Table tennis ball with markings that allow ball rotation to be detected
The table tennis ball with a specific mark point distribution allows for fast and reliable spin detection by ensuring global homogeneity and local irregularity, addressing the challenges of real-time spin determination in table tennis.
Patent Information
- Application Number
- JP2023537625
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2021-03-23
- Filing Date
- 2022-03-23
- Publication Date
- 2024-11-14
- Estimated Expiration
- 2042-03-23
AI Technical Summary
Existing methods for automatically determining the spin of a table tennis ball during flight are too slow and unreliable due to the high frequency of strokes and unpredictable orientation of the ball, making real-time spin determination impossible.
A table tennis ball with a specific distribution of mark points on its surface, characterized by a standard deviation of great circle paths between each point and its nearest neighbors, ensuring global homogeneity and local irregularity, allowing for fast and accurate spin detection.
Enables fast and reliable automatic spin determination of table tennis balls in near real-time, overcoming the limitations of previous methods by ensuring accurate orientation recognition and reduced numerical effort in image processing.
Smart Images

Figure 0007583940000016 
Figure 0007583940000017 
Figure 0007583940000018
Abstract
Description
[Technical field]
[0001] The present invention relates to a table tennis ball (hereinafter also referred to as "ball") according to the preamble of claim 1, which has a spherical ball surface and marks on the ball surface for making the ball rotation measurable and detectable, the marks including a predetermined number of mark points. Such a table tennis ball is known, for example, from WO 2020 / 096120. [Background technology]
[0002] In addition to the ball speed, the ball rotation (also called "spin") is very important in table tennis, since it has a decisive influence on the flight path of the ball and its rebound on the table and on the opponent's racket. Therefore, in particular in table tennis, there is a strong demand for coaches, for example, to determine the spin during flight as accurately as possible for the objective control and purpose-adaptive improvement of the stroke technique of table tennis players. Furthermore, in table tennis matches, there is a demand to measure and analyze the movement of the table tennis ball, preferably in real time, as part of the match hosting or sports reporting.
[0003] It is commonly known that a mark is provided on the surface of a table tennis ball in order to visually recognize the spin of the ball during flight. For example, the German Wikipedia entry for "table tennis ball" (https: / / de.wikipedia.org / wiki / Tischtennisball, version at 20:51 on April 17, 2020) depicts a table tennis training ball. A color pattern with a large color area is printed on the surface of the ball as a mark for spin recognition.
[0004] On the other hand, in the table tennis ball known from WO 2020 / 096120, the mark provided for spin detection comprises six mark points homogeneously distributed over the ball surface, each of which has a diameter between 5 mm (millimeters) and 13 mm.
[0005] Similar to table tennis, golf also requires measuring the ball rotation during flight. Here too, marks, in particular dot patterns, provided on the ball surface are used to enable measuring the ball rotation during the tee shot. Corresponding golf balls are known, for example, from CN 107543530, KR 102101512, US 2018 / 0353828, and US 7062082. The marks disclosed herein each have one or more groups with a number of closely adjacent mark points.
[0006] Alternatively, according to CN 106643662, a mark having a one-dimensional or two-dimensional structure, such as a mark line, is used to recognize the spin of the golf ball.Further alternatively, JP 2016-218014 A proposes using a manufacturer's mark on the golf ball as a mark for making the spin detectable.
[0007] Although automatic methods for measuring the spin of golf balls have already been successfully developed, no corresponding measurement method has yet been established for table tennis. The main reason is that the speed of conventional measurement methods is too slow to be effectively used in table tennis. Compared to golf, table tennis is a sport with a high frequency of strokes, and when the ball is hit, it usually passes over the net more than once per second. In addition, unlike golf, there is also the difficulty that the position and orientation of the ball are not known at the start of the measurement. While spin measurement in golf is usually performed on a tee shot where the ball is stationary before (so that the position and orientation of the ball are precisely defined), the spin of a table tennis ball is measured in flight (usually when it crosses the net). However, since the trajectory of the table tennis ball can only be predicted a priori with poor accuracy, a measurement device for calculating the spin of a table tennis ball must first find the ball before measuring the spin, which requires a considerable measurement time. Furthermore, since the orientation of the table tennis ball when it is detected by the measurement device is not known in advance, its orientation must first be determined. For this reason, up until now, measuring the spin of a table tennis ball in flight was only possible under favorable conditions and only with a long time delay compared to the course of the game. A reliable spin determination in all game situations (especially in approximately real time, i.e. approximately the time between two strokes) was not yet available. Summary of the Invention [Problem to be solved by the invention]
[0008] The object of the invention is to enable an automatic, particularly fast and more reliable determination of the ball rotation (spin) during the flight of a table tennis ball. [Means for solving the problem]
[0009] This problem is solved according to the invention by the features of claim 1. Thereby, a table tennis ball is presented with a spherical ball surface (spherical surface) and marks on the ball surface to make the ball rotation (spin) measurable and detectable. As known from WO 2020 / 096120, the marks comprise a predefined number of mark points. The number of mark points is hereinafter referred to by the variable N. The radius of the table tennis ball is hereinafter referred to by the variable R.
[0010] Preferably, the marks relevant for making it possible to detect the spin consist only of mark points. In this case, the marks are in other words pure dot patterns with no other structures than the mark points. In a variant of the invention, however, the table tennis ball may have further structures or patterns on its ball surface. Such further structures, however, do not belong to the marks relevant for the spin determination and are not evaluated during the spin determination.
[0011] According to the present invention, the distribution of the mark points on the surface of the ball is characterized by a number of criteria.
[0012] According to the first criterion, the standard deviation of the great circle path length (great circle distance) between each marked point and its three nearest neighbors is at least 12% of the average great circle path length, so that the point distances of all marked points to their respective three nearest neighbors are taken into account when calculating the standard deviation.
[0013] According to the second criterion, the minimum length of the great circle path between each marked point and its three nearest neighbors is at least 40% of the average of the lengths of these great circle paths. In other words, according to the second criterion, no great circle path between one of the marked points and one of its three nearest neighbors has a length that is less than 40% of the average of the lengths of these great circle paths.
[0014] In addition to or as an alternative to the second criterion, the point distribution is characterized by the fact that the minimum length of the great circle path between each marked point and its three nearest neighbors is at least 120% of the ball radius divided by the square root of the number of points. In other words, according to the third criterion, no great circle path between one of the marked points and one of its three nearest neighbors has a length less than 120% of the ball radius divided by the square root of the number of points.
[0015] For the second and third criteria, respectively, the point distances between every mark point and its three nearest neighbors are also taken into account.
[0016] The invention is based on the realization that the feasibility of an error-free and numerically sufficiently simple, and therefore fast, determination of the spin of a table tennis ball essentially depends on the form of the marks provided on the table tennis ball.
[0017] As a first step, the invention is based on the consideration that marks with one- or two-dimensional structures are disadvantageous for automatic spin determination, since such structures are numerically burdensome and therefore can only be detected by relatively time-consuming image recognition methods (in particular by segmentation). Point-like structures, on the other hand, can be detected with very little numerical burden. The invention therefore focuses on the development of marks that consist of or at least contain a certain number of mark points.
[0018] The homogeneous distribution of the mark points proposed by WO 2020 / 096120 is recognized as a disadvantage, since the orientation of the ball cannot be clearly recognized from the photographic image of the table tennis ball. Rather, in the case of the ball proposed there, there are six different orientations, which themselves represent marks, whereby they are indistinguishable in the photography of the ball. This redundancy of marks can lead to a comparative evaluation of time-successive images of the ball, in which mark points recognizable by different images are incorrectly assigned to one another, which can lead to errors in the spin calculation.
[0019] It has also been found that the grouping of mark points proposed by CN 107543530, KR 102101512, US 2018 / 0353828, and US 7062082 is also not suitable for determining the spin in table tennis. This is mainly due to the fact that, unlike a tee shot in golf, in table tennis the orientation of the ball is not known at the start of the measurement detection. Since the table tennis ball can enter the detection range of the measuring device in any orientation, many of the measurement points concentrated in the group are not located on the side of the ball facing the measuring device at the time of detection with a relatively high probability. Therefore, the measurement points cannot be imaged well or at all. Under such circumstances, conventional spin determination based on evaluation of time-series images of the ball can often fail. It has been recognized that this problem can be solved by a denser arrangement of the point groups on the ball surface. However, this increases the number of mark points and their dense packing increases the numerical burden on identifying the mark points, resulting in reduced accuracy of spin determination.
[0020] Considering the shortcomings of the prior art, the present invention adopts a compromise. The present invention is based on the idea of distributing the mark points as uniformly as possible globally (i.e., in relation to the whole set of mark points), but as non-uniformly and irregularly (pseudo-randomly) as possible locally (i.e., in relation to each individual mark point and its immediate neighbors).
[0021] The global homogeneity of the point distribution allows easy recognition of a certain large subgroup of mark points in all possible directions of the ball, even if the mark is composed of only a relatively small number of mark points overall. This ensures a numerically undemanding identification of the detected mark points. However, the local non-homogeneity of the point distribution simplifies the identification of the individual mark points considerably, since the individual mark points can be easily, unambiguously and reliably recognized based on their arrangement in relation to neighboring points.
[0022] The above-mentioned properties, i.e. the global homogeneity on the one hand and the local inhomogeneity on the other hand, of the point distribution of the marks according to the invention, can be particularly simply and unambiguously described on the basis of the lengths of the great circle paths between each mark point and its three nearest neighbors, the term "great circle path" meaning in this case the shortest connection between the respective considered mark points on the spherical ball surface.
[0023] The "closest neighbor" of the considered mark point ("center point") always means the mark point connected to the center point via the great circle path of the shortest length. The "second closest neighbor" and the "third closest neighbor" of the center point always mean the mark point connected to the center point via the second shortest or third shortest great circle path, respectively. According to this definition, each mark point has a close neighbor, a second closest neighbor and a third closest neighbor, and the center point can be represented according to this definition. Preferably, the marks are then selected such that for each center point, the three closest neighbors are located at different distances in pairs. Thereby, these neighbors can be unambiguously assigned to the respective center point as the close, second (2nd) closest and third (3rd) closest neighbor. However, within the scope of the present invention, it is also possible that one or more mark points are equidistant from two of their three closest neighbors. In other words, in each case the first nearest neighbor and the second nearest neighbor, and / or the second nearest neighbor and the third nearest neighbor may be equidistant from the central point.
[0024] In the following, the group of mark points formed by each considered mark point (central point) and its three nearest neighbors is also called a “four-point network”. A mark has a predefined number of such four-point networks, which corresponds to the number of mark points.
[0025] For terminological simplicity, the lengths of the great circle paths are hereinafter referred to as "point distances." Each point distance is measured along the ball surface corresponding to the run of the associated great circle path.
[0026] In what follows, the length of the great circle path connecting each center point of the considered four-point network with one of its three nearest neighbors is called the "center-point distance" and is denoted by the formula symbol Z i,jThe variable i (where i=1,2,...,N) then denotes the mark point which is to be considered as the center point in the respective relation. The variable j (where j=1,2,3) denotes the nearest neighbor, the second nearest neighbor, or the third nearest neighbor of the i-th center point. In this sense, for example, the formula symbol Z 5,2 denotes the center-point distance between the fifth mark point, considered as the center point, and its second nearest neighbor.
[0027] The consideration of the four-point network proves to be particularly advantageous when the mark points are distributed more or less homogeneously, in particular because it is recognized that the three nearest neighbors form a more or less well-defined and closed tentative shell around each central point, characterized in that they have similar point distances to each other, which are more or less clearly different from the point distances of the more distant neighbors, depending on the number of points and the homogeneity of the pattern.
[0028] The above first criterion for point distribution, i.e., that the standard deviation of the great circle path length between each mark point and its three nearest neighbors (hereinafter, σ) is at least 12% of the average great circle path length (hereinafter, μ), can therefore be stated as follows:
[0029]
number
number
number
[0030] Preferably, the center-point distance Z i,j The standard deviation σ of each of the marks P is at least 15% of the associated mean value μ. As can be seen from Equations 2 and 3, when calculating the standard deviation σ and the mean value μ, all the marks P iThe distance Z between the three nearest (center-) points i,j , i.e. Z 1,1 ,Z 1,2 ,Z 1,3 ,Z 2,1 ,Z 2,2 ,Z 2.3 ,…,Z N,1 ,Z N,2 ,Z N,3 are summed up.
[0031] A low standard deviation σ value means that at least most of the mark points P i For each considered mark point P, the three nearest neighbors are considered as the center point. i or, in other words, at least most of the mark points P i is located at a very similar distance from its three nearest neighbors. Therefore, the characteristic quantities σ and σ / μ respectively represent measures of the homogeneity of the point distribution. The above minimum standard deviation ensures that the point distribution in the four-point network does not become too homogeneous.
[0032] The second criterion of point distribution mentioned above, i.e., the minimum length of the great circle path between each marked point and its three nearest neighbors, i.e., the minimum (center-) point distance d on the ball surface. min The criterion that is at least 30% of the average length of the great circle path can be stated as follows:
[0033]
number
[0034] Preferably, this minimum (i.e., smallest) point distance d min is further at least 50%, in particular at least 60%, of the associated mean value μ. As can be seen from equation 4, the second criterion is that all the mark points P i The distance Z between the three nearest (center-) points of each i,j This also applies to Z 1,1 ,Z 1,2 ,Z 1,3,Z 2,1 ,Z 2,2 ,Z 2,3 ,…,Z N,1 ,Z N,2 ,Z N,3 This applies to:
[0035] The second criterion for this point distribution is to exclude mark points with too small distances between them, thereby eliminating clumping, especially of mark points, thereby ensuring an approximate global homogeneity of the point distribution.
[0036] A further measure of the global homogeneity of the point distribution is the minimum (center-) point distance d with respect to the ball radius R. min If the N mark points on the ball surface are distributed as uniformly as possible, the mark point P i are on average as far apart as possible from each other; that is, point pairs with extremely small point distances (i.e. point groups of closely spaced mark points) are, in other words, avoided. As an approximation, it can be seen that in the best homogeneous distribution, each mark point is located in a different 1 / Nth part of the ball surface, so that each mark point is more or less centrally located on its assigned subsurface of the ball surface on which no other mark points are located, each of these subsurfaces covering an area A of 1 / Nth of the ball surface. p has.
[0037]
number
[0038] Now, assuming that these partial surfaces are nearly flat circular surfaces, we can use Equation 5 and the formula for the area of a circle (A p ≒0.25 π d G 2 ) and the diameter of the partial surface d G can be estimated.
[0039]
number
[0040] Therefore, the diameter d G is a measure of the typical distance from the uniformly distributed mark points to their respective neighbors on the ball surface. As can be recognized from Equation 6, this typical distance decreases approximately by the square root of the number of points N. Since the size of the ball surface is limited, it is usually not possible to increase the distance between the mark points without decreasing the distance between the points to other points on the ball surface. Therefore, the diameter d G The minimum point distance d to min The ratio of is a measure of how much the point distribution deviates from global homogeneity.
[0041] According to the third criterion above, the minimum point distance d min is the typical distance d calculated according to Eq. G or, equivalently, at least 120% (i.e., 1.2 times) the ball radius R divided by the square root of the number of points N.
[0042]
number
[0043] Preferably, the minimum point distance d min is the typical distance d G is at least 37.5%, in particular at least 45%. Equivalently, the minimum point distance d min is preferably at least 150% (i.e. 1.5 times) and in particular at least 180% (i.e. 1.8 times) of the ball radius R divided by the square root of the number of points N. In a pattern with 18 mark points (N=18), the minimum point distance d min is therefore at least 28%, preferably at least 35%, in particular at least 42% of the ball radius R.
[0044] As stated above, the second and third criterion are considered within the scope of the present invention either alternately with one another or in combination with one another, but always in combination with the first criterion, for characterizing a ball pattern according to the present invention.
[0045] The length of the great circle path connecting the three nearest neighbors is the center-point distance Z i,j In distinction from "perimeter-point distance" Q i,k In this case, the variable i again indicates the mark point to be considered as the center point, while the variable k (where k=1, 2, 3) is expressed as follows:
[0046] For k=1, the point distance between the nearest neighbor and the second nearest neighbor, For k=2, the point distance between the second nearest neighbor and the third nearest neighbor, When k=3, it represents the point distance between the third nearest neighbor and the nearest neighbor of the i-th mark point. In this sense, the formula symbol Q 5,2 denotes the perimeter-point distance between the second nearest neighbor and the third nearest neighbor of the fifth mark point considered as the center point.
[0047] In a preferred embodiment of the present invention, the mark points are distributed on the ball surface according to a fourth criterion, namely, the perimeter-point distance Q of each four-point network. i,k The difference in length (hereinafter ΔQ i ) is, in other words, the difference between the longest and shortest great circle paths between the three nearest neighbors of each marked point is the center-point distance Z i,j The mark points are distributed so that the average value μ of is greater than 30%.
[0048] For each i=1,2,...,N
number
[0049] This fourth property of the point distribution according to formula 8 can advantageously be used in a table tennis ball according to the preamble of claim 1 in order to characterize a particularly preferred point distribution, either in combination with or apart from one of the features of the markings described above, and in this respect is also considered as an independent invention. i is greater than 35% of the mean value μ, and in particular even greater than 40%.
[0050] Difference length ΔQ i The preferred lower limit of this difference length ΔQ i describes how much the triangle formed by the three nearest neighbors of a four-point network deviates from an equilateral triangle and thus from a locally homogeneous point distribution. As a result, the difference length ΔQ i A sufficiently large value of is recognized to be a particularly meaningful measure of the desired local heterogeneity of the point distribution.
[0051] In a preferred embodiment of the invention, the number of mark points is between 13 and 25, preferably between 16 and 21, in particular 18 or 19. This is based on experience that a number of 18 or 19 mark points distributed on the ball surface allows an optimized spin detection in terms of numerical effort and reliability.
[0052] For marks with a small number of mark points, the situation occurs more frequently where the mark points are not visible enough in the image of the ball to clearly identify recognizable mark points, which can lead to inaccurate or erroneous spin calculations.
[0053] On the other hand, as the number of mark points increases, the number of possibilities that must be compared to identify a group of measured and detected mark points increases disproportionately, so that marks using 19 or more mark points impose a large numerical burden on spin determination.
[0054] However, the intended benefits of the present invention can be achieved, albeit to a lesser extent, with marks having 16, 17, 20 or 21 mark points, and even less so with marks including 13-15 or 22-25 mark points.
[0055] Preferably, the diameter of each mark point is selected to be 10% to 24%, preferably 15% to 20%, in particular 17.5% of the ball radius. The diameter of each mark point, in absolute terms, is preferably 2.0 mm to 4.8 mm, preferably 3.0 mm to 4.0 mm, in particular 3.5 mm. This dimension of the mark points has proven to be particularly advantageous, since, on the one hand, the mark points are sufficiently large to be reliably detectable on the image of the ball even under unfavourable conditions (e.g. when the ball is relatively far from the measuring device). On the other hand, the mark points dimensioned in this way are sufficiently small that in a useful approximation they can be treated as point-like (i.e. dimensionless) structures, which allows a numerically simple image evaluation. Furthermore, in any embodiment, the identical shape and size of all the mark points is advantageous for a simple image evaluation.
[0056] In an advantageous embodiment, the marking points are characterized by infrared absorption or infrared re-emission properties that are different from the rest of the ball surface. The marking points therefore stand out in contrast to the rest of the ball surface in the infrared (IR) range of the electromagnetic radiation spectrum, in particular in an infrared image of the table tennis ball. In particular, the marking points are provided on the ball surface by an infrared-sensitive lacquer (wherein this infrared-sensitive lacquer is used only for the marking points and not for other marks or patterns possibly present on the ball surface). This feature allows automatic spin determination by photographing the ball with an infrared camera that is particularly insensitive to disturbances in the visible light spectral range (e.g. dazzling light from headlights or flashlights). Furthermore, the contrasting infrared absorption or infrared re-emission properties also allow selective photographing of the marks for spin determination purposes. In addition to the marks, the ball can thereby be freely provided with further patterns or printings in areas that do not interfere with spin determination.
[0057] In a further embodiment of the invention, the markings are made in such a way that the marking points stand out in contrast from the background of the rest of the ball surface by using a predefined common color in the visible region of the electromagnetic radiation spectrum. This feature allows selective photographing of the markings for spin determination purposes, for example by taking a color-filtered photograph of the table tennis ball or by subsequent image processing. This feature therefore also allows the ball to be provided with further patterns or printing (in one or more other colors) without interfering with the spin determination.
[0058] In a particularly advantageous embodiment, the table tennis ball is provided with a mark having 18 mark points. In the following, these mark points are collectively referred to as P i (where i = 1, 2, ..., 18). Each of the 18 mark points is compared to the radius R of the ball and the center-point distance Z to its nearest neighbors, detailed below. i,j They are located at.
[0059] [Table 1] TIFF0007583940000010.tif255166TIFF0007583940000011.tif99166
[0060] Mark point P according to Table 1 i may be used advantageously in combination with one of the features of the marking described above, or separately therefrom, in a table tennis ball according to the preamble of claim 1, and in this respect may also be regarded as an independent invention.
[0061] Preferably, the neighbors of each 4-point network have a perimeter-point distance Q i,k They are located at.
[0062] [Table 2] TIFF0007583940000013.tif255167TIFF0007583940000014.tif148170
[0063] Hereinafter, embodiments of the present invention will be described in more detail with reference to the drawings. [Brief description of the drawings]
[0064] [Figure 1] FIG. 1 is a plan view, from one of six mutually orthogonal viewing directions, of a table tennis ball having a spherical ball surface and markings thereon to enable measurement and detection of the spin of the ball during flight, the markings being formed from 18 mark points distributed on the ball surface. [Diagram 2] FIG. 2 is a plan view of the table tennis ball from another one of six mutually orthogonal viewing directions. [Diagram 3] FIG. 2 is a plan view of the table tennis ball from another one of six mutually orthogonal viewing directions. [Figure 4]FIG. 2 is a plan view of the table tennis ball from another one of six mutually orthogonal viewing directions. [Diagram 5] FIG. 2 is a plan view of the table tennis ball from another one of six mutually orthogonal viewing directions. [Figure 6] FIG. 2 is a plan view of the table tennis ball from another one of six mutually orthogonal viewing directions. [Figure 7] FIG. 2 shows a subgroup of marks ("four-point network") formed by a first mark point of a mark taken as the central point and its three closest neighbors in a representation according to FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0065] In all the figures, parts, sizes and structures that correspond to each other are always given the same reference numbers.
[0066] In Fig. 1 to Fig. 6, a (table tennis) ball 2 is shown in six plan views from six mutually orthogonal viewing directions. The spatial relationships of the views shown in Fig. 1 to Fig. 6 are represented in these figures by arrows respectively labelled with Roman numerals. The Roman numerals respectively indicate the viewing direction underlying the figure corresponding to the number. Thus, for example, the arrow indicated by the Roman numeral II in Fig. 1 indicates the viewing direction from which the view according to Fig. 2 is obtained.
[0067] The ball 2 has, in a typical embodiment, a hollow spherical shell 4, in particular made of plastic. The ball 2 has in particular a ball radius R of 20 mm (millimeters) or a ball diameter of 40 mm, which complies with the rules of table tennis.
[0068] In order to be able to measure the ball rotation (spin) during the flight of the ball 2, the spherical ball surface 6 of the outer shell 4 (and thus the entire ball 2) is provided with marks 8. In the illustrated embodiment, the marks 8 are provided at 18 mark points P i(where i=1, 2, ..., 18). For convenience of explanation, the marks 8 are shown in schematic form in Figs. 1 to 6. Note that the mark points P i The perspective distortion is not shown.
[0069] Each mark point P i is formed by a circular area with a diameter d (Fig. 7) of 3.5 mm (corresponding to 17.5% of the ball radius R). i is then provided on the ball surface 6 by means of a strongly IR-absorbing and / or IR-reemitting lacquer, a strongly IR-absorbing and / or IR-reemitting printing ink / ink (see for example DE 10 2008 049 595 A1) or another strongly IR-absorbing and / or IR-reemitting coating. i may then have a color that forms a color contrast with the surrounding ball surface 6. In this case, the mark point P i is also visually distinct from the surrounding ball surface 6. However, preferably, the mark point P i is transparent to the human eye or has the same or similar color as the surrounding ball surface 6. In the latter case, the mark point P i is clearly visible only in an infrared image of the ball 2, but is invisible, or at least unnoticeable, to a human observer and in photographs of the ball 2.
[0070] By the above-mentioned IR-sensitive implementation of mark 8, mark point P iwill stand out in strong contrast from the rest of ball surface 6 in an IR image of ball 2. Optionally, ball 2 is additionally provided with an additional structure (e.g., a print or pattern) that is provided on ball surface 6 by at least one non- or low-IR-absorbency ink. Mark 8 also stands out strongly against this possibly present additional structure in the IR image. As a result, in the reverse inference, the possibly present structure does not detract from or falsify the information content conveyed by mark 8 regarding the orientation of ball 2 in space.
[0071] Mark point P on the sphere surface 6 i The position of in spherical coordinates, i.e. by polar angle θ and azimuthal angle φ, is given in the table below:
[0072] [Table 3]
[0073] In this arrangement, the mark point P i are, in a global perspective and in a rough approximation, uniformly distributed on the spherical surface 6. However, for an ideal uniform distribution, the mark points P i are pseudo-randomly offset in a recognizable manner.
[0074] The properties of such point distribution become particularly clear when considering a local point environment called a four-point network 10. The four-point network 10 has a mark point P i and its three nearest neighbors P i The mark point P is formed by i Depending on the number of 4-points, the mark 8 can be divided into 18 overlapping 4-point networks 10.
[0075] A four-point network 10 for a first mark point P1 is shown as an example in FIG. The nearest neighboring point 14 of the mark point P1 is determined by the mark point P2 as follows: The second nearest neighbor 16 of the mark point P1 is determined by the mark point P5 as follows: The third nearest neighbor 18 of the mark point P1 is the mark point P 17 By It is recognized that a structure is formed.
[0076] As defined above, the central point 12 of each four-point network 10 (in the example shown in FIG. 7, the mark point P1) and its adjacent points 14, 16, and 18 (i.e., in the example shown in FIG. 7, the mark points P2, P5, and P6, respectively) are 17 ) and the length of the great circle path 20 connecting the center point distance Z i,j On the other hand, the length of the great circle path 20 connecting the neighboring points 14, 16, and 18 of each four-point network 10 is the perimeter-point distance Q i,k For the four-point network 10 of the first mark point P1, the corresponding center-point distance Z 1,1 ,Z 1,2 ,Z 1,3 , and the perimeter-point distance Q 1,1 ,Q 1,2 ,Q 1,3 is entered in Figure 7.
[0077] The 18 mark points P shown in Table 3 i For the configuration, the center-point distance Z i,j The values of are taken from the right column of Table 1.
[0078] According to Equation 3, the center-point distance Z i,j The average value μ of each mark point P corresponds to 0.80 times the ball radius R (i.e. μ / R=0.80). If the ball radius R is 20 mm, i The average point distance from its three nearest neighbors is thereby 16.1 mm (i.e. μ = 16.1 mm). For the standard deviation σ according to Equation 2, a value of 0.126 was obtained for the point distribution in Table 3. Thus, the standard deviation σ normalized to the mean value μ is 15.7% (i.e. σ / μ = 15.7%). The minimum (center-) point distance d minis point P9 and point P 14 , which is 10.1 mm for the above ball radius. This corresponds to 63% of the average value μ (i.e., d min =min{Z i,j |i=1,...,N;j=1,2,3}=63%·μ).
[0079] minimum point distance d min is the representative distance d calculated according to Eq. G (6), or, equivalently, at least 214% of the quotient of the ball radius R divided by the square root of the number of points N, or 50.5% of the ball radius R.
[0080] For the characteristic quantity σ / μ, the point distribution in Table 3 is the most ideal mark point P i , and there are 18 such mark points P i The comparative value of σ / μ=10.1% for a homogeneous distribution of σ / μ was confirmed by tests. On the other hand, the point distribution shown in Table 3 also differs from the average random point distribution in that the standard deviation σ normalized by the mean value μ is significantly larger. However, the minimum center-point distance regularly does not meet the condition according to Eq. 4.
[0081] Perimeter-point distance Q of point arrangement in Table 3 i,k is obtained from the right column of Table 2. The difference length ΔQ of this point arrangement normalized by the mean value μ according to Equation 8 i is 42.8% or more in all four-point networks10 (min{ΔQ i / μ;i=1,2,...,18}=ΔQ 10 / μ=42.8%).
[0082] This characteristic quantity ΔQ indicates a measure of the local irregularity of the point distribution. i The minimum value of is the ideal uniform distribution or the 18 mark points P i is significantly larger than the average random placement case.
[0083] Thereby, the point distribution according to table 3 is distinguished by a remarkable homogeneity globally, but by a particularly high degree of irregularity locally. Both features, in combination, result in a mark point P i This facilitates clear identification of the spin, thereby providing a decisive advantage in determining the spin.
[0084] To determine the spin of the ball 2, a time sequence of images of the ball 2 in flight is taken by a camera, whereby a mark point P i In order to be able to take advantage of the strong contrast between the two, an infrared camera is preferably used. The camera is preferably positioned to the side of the table at net height, whereby the camera is oriented parallel to the transverse direction of the table.
[0085] The captured images of the ball 2 are evaluated, for example using the method known from US Pat. No. 7,062,082, in order to determine the spin (in particular with regard to the axis of rotation, direction of rotation and speed of rotation of the ball 2).
[0086] The mark 8 according to the invention, in particular in the embodiment shown in the figures and detailed in table 3, is numerically particularly undemanding and thereby allows a particularly fast spin determination, which for the first time allows the judicious use of automatic spin determination methods in table tennis, in particular in quasi-real time.
[0087] The invention is particularly evident from the above-described embodiments, but is not limited thereto. Further embodiments of the invention can be derived from the claims and the above description. [Explanation of symbols]
[0088] 2. Ball 4. Outer shell 6. Ball Surface 8 Marks 10 Four-point network 12 center point 14 (closest) neighbors 16 (second nearest) neighbors 18 (3rd nearest) neighbors 20 Great Circle Path d diameter (of the mark point) P i Mark points (i=1,2,...,18) Z i,j Center-point distance (i=1,2,...,18;j=1,2,3) Q i,k Perimeter-Point Distance(i=1,2,...,18;k=1,2,3) R ball radius
Claims
1. A table tennis ball (2) having a spherical ball surface (6) and marks (8) provided on said ball surface (6) for enabling the ball rotation to be measured and detected, The mark (8) is a predetermined number (N) of mark points (P i In the table tennis ball (2), The mark point (P i ) on the ball surface (6), Each of the mark points (P i ) and its three nearest neighbors (14, 16, 18). i,j The standard deviation of the length (Z i,j ) is at least 12% of the average value (μ) of Each of the mark points (P i The minimum length of the great circle path (20) between the point (Z ) and its three nearest neighbors (14, 16, 18) is the length (Z i,j ) is at least 40% of said average value (μ) of said mark points (P i ) by the square root of the number (N) of A table tennis ball (2) characterized by being distributed.
2. The mark point (P i ) on the ball surface (6), Each of the mark points (P i ) and its three nearest neighbors (14, 16, 18). i,j The standard deviation of the length (Z i,j 2. The table tennis ball (2) according to claim 1, wherein the average (μ) of the surface roughness (μ) is at least 15% of the average (μ).
3. Each of the mark points (P i The minimum length of the great circle path (20) between the point (Z ) and its three nearest neighbors (14, 16, 18) is the length (Z i,j 3. The table tennis ball (2) according to claim 1 or 2, wherein the average value (μ) of the surface roughness (μ) is at least 50% of the surface roughness (μ).
4. Each of the mark points (P i 4. The table tennis ball (2) according to claim 1, wherein the minimum length of the great circle path (20) between a point (11) and its three nearest neighbors (14, 16, 18) is at least 150% of the ball radius (R) divided by the square root of the number of points (N).
5. The ball has a spherical surface (6) and marks (8) provided on the surface (6) for enabling the ball rotation to be measured and detected, The mark (8) is a predetermined number of mark points (P i ), The mark point (P i ) on the ball surface (6), Each of the mark points (P i The length (Q i,k ) difference length (ΔQ i ) is each of the mark points (P i ) and its three nearest neighbors (14, 16, 18). i,j The table tennis ball (2) according to any one of claims 1 to 4, wherein the average value (μ) of ...
6. The mark point (P i 6. The table tennis ball (2) according to any one of claims 1 to 5, wherein the number of ribs (13) is between 13 and 25.
7. Each of the mark points (P i The table tennis ball (2) according to any one of claims 1 to 6, wherein the diameter (d) of the first vertex (12) is between 10% and 24% of the ball radius (R).
8. Each of the mark points (P i The table tennis ball (2) according to any one of claims 1 to 7, wherein the diameter (d) of the first spur (12) is between 2.0 mm and 4.8 mm.
9. All mark points (P i 9. The table tennis ball (2) according to any one of claims 1 to 8, wherein all of the first and second surfaces have the same shape and size.
10. The mark point (P i ) have infrared absorption and / or infrared re-emission properties different from the rest of the ball surface (6), so that the mark points (P i 10. The table tennis ball (2) according to any one of claims 1 to 9, wherein the first and second surfaces (22, 24) of the ball (10) are in contrast to each other.
11. The mark point (P i ) has a different color than the rest of the ball surface (6), so that the mark points (P i 11. The table tennis ball (2) according to any one of the preceding claims, wherein the first and second surfaces (21, 22) of the ball (10) are in contrast to each other.
12. A ball having a spherical surface (6) and marks (8) provided on the ball surface (6) for enabling the ball rotation to be measured and detected, The mark (8) includes a predetermined number of mark points (P i ); The mark points (P i ) are located on the ball surface (6) The table tennis ball (2) is distributed such that a difference length (ΔQ i ) of the lengths (Q i,k ) of the great circle paths (20) connecting each of the marked points (P i ) with its three nearest neighbors (14, 16, 18) is greater than the average value (μ) of the lengths (Z i,j ) of the great circle paths (20) between each of the marked points (P i ) and its three nearest neighbors (14, 16, 18).
13. The table tennis ball (2) according to claim 12, wherein the number of mark points (P i ) is 13 to 25.
14. A table tennis ball (2) as described in claim 12 or 13, wherein the diameter (d) of each of the mark points (P i ) is 10% to 24% of the ball radius (R).
15. A table tennis ball (2) as described in any one of claims 12 to 14, wherein the diameter (d) of each of the mark points (P i ) is between 2.0 mm and 4.8 mm.
16. A table tennis ball (2) as described in any one of claims 12 to 15, wherein all the mark points (P i ) have the same shape and size.
17. A table tennis ball (2) as claimed in any one of claims 12 to 16, wherein the mark points (P i ) have infrared absorption and / or infrared re-emission properties which differ from the remainder of the ball surface (6), such that the mark points (P i ) stand out in contrast from the remainder of the ball surface (6) in the infrared region of the electromagnetic radiation spectrum, in particular in an infrared image of the table tennis ball (2).
18. A table tennis ball (2) as claimed in any one of claims 12 to 17, wherein the mark points (P i ) have a different colour from the rest of the ball surface (6), whereby in the visible range of the electromagnetic radiation spectrum, in particular in a colour-filtered photograph of the table tennis ball (2), the mark points (P i ) stand out in contrast from the rest of the ball surface (6).
19. The ball has a spherical surface (6) and marks (8) provided on the ball surface (6) to enable ball rotation to be measured and detected, and the marks (8) are arranged at a predetermined number of mark points (P i ), The mark (8) has 18 mark points (P i ), The first mark point (P 1 ) as a starting point, Its nearest neighbor (P 2 ) the length (Z 1,1 ) to the ball radius (R) is 0.54±20%; Its second nearest neighbor (P 5 ) the length (Z 1,2 ) to the ball radius (R) is 0.91±20%; Its third nearest neighbor (P 17 ) the length (Z 1,3 ) to the ball radius (R) is 0.99±20%; The second mark point (P 2 ) as a starting point, Its nearest neighbor (P 1 ) the length (Z 2,1 ) to the ball radius (R) is 0.54±20%; Its second nearest neighbor (P 4 ) the length (Z 2,2 ) to the ball radius (R) is 0.72±20%; Its third nearest neighbor (P 5 ) the length (Z 2,3 ) to the ball radius (R) is 0.83±20%; The third mark point (P 3 ) as a starting point, Its nearest neighbor (P 10 ) the length (Z 3,1 ) to the ball radius (R) is 0.78±20%; Its second nearest neighbor (P 15 ) the length (Z 3,2 ) to the ball radius (R) is 0.84±20%; Its third nearest neighbor (P 8 ) the length (Z 3,3 ) to the ball radius (R) is 0.87±20%; The fourth mark point (P 4 ) as a starting point, Its nearest neighbor (P 2 ) the length (Z 4,1 ) to the ball radius (R) is 0.72±20%; Its second nearest neighbor (P 12 ) the length (Z 4,2 ) to the ball radius (R) is 0.84±20%; Its third nearest neighbor (P 5 ) the length (Z 4,3 ) to the ball radius (R) is 0.85±20%; The fifth mark point (P 5 ) as a starting point, Its nearest neighbor (P 12 ) the length (Z 5,1 ) to the ball radius (R) is 0.79±20%; Its second nearest neighbor (P 2 ) the length (Z 5,2 ) to the ball radius (R) is 0.83±20%; Its third nearest neighbor (P 4 ) the length (Z 5,3 ) to the ball radius (R) is 0.85±20%; The sixth mark point (P 6 ) as a starting point, Its nearest neighbor (P 12 ) the length (Z 6,1 ) to the ball radius (R) is 0.67±20%; Its second nearest neighbor (P 18 ) the length (Z 6,2 ) to the ball radius (R) is 0.73±20% Its third nearest neighbor (P 15 ) the length (Z 6,3 ) to the ball radius (R) is 0.94±20%; The seventh mark point (P 7 ) as a starting point, Its nearest neighbor (P 16 ) the length (Z 7,1 ) to the ball radius (R) is 0.59±20%; Its second nearest neighbor (P 17 ) the length (Z 7,2 ) to the ball radius (R) is 0.76±20%; Its third nearest neighbor (P 15 ) the length (Z 7,3 ) to the ball radius (R) is 0.90±20%, The eighth mark point (P 8 ) as a starting point, Its nearest neighbor (P 15 ) the length (Z 8,1 ) to the ball radius (R) is 0.82±20%; Its second nearest neighbor (P 3 ) the length (Z 8,2 ) to the ball radius (R) is 0.87±20%; Its third nearest neighbor (P 11 ) the length (Z 8,3 ) to the ball radius (R) is 0.89±20%; The ninth mark point (P 9 ) as a starting point, Its nearest neighbor (P 14 ) the length (Z 9,1 ) to the ball radius (R) is 0.51±20%; Its second nearest neighbor (P 18 ) the length (Z 9,2 ) to the ball radius (R) is 0.91±20%; Its third nearest neighbor (P 8 ) the length (Z 9,3 ) to the ball radius (R) is 0.95±20%; The tenth mark point (P 10 ) as a starting point, Its nearest neighbor (P 13 ) the length (Z 10,1 ) to the ball radius (R) is 0.75±20%; Its second nearest neighbor (P 17 ) the length (Z 10,2 ) to the ball radius (R) is 0.75±20%; Its third nearest neighbor (P 3 ) the length (Z 10,3 ) to the ball radius (R) is 0.78±20%; The eleventh mark point (P 11 ) as a starting point, Its nearest neighbor (P 13 ) the length (Z 11,1 ) to the ball radius (R) is 0.82±20%; Its second nearest neighbor (P 10 ) the length (Z 11,2 ) to the ball radius (R) is 0.88±20%; Its third nearest neighbor (P 8 ) the length (Z 11,3 ) to the ball radius (R) is 0.89±20%; The 12th mark point (P 12 ) as a starting point, Its nearest neighbor (P 6 ) the length (Z 12,1 ) to the ball radius (R) is 0.67±20%; Its second nearest neighbor (P 5 ) the length (Z 12,2 ) to the ball radius (R) is 0.79±20%; Its third nearest neighbor (P 4 ) the length (Z 12,3 ) to the ball radius (R) is 0.84±20%; The 13th mark point (P 13 ) as a starting point, Its nearest neighbor (P 10 ) the length (Z 13,1 ) to the ball radius (R) is 0.75±20%; Its second nearest neighbor (P 11 ) the length (Z 13,2 ) to the ball radius (R) is 0.82±20%; Its third nearest neighbor (P 1 ) the length (Z 13,3 ) to the ball radius (R) is 1.02±20%, The fourteenth mark point (P 14 ) as a starting point, Its nearest neighbor (P 9 ) the length (Z 14,1 ) to the ball radius (R) is 0.51±20%; Its second nearest neighbor (P 4 ) the length (Z 14,2 ) to the ball radius (R) is 0.92±20%; Its third nearest neighbor (P 11 ) the length (Z 14,3 ) to the ball radius (R) is 0.98±20%; The 15th mark point (P 15 ) as a starting point, Its nearest neighbor (P 8 ) the length (Z 15,1 ) to the ball radius (R) is 0.82±20%±20%, Its second nearest neighbor (P 3 ) the length (Z 15,2 ) to the ball radius (R) is 0.84±20%; Its third nearest neighbor (P 7 ) the length (Z 15,3 ) to the ball radius (R) is 0.90±20%, The 16th mark point (P 16 ) as a starting point, Its nearest neighbor (P 7 ) the length (Z 16,1 ) to the ball radius (R) is 0.59±20%; Its second nearest neighbor (P 17 ) the length (Z 16,2 ) to the ball radius (R) is 0.74±20%; Its third nearest neighbor (P 5 ) the length (Z 16,3 ) to the ball radius (R) is 1.10±20%, The seventeenth mark point (P 17 ) as a starting point, Its nearest neighbor (P 16 ) the length (Z 17,1 ) to the ball radius (R) is 0.74±20%; Its second nearest neighbor (P 10 ) the length (Z 17,2 ) to the ball radius (R) is 0.75±20%; Its third nearest neighbor (P 7 ) the length (Z 17,3 ) to the ball radius (R) is 0.76±20%; The 18th mark point (P 18 ) as a starting point, Its nearest neighbor (P 6 ) the length (Z 18,1 ) to the ball radius (R) is 0.73±20%; Its second nearest neighbor (P 12 ) the length (Z 18,2 ) to the ball radius (R) is 0.88±20%; Its third nearest neighbor (P 9 ) the great circle path (20) length (Z 18,3 ) to the ball radius (R) is 0.91±20% (2).
20. The first mark point (P 1 ) The nearest neighbor (P 2 ) and the second nearest neighbor (P 5 ) the length (Q 1,1 ) to the ball radius (R) is 0.83±20%; The second nearest neighbor point (P 5 ) to its third nearest neighbor (P 17 ) the length (Q 1,2 ) to the ball radius (R) is 1.57±20%; The third nearest neighbor point (P 17 ) to its nearest neighbor (P 2 ) the length (Q 1,3 ) to the ball radius (R) is 1.53±20%; The second mark point (P 2 ) The nearest neighbor (P 1 ) and the second nearest neighbor (P 4 ) the length (Q 2,1 ) to the ball radius (R) is 1.24±20%; The second nearest neighbor point (P 4 ) to its third nearest neighbor (P 5 ) the length (Q 2,2 ) to the ball radius (R) is 0.85±20%; The third nearest neighbor point (P 5 ) to its nearest neighbor (P 1 ) the length (Q 2,3 ) to the ball radius (R) is 0.91±20%; The third mark point (P 3 ) The nearest neighbor (P 10 ) and the second nearest neighbor (P 15 ) the length (Q 3,1 ) to the ball radius (R) is 1.61±20%; The second nearest neighbor point (P 15 ) to its third nearest neighbor (P 8 ) the length (Q 3,2 ) to the ball radius (R) is 0.82±20%; The third nearest neighbor point (P 8 ) to its nearest neighbor (P 10 ) the length (Q 3,3 ) to the ball radius (R) is 1.45±20%; The fourth mark point (P 4 ) The nearest neighbor (P 2 ) and the second nearest neighbor (P 12 ) the length (Q 4,1 ) to the ball radius (R) is 1.39±20%; The second nearest neighbor point (P 12 ) to its third nearest neighbor (P 5 ) the length (Q 4,2 ) to the ball radius (R) is 0.79±20%; The third nearest neighbor point (P 5 ) to its nearest neighbor (P 2 ) the length (Q 4,3 ) to the ball radius (R) is 0.83±20%; The fifth mark point (P 5 ) The nearest neighbor (P 12 ) and the second nearest neighbor (P 2 ) the length (Q 5,1 ) to the ball radius (R) is 1.39±20%; The second nearest neighbor point (P 2 ) to its third nearest neighbor (P 4 ) the length (Q 5,2 ) to the ball radius (R) is 0.72±20%; The third nearest neighbor point (P 4 ) to its nearest neighbor (P 12 ) the length (Q 5,3 ) to the ball radius (R) is 0.84±20%; The sixth mark point (P 6 ) The nearest neighbor (P 12 ) and the second nearest neighbor (P 18 ) the length (Q 6,1 ) to the ball radius (R) is 0.88±20%; The second nearest neighbor point (P 18 ) to its third nearest neighbor (P 15 ) the length (Q 6,2 ) to the ball radius (R) is 0.97±20%; The third nearest neighbor point (P 15 ) to its nearest neighbor (P 12 ) the length (Q 6,3 ) to the ball radius (R) is 1.57±20%; The seventh mark point (P 7 ) The nearest neighbor (P 16 ) and the second nearest neighbor (P 17 ) the length (Q 7,1 ) to the ball radius (R) is 0.74±20%; The second nearest neighbor point (P 17 ) to its third nearest neighbor (P 15 ) the length (Q 7,2 ) to the ball radius (R) is 1.60±20%, The third nearest neighbor point (P 15 ) to its nearest neighbor (P 16 ) the length (Q 7,3 ) to the ball radius (R) is 1.38±20%; The eighth mark point (P 8 ) The nearest neighbor (P 15 ) and the second nearest neighbor (P 3 ) the length (Q 8,1 ) to the ball radius (R) is 0.84±20%; The second nearest neighbor point (P 3 ) to its third nearest neighbor (P 11 ) the length (Q 8,2 ) to the ball radius (R) is 1.00±20%, The third nearest neighbor point (P 11 ) to its nearest neighbor (P 15 ) the length (Q 8,3 ) to the ball radius (R) is 1.59±20%; The ninth mark point (P 9 ) The nearest neighbor (P 14 ) and the second nearest neighbor (P 18 ) the length (Q 9,1 ) to the ball radius (R) is 1.39±20%; The second nearest neighbor point (P 18 ) to its third nearest neighbor (P 8 ) the length (Q 9,2 ) to the ball radius (R) is 0.96±20%; The third nearest neighbor point (P 8 ) to its nearest neighbor (P 14 ) the length (Q 9,3 ) to the ball radius (R) is 1.33±20%; The tenth mark point (P 10 ) The nearest neighbor (P 13 ) and the second nearest neighbor (P 17 ) the length (Q 10,1 ) to the ball radius (R) is 1.22±20%; The second nearest neighbor point (P 17 ) to its third nearest neighbor (P 3 ) the length (Q 10,2 ) to the ball radius (R) is 1.11±20%; The third nearest neighbor point (P 3 ) to its nearest neighbor (P 13 ) the length (Q 10,3 ) to the ball radius (R) is 1.45±20%; The eleventh mark point (P 11 ) The nearest neighbor (P 13 ) and the second nearest neighbor (P 10 ) the length (Q 11,1 ) to the ball radius (R) is 0.75±20%; The second nearest neighbor point (P 10 ) to its third nearest neighbor (P 8 ) the length (Q 11,2 ) to the ball radius (R) is 1.45±20%; The third nearest neighbor point (P 8 ) to its nearest neighbor (P 13 ) the length (Q 11,3 ) to the ball radius (R) is 1.71±20%; The twelfth mark point (P 12 ) The nearest neighbor (P 6 ) and the second nearest neighbor (P 5 ) the length (Q 12,1 ) to the ball radius (R) is 1.21±20%; The second nearest neighbor point (P 5 ) to its third nearest neighbor (P 4 ) the length (Q 12,2 ) to the ball radius (R) is 0.85±20%; The third nearest neighbor point (P 4 ) to its nearest neighbor (P 6 ) the length (Q 12,3 ) to the ball radius (R) is 1.51±20%; The thirteenth mark point (P 13 ) The nearest neighbor (P 10 ) and the second nearest neighbor (P 11 ) the length (Q 13,1 ) to the ball radius (R) is 0.88±20%; The second nearest neighbor point (P 11 ) to its third nearest neighbor (P 1 ) the length (Q 13,2 ) to the ball radius (R) is 1.84±20%; The third nearest neighbor point (P 1 ) to its nearest neighbor (P 10 ) the length (Q 13,3 ) to the ball radius (R) is 1.33±20%; The fourteenth mark point (P 14 ) The nearest neighbor (P 9 ) and the second nearest neighbor (P 4 ) the length (Q 14,1 ) to the ball radius (R) is 1.12±20%; The second nearest neighbor point (P 4 ) to its third nearest neighbor (P 11 ) the length (Q 14,2 ) to the ball radius (R) is 1.88±20%; The third nearest neighbor point (P 11 ) to its nearest neighbor (P 9 ) the length (Q 14,3 ) to the ball radius (R) is 1.05±20%, The fifteenth mark point (P 15 ) The nearest neighbor (P 8 ) and the second nearest neighbor (P 3 ) the length (Q 15,1 ) to the ball radius (R) is 0.87±20%; The second nearest neighbor point (P 3 ) to its third nearest neighbor (P 7 ) the length (Q 15,2 ) to the ball radius (R) is 0.90±20%, The third nearest neighbor point (P 7 ) to its nearest neighbor (P 8 ) the length (Q 15,3 ) to the ball radius (R) is 1.58±20%; The sixteenth mark point (P 16 ) The nearest neighbor (P 7 ) and the second nearest neighbor (P 17 ) the length (Q 16,1 ) to the ball radius (R) is 0.76±20%; The second nearest neighbor point (P 17 ) to its third nearest neighbor (P 5 ) the length (Q 16,2 ) to the ball radius (R) is 1.57±20%; The third nearest neighbor point (P 5 ) to its nearest neighbor (P 7 ) the length (Q 16,3 ) to the ball radius (R) is 1.67±20%; The seventeenth mark point (P 17 ) The nearest neighbor (P 16 ) and the second nearest neighbor (P 10 ) the length (Q 17,1 ) to the ball radius (R) is 1.47±20%; The second nearest neighbor point (P 10 ) to its third nearest neighbor (P 7 ) the length (Q 17,2 ) to the ball radius (R) is 1.21±20%; The third nearest neighbor point (P 7 ) to its nearest neighbor (P 16 ) the length (Q 17,3 ) to the ball radius (R) is 0.59±20%; The 18th mark point (P 18 ) The nearest neighbor (P 6 ) and the second nearest neighbor (P 12 ) the length (Q 18,1 ) to the ball radius (R) is 0.67±20%; The second nearest neighbor point (P 12 ) to its third nearest neighbor (P 9 ) the length (Q 18,2 ) to the ball radius (R) is 1.37±20%; The third nearest neighbor point (P 9 ) to its nearest neighbor (P 6 ) the length (Q 18,3 20. The table tennis ball (2) according to claim 19, wherein the ratio of the radius (R) of the ball to the radius (R) is 1.62±20%.
Citation Information
Patent Citations
Table tennis ball practice for seal -
JP1985195063U
Method and device for measuring rotation direction and amount of rotation in sphere traveling in space, and sphere suited for measuring rotation direction and amount of rotation when traveling in space
JP2006234485A
Golf ball and golf ball set
JP2008259721A
System and Method for Measuring Flight Information of Spherical Objects
JP2012517603A
Method and apparatus for ball flight analysis
US20180353828A1
