markers 38 in the obtained head image. This pointing can be done automatically or manually.
Step St5: The computer calculates the three-dimensional position of each marker 38, and calculates the position, posture, and the like of the head 34a based on these three-dimensional positions.
Step St6: The calculated result is output.

  In FIG. 2, what is indicated by a double arrow B1 is the rear distance of the camera CM1 from the ball center point. Preferably, the rear distance B1 is 0 or more. That is, the position of the camera CM1 in the X-axis direction is the same as the center point of the ball 36 or behind the center point of the ball 36. In the embodiment of FIG. 2, the position of the camera CM <b> 1 is behind the center point of the ball 36. The position of the camera CM1 suppresses the marker from being hidden by the ball in the image near the impact. Therefore, measurement accuracy can be improved.

  Note that the position of the camera CM1 can be determined by the center point of the lens. This center point may be the center point of the lens surface.

  From the viewpoint of measurement accuracy, more preferably, the rear distance B1 is greater than zero. That is, more preferably, the camera CM1 is located behind the center of the ball 36 by a distance B1. The embodiment of FIG. 2 shows this more preferred form.

  FIG. 3 shows a measuring apparatus 50 according to the second embodiment. The measuring device 50 is the same as the measuring device 10 except for the position of the camera CM1.

  4 and 5 show the position of the camera CM1 in the measuring device 50. FIG. 4 is a view from the front of the player P, and FIG. 5 is a view from above. Compared with the embodiment of FIG. 2, the rear distance B1 is increased in the embodiment of FIG. Therefore, the influence of the ball 36 on the head image can be further reduced. This can contribute to an improvement in measurement accuracy.

  From the viewpoint of eliminating the influence of the ball in the head image, the rear distance B1 is preferably equal to or greater than 0, more preferably greater than 0, more preferably equal to or greater than 3 cm, and still more preferably equal to or greater than 5 cm. From the viewpoint of obtaining a clear image, the rear distance B1 is preferably 200 cm or less, and more preferably 30 cm or less. In the first embodiment, the rear distance B1 is 8 cm.

  In FIGS. 2 and 4, what is indicated by a double arrow H1 is the height of the camera CM1 from the ground. In order to obtain a clear head image, an excessive height H1 is not preferable. From this viewpoint, the height H1 is preferably 300 cm or less, more preferably 250 cm or less, and more preferably 200 cm or less. This height H1 is measured along the Z-axis direction (vertical direction).

  The position of the camera CM1 in the Y-axis direction is preferably a position where the head image near the impact is not hidden by the player P. Here, the Y coordinate (distance S1 in FIG. 5) on the front side (front of the player P) from the center of the ball 36 is negative, and the Y coordinate on the rear side (back of the player P) from the center of the ball 36 is positive. It is said. From the viewpoint of suppressing the influence of the player P image on the head image, the Y coordinate of the position of the camera CM1 is preferably −200 cm to +200 cm, more preferably −100 cm to +100 cm, and −50 cm to 0 cm. Is more preferable. The Y coordinate of the center of the ball 36 is set to 0.

[Head vertical width of wood type head]
In the present application, a head vertical width F1 is defined. FIG. 6 shows the head vertical width F1 of the wood-type head Wh. In the measurement of the head vertical width F1, a plane HP (not shown) perpendicular to the horizontal plane h is considered. The head Wh is placed on the horizontal plane h, and the shaft axis line z1 is placed in the plane of the vertical plane HP. Furthermore, the angle θ between the shaft axis and the vertical plane HP is set to 60 °. In this state, the height from the horizontal plane h to the top of the crown is the head vertical width F1. The head vertical width F1 is measured along a direction perpendicular to the horizontal plane h.

[Head vertical width of iron type head]
FIG. 7 shows the head vertical width F1 of the iron type head Ih. In the measurement of the head vertical width F1, a plane HP (not shown) perpendicular to the horizontal plane h is considered. The head Ih is placed on the horizontal plane h, and the shaft axis line z1 is placed in the plane of the vertical plane HP. Further, the face line gv is made parallel to the horizontal plane h. In this state, the height from the horizontal plane h to the blade uppermost portion is the head vertical width F1. The head vertical width F1 is measured along a direction perpendicular to the horizontal plane h.

  In a head having a crown, such as a utility type head or a hybrid type head, the above-mentioned wood type head reference can be adopted as the head vertical width F1. Further, for the head vertical width F1 of the head having no crown, the above-mentioned reference of the iron type head can be adopted.

  FIGS. 8A and 9 show an example of the head image Pc1 in the vicinity of the impact. In FIG. 8, the description of the shaft and the like is omitted, and only the head is drawn. The head image Pc1 is taken by the measuring device 10.

  In this embodiment, as the four markers 38, a marker mk1, a marker mk2, a marker mk3, and a marker mk4 are used. All the markers mk1 to mk4 are arranged on the crown of the head 34a. These markers 38 are white and have a diagonal line drawn on a square tape. For example, the intersection of the diagonal lines is the position of the marker 38. The centroid of the marker 38 may be the position of the marker 38. Measurement accuracy can be further improved by determining the position of the marker 38 in detail.

[Marker interval]
In the present application, a three-dimensional distance between markers is defined as a marker interval. That is, the actual distance between the markers is the marker interval. Marker spacing can be determined for all marker combinations. For example, if there are four markers, there are six marker intervals.

  FIG. 8B shows the marker interval in the head of FIG. All combinations are considered in determining the marker spacing. Since there are four markers 38 in this head, there are six marker intervals. That is, in the head 34a, there are a marker interval D12, a marker interval D13, a marker interval D14, a marker interval D23, a marker interval D24, and a marker interval D34. The marker interval D12 is a distance between the marker mk1 and the marker mk2. The marker interval D13 is a distance between the marker mk1 and the marker mk3. The marker interval D14 is a distance between the marker mk1 and the marker mk4. The marker interval D23 is a distance between the marker mk2 and the marker mk3. The marker interval D24 is a distance between the marker mk2 and the marker mk4. The marker interval D34 is a distance between the marker mk3 and the marker mk4.

[Number of combinations Cn]
It is assumed that the combination of markers whose marker interval is equal to or greater than the head vertical width F1 is the Cn set. From the viewpoint of measurement accuracy, the number of combinations Cn is preferably 3 or more, and more preferably 4 or more. From the viewpoint of avoiding excessively complicated calculations, the number of combinations Cn is preferably 190 or less, and more preferably 45 or less.

[Marker formation region R1, area Sm]
In the head image Pc1, a marker formation region R1 is defined (see FIG. 8). This marker formation region R1 is indicated by the one-dot chain line hatching in FIG. The marker formation region R1 is defined by connecting a plurality of markers mk1 to mk4 with a straight line SL in the head image Pc1. When the number of the markers 38 is N, the marker forming region R1 is an N-gon. The area of the marker formation region R1 is Sm. This area Sm is an area in the head image Pc1 (two-dimensional image).

[Head entire region R2, area Sh]
Further, in the head image Pc1, the entire head region R2 is defined (see FIG. 9). The entire head region R2 is indicated by broken line hatching in FIG. The entire head region R2 is defined by the outer contour line Ct of the head 34a in the head image Pc1. The portion hidden by the shaft and the ferrule does not constitute the outer contour line Ct of the head 34a. That is, the outer contour line Ct is a line existing in the head image Pc1. The area of the entire head region R2 is Sh. This area Sh is also an area in the head image Pc1 (two-dimensional image).

[Sm / Sh]
From the viewpoint of improving measurement accuracy, Sm / Sh is preferably 0.25 or more, more preferably 0.5 or more, and further preferably 0.75 or more. Although Sm / Sh may be 1, Sm / Sh is usually 0.8 or less in consideration of the head shape and the like.

[Number of markers N]
The number N of the markers 38 is plural. From the viewpoint of measurement accuracy, the number N of the markers 38 is preferably 3 or more, and more preferably 4 or more. When the number N is excessive, calculation is complicated. In this respect, the number N is preferably 20 or less, and more preferably 10 or less. In the embodiment of FIGS. 8 and 9, the number N is four.

  By increasing Sm / Sh, even when the number N of markers is small, highly accurate measurement can be performed. Further, by increasing Sm / Sh, it is possible to measure with high accuracy even when the rear distance B1 is 0 or more.

[Centroid Gh, straight line L1, straight line L2]
In the head image Pc1, the centroid Gh of the outer contour line Ct of the head 34a is determined. A straight line L1 and a straight line L2 that are orthogonal to each other with the centroid Gh as an intersection are determined. The straight line L1 and the straight line L2 are arbitrary straight lines. That is, the straight line L1 and the straight line L2 can be any straight line as long as they pass through the centroid Gh and are orthogonal to each other. The straight line L1 and the straight line L2 can be determined innumerably. The straight line L1 and the straight line L2 shown in FIG. 9 are examples.

[Division by straight line L1 and straight line L2]
As shown in FIG. 9, the head image Pc1 is divided into four by the straight line L1 and the straight line L2. In the present embodiment, the head image Pc1 is divided into a first section A1, a second section A2, a third section A3, and a fourth section A4. In the present embodiment, the marker mk4 is located in the first section A1, the marker mk1 is located in the second section A2, the marker mk2 is located in the third section A3, and the marker mk3 is located in the fourth section A4. .

[Diagonal position]
As shown in FIG. 9, the marker mk4 and the marker mk2 are arranged in each of the two sections A1 and A3 at the diagonal positions among the four sections. Further, among the four sections, the marker mk1 and the marker mk3 are arranged in each of the two sections A2 and A4 that are diagonal positions. With these arrangements, the markers 38 are dispersed, and the measurement accuracy can be improved. In the present embodiment, at least one marker 38 is arranged in each of the four sections A1, A2, A3, and A4. Therefore, measurement accuracy can be further improved.

  A measurement error may occur due to the number of pixels in the head image or a deviation in pointing. By increasing the interval between the markers 38, the ratio of the deviation amount with respect to the marker interval is reduced. Therefore, the error can be reduced in the calculation in step (c).

  In the present embodiment, in the step (c), the position and posture of the head near the impact are calculated by analyzing only one head image Pc1. Therefore, it is not necessary to acquire a plurality of head images. The number of cameras CM1 may be one. Therefore, measurement, calculation, and apparatus are simplified. These simplifications improve the convenience of measurement and can reduce the cost of the apparatus.

  In the above embodiment, the head 34a is a wood type. The head 34a has a crown. The crown is relatively wide. By providing the markers 38 on the crown, the distance between the markers 38 can be increased. Providing the marker 38 on the crown is advantageous for increasing Sm / Sh.

  In the embodiment, three or more markers are arranged outside the face surface. Therefore, a situation in which these markers 38 are hidden by the ball 36 and the head posture is not calculated does not occur. In this case, it is easy to increase the number of combinations Cn. Therefore, measurement accuracy can be improved.

  When the head has a crown, all of the plurality of markers 38 may be arranged on the crown. In this case, the marker 38 is not hidden by the ball 36. Therefore, the effect obtained when the rear distance B1 is 0 or more can be further improved. Further, since the crown is relatively wide, it is easy to increase the number of combinations Cn. Therefore, measurement accuracy can be improved.

  When the rear distance B1 is 0 or more, the face surface may not be captured in the head image. In the present embodiment, measurement is possible even when the face surface is not captured in the head image.

  In the step (c), preferably, the position and orientation of the head 34a are calculated based on the position of the marker 38 in the head image Pc1. This calculation method can preferably calculate the position and orientation of the head 34a from one head image.

  In the preferred step (c), the three-dimensional coordinates of each of the plurality of markers 38 are calculated from the head image Pc1. The DLT method is known as an example of this calculation method. DLT is an abbreviation for “Direct Linear Transformation”.

  The DLT method is a method for obtaining three-dimensional spatial coordinates using a plurality of images viewed from different directions. In the DLT method, three-dimensional coordinates are reconstructed based on an image of a point (control point) whose three-dimensional coordinates are known. The DLT method is less versatile and has high versatility.

  This DLT method is illustrated by the following formula and FIG. FIG. 14 shows the relationship between real space coordinates and film surface coordinates.

FIG. 14 shows the coordinates (X, Y, Z) in the real space and the coordinates (U, V) on the film surface (U, V) when the point P in the real space (Object space) is photographed by the camera. Show the relationship. Here, the point O is the lens center point of the camera, and the coordinate system X′Y′Z ′ is a coordinate system in which the point O is the origin and the on-film coordinate system UV is parallel to the X ′ axis and the Y ′ axis. Yes, L is the distance on the Z ′ axis from the point O to the point P, F is the distance on the Z ′ axis from the point O to the point Q (mapping of the point P), and the point (U ₀ , V ₀ ) is the intersection of the straight line including the Z ′ axis and the film surface. In this case, the following explanation and mathematical formula are established.

In order to obtain these 11 camera constants, six or more points with known real space coordinates (X, Y, Z) and film surface coordinates (U, V) are photographed by the camera, and (X, By substituting Y, Z) and (U, V) into the above equation, a total of 12 or more equations are created, and 11 camera constants are obtained by the least square method. The operation for obtaining the camera constant is called calibration. If eleven camera constants are obtained, the coordinates (U, V) on the film surface of the points with known real space coordinates (X, Y, Z) can be obtained. On the other hand, to obtain the real space coordinates (X, Y, Z) from the coordinates (U, V) on the film surface, the same point was obtained using two or more cameras with known camera constants. By substituting (U ₁ , V ₁ ) (U ₂ , V ₂ )... Into the above equation, four or more equations are created, and (X, Y, Z) is obtained by the least square method. Such a method is called a DLT method.

  In the above method, two cameras are required to obtain the coordinates of one independent point in the space, and the coordinates of the point cannot be obtained from one camera image. However, with respect to the coordinates of a plurality of points fixed to a large object, if the positional relationship between the points (that is, the coordinates of each point in the object coordinate system) is known, the relational expression is expressed as a spatial coordinate (X, In addition to the equation for obtaining (Y, Z), the coordinates of a plurality of points (that is, the position and orientation of the object) can be obtained. The coordinates of each point (marker) in the object coordinate system (head coordinate system) are obtained, for example, by measuring the three-dimensional shape of the object (head). Since the purpose of this embodiment is to determine the position and orientation of the club head, the spatial coordinates (X, Y, Z) of the representative point of the object and the orientation (α, β, γ) of the object are unknown below. The case of solving as

In the above formula _{(F3), R} i (x) denotes the x component of the _{R i, R} i (y) represents the y component of _{R i, R} i (z) denotes the z-component of the _{R i} .

  These simultaneous equations (F3) having six unknowns and 2n equations can be solved by the Newton-Raphson method (solution of nonlinear simultaneous equations). Thereby, six unknowns, that is, the position of the object representative point and the posture of the object can be obtained. Note that α, β, and γ indicate coordinate conversion angles from the space coordinate system to the object coordinate system. In the present embodiment, the coordinate system Cs1 is obtained by rotating the spatial coordinate system XYZ about the Z axis by α °, and the coordinate system Cs2 is obtained by rotating the coordinate system Cs1 by β ° about the Y axis. When the coordinate system Cs3 is rotated by γ ° around the X axis to obtain the coordinate system Cs3, the coordinate system Cs3 coincides with the object coordinate system.

  The 2n simultaneous equations described above are defined as an equation (F3). A method for solving these equations (F3) by the Newton-Raphson method is shown below.

  The above method is an example of a so-called gradient method. Since this gradient method is simpler in calculation than the genetic algorithm used in the above-mentioned Japanese Patent Application Laid-Open No. 2004-61483, it is effective, for example, in a scene where the calculation result is to be displayed in real time. For example, in store fitting, there is a need to display the calculation result in real time and to show the calculation result to the customer on the spot. The above simple calculation method is suitable for displaying results in real time.

  In this way, based on the relative relationship between points whose coordinates in the object coordinate system are known, the spatial coordinates and the three-dimensional posture of the representative point of the head can be calculated from one head image.

  Hereinafter, the effects of the present invention will be clarified by examples. However, the present invention should not be construed in a limited manner based on the description of the examples.

[Test with wood-type head]

[Example 1]
A wood-type golf club head was prepared. A No. 1 wood head (driver head) was used. A plurality of markers were attached to the crown of the head. Each marker was approximately square. A white marker was used. Each marker has a diagonal line, and the intersection of the diagonal lines is pointed.

  This head was fixed to a head fixing jig, and the state of the head near the impact was reproduced. This head was photographed using the measuring apparatus 10 described above. FIG. 10 shows an example of a captured head image. The position of the head camera CM1 was such that the rear distance B1 was 8 cm, the height H1 was 110 cm, and the Y coordinate (distance S1) was −30 cm. This camera position is the camera position when it is assumed that the ball is in contact with the face center of the fixed head. That is, the head image in FIG. 10 can be said to be an image that reproduces the moment when the face contacts the ball in impact (immediately before the ball is crushed and deformed).

  A head image was taken for each set value of the jig face angle and the jig lie angle. That is, in Example 1, 19 head images were taken. Measurement values were calculated for each of these head images. All measured values were obtained from head images taken with one camera. In Examples 2 to 4 described later, the same head image as that in Example 1 was used.

  As shown in the head image of FIG. 10, a plurality of markers were also provided on a jig around the head. These markers are for calibration.

  Using the head image of FIG. 10 and the like, the posture of the head was calculated in steps St1 to St6 described above. Specifically, the face angle and the lie angle were calculated.

  In the head fixing jig, it is possible to set a desired lie angle and face angle. With this head fixing jig, the face angle and the lie angle were set to predetermined values and measured. The measurement error was verified by the difference between the measured value obtained from the head image and the actual set value. Multiple levels of testing were performed with different face and lie angle settings.

  The above-mentioned Newton-Raphson method was used for the calculation. A program capable of this calculation is installed in the hard disk, and the CPU calculates the three-dimensional coordinates of each marker according to the program and calculates the face angle and the lie angle based on these three-dimensional coordinates. Executed.

  FIG. 11 shows the identification number of the marker assigned to the wood type head. In Example 1, four markers of No. 15, No. 18, No. 28 and No. 32 were used (see FIG. 11). As shown in Table 9 below, the head vertical width F1 was 60.5 mm, and the number of combinations Cn was 5. In Example 1, Sm / Sh was 0.6 in the head image in the reference state (jig face angle 0 ° and jig lie angle 0 °). The results of Example 1 are shown in Table 1 below.

  In the head fixing jig, the set lie angle and face angle are displayed. The face angle displayed by the head fixing jig is referred to as a jig face angle. Similarly, the lie angle displayed on the head fixing jig is referred to as a jig lie angle. The jig face angle is slightly different from the actual face angle (absolute face angle). The jig lie angle is slightly different from the actual lie angle (absolute lie angle). These differences are obtained by measuring a reference state where the set jig lie angle and jig face angle are set to 0 °. In Example 1, the measurement face angle in the reference state was 0.25 °, and the measurement lie angle in the reference state was −3.09 °. Using these values, the measured values were converted into jig reference values. In Examples 2 to 4 to be described later, these values were used to convert the measured value to the jig reference value.

  The absolute face angle calculated based on the measurement was converted into a jig face angle. Then, the converted value was compared with the set value (jig face angle) in the jig. Similarly, the absolute lie angle calculated based on the measurement was converted into a jig lie angle, and the converted value was compared with a set value (jig lie angle) in the jig. That is, the set value and the measured value were aligned with the jig reference value and then compared. The absolute value of the difference between these jig reference values is shown in the column of “deviation from set value” in Tables 1 to 8.

[Example 2]
In Example 2, four markers No. 16, No. 17, No. 36 and No. 37 were used (see FIG. 11). The other measurements were obtained in the same manner as in Example 1. As shown in Table 10 described later, the head vertical width F1 was 60.5 mm, and the number of combinations Cn was 4. In Example 2, Sm / Sh was 0.3 in the head image in the reference state (jig face angle 0 ° and jig lie angle 0 °). The results of Example 2 are shown in Table 2 below.

[Example 3]
In Example 3, four markers No. 21, No. 26, No. 28 and No. 34 were used (see FIG. 11). The other measurements were obtained in the same manner as in Example 1. As shown in Table 11 below, the head vertical width F1 was 60.5 mm, and the number of combinations Cn was 4. In Example 3, Sm / Sh was 0.4 in the head image in the reference state (jig face angle 0 ° and jig lie angle 0 °). The results of Example 3 are shown in Table 3 below.

[Example 4]
In Example 4, four markers No. 24, No. 25, No. 30 and No. 31 were used (see FIG. 11). The other measurements were obtained in the same manner as in Example 1. As shown in Table 12 described later, the head vertical width F1 was 60.5 mm, and the number of combinations Cn was 0. In Example 4, Sm / Sh was 0.1 in the head image in the reference state (jig face angle 0 ° and jig lie angle 0 °). The results of Example 4 are shown in Table 4 below.

[Test with iron-type head]

[Example 5]
An iron-type golf club head was prepared. A 7-iron head was used. A plurality of markers were attached to the crown of the head. Each marker was approximately square. A white marker was used. Each marker has a diagonal line, and the intersection of the diagonal lines is pointed.

  This head was fixed to a head fixing jig, and the state of the head near the impact was reproduced. This head was photographed using the measuring apparatus 10 described above. The obtained head image is shown in FIG. The camera CM1 has a rear distance B1 of 8 cm, a height H1 of 110 cm, and the Y coordinate (distance S1) of −30 cm. This camera position is the camera position when it is assumed that the ball is in contact with the face center of the fixed head.

  Using the head image of FIG. 12 and the like, the face angle and lie angle were calculated in the same manner as in Example 1.

  A head image was taken for each set value of the jig face angle and the jig lie angle. That is, in Example 5, 19 head images were taken. Measurement values were calculated for each of these head images. All measured values were obtained from head images taken with one camera. In Examples 6 to 8 described later, the same head image as in Example 5 was used.

  FIG. 13 shows the identification number of the marker assigned to the iron type head. In Example 5, four markers No. 4, No. 7, No. 17, and No. 20 were used (see FIG. 13). As shown in Table 13 to be described later, the head vertical width F1 was 48.0 mm, and the number of combinations Cn was 5. In Example 5, Sm / Sh was 0.6 in the head image in the reference state (jig face angle 0 ° and jig lie angle 0 °). The results of Example 5 are shown in Table 5 below.

   In Example 5, the measurement face angle in the reference state was 0.30 °, and the measurement lie angle in the reference state was 0.26 °. Using these values, the measured values were converted into jig reference values. In Examples 6 to 8 to be described later, these values were used to convert the measured values to the jig reference values.

[Example 6]
In Example 6, four markers No. 5, No. 6, No. 18, and No. 19 were used (see FIG. 13). The other measurements were obtained in the same manner as in Example 5. As shown in Table 14 below, the head vertical width F1 was 48.0 mm, and the number of combinations Cn was 2. In Example 6, Sm / Sh was 0.2 in the head image in the reference state (jig face angle 0 ° and jig lie angle 0 °). The results of Example 6 are shown in Table 6 below.

[Example 7]
In Example 7, four markers of No. 3, No. 8, No. 11, and No. 20 were used (see FIG. 13). The other measurements were obtained in the same manner as in Example 5. As shown in Table 15 below, the head vertical width F1 was 48.0 mm, and the number of combinations Cn was 4. In Example 7, Sm / Sh was 0.3 in the head image in the reference state (jig face angle 0 ° and jig lie angle 0 °). The results of Example 7 are shown in Table 7 below.

[Example 8]
In Example 8, four markers of No. 9, No. 10, No. 12, and No. 13 were used (see FIG. 13). The other measurements were obtained in the same manner as in Example 5. As shown in Table 16 described later, the head vertical width F1 was 48.0 mm, and the number of combinations Cn was 0. In Example 8, Sm / Sh was 0.07 in the head image in the reference state (jig face angle 0 ° and jig lie angle 0 °). The results of Example 8 are shown in Table 8 below.

  Table 9 shows all the marker intervals in Example 1. Table 10 shows all of the marker intervals in Example 2. Table 11 shows all of the marker intervals in Example 3. Table 12 shows all of the marker intervals in Example 4. Table 13 shows all of the marker intervals in Example 5. Table 14 shows all of the marker intervals in Example 6. Table 15 shows all of the marker intervals in Example 7. Table 16 shows all of the marker intervals in Example 8.

  Tables 1 to 8 show an average value (average) and a maximum value (max) of deviation from the set value. It can be said that the smaller these values, the smaller the measurement error. As the number of combinations Cn increases, the measurement error tends to be suppressed.

  From the evaluation results of Tables 1 to 8, the superiority of the present invention is clear.

  The method described above can be applied to detection of the posture and / or position of the head.

10, 50 ... Measuring device 18 ... Tee CM1 ... Head camera (camera)
22 ... Ball camera 23, 23a, 23b ... Trigger sensor 24, 24a, 24b ... First sensor 26, 26a, 26b ... Second sensor 28, 28a, 28b, 29, 29a, 29b ..Strobe 30 ... Control device 32 ... Information processing device 34 ... Golf club 36 ... Golf ball 38 ... Marker

Claims (5)

A head measurement method including the following steps (a), (b) and (c),
(A) A step of preparing a golf club having a plurality of markers on the head.
(B) A step of capturing the head with a two-dimensional camera arranged at a position behind the ball center point to obtain a head image in the vicinity of the impact.
(C) analyzing the head image and calculating the position and orientation of the head near the impact.
In the step (c), by analyzing the head image photographed from one direction by the one two-dimensional camera, the three-dimensional position and three-dimensional posture of the head near the impact are calculated,
Based on the relative positional relationship between the plurality of markers, the three-dimensional position and the three-dimensional posture are calculated from the head image,
A head measurement method in which the number N of the markers is 3 or more and 20 or less. When the three-dimensional distance between the markers is the marker interval,
The head measurement method according to claim 1, wherein there are three or more combinations of markers such that the marker interval is equal to or greater than the head vertical width. In the head image, the area of the marker formation region defined by connecting the plurality of markers with straight lines is Sm,
In the head image, when the area of the entire head area defined by the outer contour of the head is Sh,
The head measurement method according to claim 1, wherein Sm / Sh is 0.25 or more. In the head image, when the centroid Gh of the outer contour line of the head is determined, and the head image is divided into four by the straight lines L1 and L2 orthogonal to each other with the centroid Gh as an intersection,
4. The head measurement method according to claim 1, wherein at least one of the markers is arranged in each of two of the four sections in a diagonal position. 5.   The head measurement method according to claim 1, wherein three or more of the markers are arranged outside the face surface.