JPH05253324A - Motion analyzer for sports - Google Patents

Abstract

PURPOSE:To provide an apparatus to analyze velocity, direction, distance or the like of motion of players, balls and the like every moment based on a TV image signal. CONSTITUTION:A corresponding relationship of an image of a moving object is determined by a decision means 13c among field image data composing one frame of a TV image signal of the moving object obtained with an interlace scan image pickup tube type without shutter. Kinetic vectors of the moving body are calculated by a calculation means 13d based on the corresponding relationship determined and the field image data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sports motion analysis device, and more particularly to a sports motion analysis device suitable for simultaneously analyzing each motion state of a ball and a large number of players in soccer, rugby, American football and the like. Regarding

[0002]

2. Description of the Related Art Conventionally, it is often necessary to check the movements of a player and / or a ball in various sports in order to instruct a player. For this purpose, for example, in baseball, a speed gun is used to measure the speed of the ball, and in baseball or other competitions, time-dependent changes in the movements of players, etc. The method of examining is also widely adopted.

[0003]

However, although the speed gun is suitable for investigating the movement of a ball having a high speed, it is possible to analyze the movement of a large number of players such as soccer and rugby at a time. It is not suitable for such applications. On the other hand, analysis using disassembled photographs is used for such purposes, but it is necessary to take a number of consecutive photographs, and in order to analyze the movement of each of many athletes from these photographs. It requires a lot of man-hours and is not practical. The purpose of the present invention is therefore to provide a large number of players who are playing,
(EN) Provided is a sports motion analysis device capable of electrically analyzing the speed, direction, distance, etc. of a player, a ball, etc., at any given time based on a television image signal obtained by photographing the motion state of the ball, etc. Especially.

[0004]

A feature of the present invention for solving the above-mentioned problems is to provide a sports motion analysis apparatus for obtaining motion data of a player or a moving body such as a ball.
Each field constituting one frame of the television image signal in response to a television image signal including a two-dimensional motion image of a moving body in a required field of view obtained by a television camera of an interlaced scanning camera tube system without a shutter. Determining means for determining the correspondence between the images of the moving objects between the field image data by ANDing the image data, and the above-mentioned motion based on the correspondence relationship and each field image data determined by the determining means. And a calculation means for calculating each motion vector of the body.

[0005]

A pair of field image data including a two-dimensional motion image of a moving body obtained by photographing a moving body such as a player or a ball with a non-shutter interlaced scanning camera tube type television camera is used. Be done. One field image data is obtained based on the first field signal, for example, and the other field image data is obtained based on the second field signal, for example.
It can be obtained based on the field signal. The first field signal and the second field obtained by scanning one frame with this television camera are:
There is a time lag such that the scanning periods partially overlap. Therefore, the image of each moving body represented by the one field image data and the image of each moving body represented by the other field image data are the same as the moving length of the moving body corresponding to the temporal shift. It is obtained as two images shifted in the movement direction. As described above, each image of the same moving body in the pair of field images obtained by scanning with the non-shutter interlaced image pickup tube type television camera has an overlapping portion when configured as one frame image. Therefore, by finding the overlap between the images of the moving body, the images of the same moving body can be associated with each other between different field image data. This association is
This is performed based on the result of the logical product operation of the two field image data, and the calculation output means calculates the movement vector of the image of the corresponding pair of moving bodies as the movement vector.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings. FIG. 1 shows the configuration of an embodiment of a motion analysis apparatus 10 according to the present invention for analyzing the movements of a large number of players A to H and a ball I performing a ball game on the court 1. Motion analysis device 1
0 has a television camera 11 for obtaining the movements of the players A to H and the ball I on the court 1 as electric signals. In the illustrated embodiment, the television camera 11 is installed by being mounted on a ceiling (not shown) directly above the court 1 by appropriate means, and the players A to H, which are moving bodies, are seen from this position. And the two-dimensional movement of the ball I. However, the position of the television camera 11 is not limited to just above the court 1 and may be a position where the court 1 is viewed obliquely from above or from the side.

The television camera 11 is a 2: 1 interlace scanning type image pickup tube type television camera without a shutter, and outputs the state of a moving body captured by the television camera 11 as a corresponding video signal. Therefore, the image of one frame obtained by this is obtained by combining one set of field images. The relationship between these field images will be described below.

When the field of view captured by the television camera 11 is scanned by the interlace system, the field of view is first scanned by scanning the odd-numbered scanning lines to obtain an odd field signal. In this case, since there is no shutter,
The above-mentioned scanning is performed while each moving body is moving.
Next, scanning in the same field of view is performed by scanning along even-numbered scanning lines, and even-numbered field signals are obtained.

As shown in FIG. 6A, this interlaced scanning is performed in synchronization with each pulse P1, P2, ... Of the vertical synchronizing signal output every 1/60 second, in an odd field. Suppose that a scan of 1 is performed in time slot T1 of 1/30 second duration, the corresponding even field scan is also performed in time slot T2 of 1/30 second duration. Therefore, in this case, the time slots T1 and T2 overlap by 1/60 seconds.

The odd field signal thus obtained in the time slot T1 is applied to the time slot Ta of the video signal VS shown in FIG. 7B, while the even field signal obtained in the time slot T2 is This is applied to the time slot Tb of the video signal VS shown in FIG. 6B, and the television video signal VS is produced in this manner.

Returning to FIG. 1, the television video signal VS
Is input to the signal processing device 12, where it is separated into an odd field video signal VS1 and an even field video signal VS2. The signals VS1 and VS2 are sent to the computer 13
And image data processing for motion analysis of each moving body is performed in accordance with a predetermined processing program stored in the memory 14.

FIG. 2 is a functional diagram showing the contents of image data processing executed by the computer 13. The image data processing executed in the computer 13 will be described based on this functional diagram. Odd field video signal VS1 and even field video signal VS2
Is given to the first computing means 13a and the second computing means 13b, respectively, and the motion image data D1 and D2 for one frame according to the signals VS1 and VS2 are respectively obtained.

The motion image data D1 is, for example, as shown in FIG. 7A, every other scanning line ,.
, Are interpolated using the contents of the scanning line, ... As the contents of the scanning line ,. ) Is created as data indicating one frame of image information.

On the other hand, the motion image data D2 is the scanning lines included in the even field video signal VS2, ...
Based on the image information of, the interpolation similar to the interpolation shown in FIG. 7B is created as data indicating the image information of one frame.

Therefore, the motion images A1 to I1 in the motion image according to the motion image data D1 based on the image information of the odd field and the motion images A2 to I2 in the motion image according to the data motion image D2 based on the even field image information. As shown in FIG. 4, there is a slight deviation along the movement direction of each moving body, and only a part of them overlap.

The set of motion image data D1 and D2 thus obtained is given to the AND operation means 13c, where each motion image according to one motion image data D1 and the other motion image data D2. An AND operation is executed between each moving body image according to Then, the combination data indicating the combination of the moving body images in which the logical product image according to the logical product operation result exists is given to the motion vector calculation means 13d.

The motion vector calculation means 13d is further provided with motion image data D1 and D2, and the barycentric position of each moving body image is calculated based on these. Further, the motion vector of the position of the center of gravity in the pair of moving body images based on the combination data indicating the combination of the moving body images is calculated as the motion vector of the moving body (see FIG. 5).

In this way, the motion vector of each of the player and / or the ball of interest is calculated by the motion vector calculation means 13d, and the calculated motion vector data is used as the motion analysis data of the player and / or the ball. It is output from 13d.

Next, the computer 13 shown in FIG.
Based on the flowchart of the motion analysis program stored in the internal memory 14, the data processing for the motion analysis will be described in more detail.

When the motion analysis program 20 starts to be executed, first, in step 21, the odd field video signal VS1 and the even field video signal VS2 are taken into the computer 13, and in step 22, motion images according to these signals are obtained. Motion image data D1 and D shown
2 are created respectively.

The motion image data D1 corresponds to the odd field image shown by the solid line in FIG. 4, while the motion image data D2.
Also corresponds to the even field image shown by the dotted line in FIG. In the following description, the image represented by the motion image data D1 is referred to as the motion image at the first time, and the image represented by the motion image data D2 is referred to as the motion image at the second time.

Next, in step 23, data processing for binarizing the respective motion image data D1 and D2 is performed. The data processing for this binarization is performed by using a threshold value set appropriately,
This is performed by converting 2 into binary image data.
As a result, binarized data Da and Db indicating the binarized motion state of the moving body are obtained.

At the next step 24, unnecessary information removal, that is, noise removal processing is executed based on these binarized data Da and Db. This processing is noise removal processing on the outer periphery of the moving body image due to expansion and contraction. A process of removing an image having a number of pixels equal to or less than a preset number of pixels as noise. Is included. However, it goes without saying that other appropriate noise removal processing may be performed in addition to these or in place of some of them.

Next, in step 25, step 2
Labeling A1 and B on each moving body image in each moving image at the first time and the second time left as a result of the noise removal processing of No. 4
1, C1, ..., A2, B2, C2, ... (See FIG. 4), the center of gravity and the area are calculated for all the labeled moving body images.

After that, the process proceeds to step 26, where a logical product is calculated between each moving body image of the moving image at the first time and each moving body image of the moving image at the second time. Next, at step 27, as a result of the logical product operation, combination data having a value and showing a combination of the label of the moving body image at the first time and the label of the moving body image at the second time is obtained.

For example, the moving body image of the label B1 and the label B
2 and the positional relationship shown in FIG. 5, the result of the logical product of these moving body images represents the area of the hatched overlapping portion in FIG. Is larger than a predetermined value, both moving body images are considered to be moving body images formed by the same moving body. Therefore,
In this case, combined data indicating a set of moving body images indicated by the labels B1 and B2 is obtained. This calculation is performed for all the combinations to determine the combination of the moving body images.

In step 28, in order to determine whether or not each combination of the moving body images obtained as described above is valid, it is checked whether or not the following conditions are satisfied. (A) Each of the moving body images is combined with only one of the moving body images at the other time. (B) The difference between the areas of one set of moving body images is within 20%. (C) The straight line connecting the centers of gravity of the pair of moving body images passes through the logical product image obtained by these. Only the combination data that satisfies each of the conditions (a), (b), and (c) is captured as the correct combination data, and any combination that does not satisfy any one of the conditions (a) to (c). Matching data is removed.

In step 29, of the pair of moving body images shown by the combination data of which the validity is recognized, for example, as shown in FIG. 5, from the center of gravity W1 of the moving body image B1 at the first time. Corresponding to this, the moving body image B at the second time
A vector VC toward the center-of-gravity position W2 of 2 is calculated, and this vector VC is set as the motion vector of the player B who is a moving body at this point. In this way, the motion vector is calculated as described above for the moving body image pairs indicated by all the combined data.

As a result, the motion vectors of the players A to H and the ball I in the image are clarified, and in step 30, the motion directions and speeds of the players A to H and the ball I are displayed on the display device. 15 is displayed. In this display, the data of each motion vector may be displayed side by side on the display tube, and the direction and size of each motion vector are indicated by an arrow, and the arrow is superimposed on the position of each moving body image in FIG. It may be a graphical display that is also displayed. Further, the display device may be a liquid crystal display device, a printer, etc., in addition to the cathode ray tube display device.

As described above, in the apparatus shown in FIG. 1, since the video output signal obtained by using the television camera of the interlaced scanning image pickup tube system without the shutter is used, the first time obtained by the movement of each moving body is obtained. Of the correspondence between the moving body image of the moving image and the moving body image of the moving image at the second time,
It can be known very easily by calculating the logical product between them. Based on this correspondence relationship, the required motion vector of the moving body is calculated, and the motion analysis of the player and / or the ball is performed by electrical means. be able to.

In the above embodiment, the image pickup tube type television camera for photographing the moving body is used, but as can be seen from the above description, there is no shutter using the MOS type image pickup element or the CCD image pickup element. Even if an interlaced scanning type television camera is used, the same effect can be obtained.

Further, in the above embodiment, the case where only one television camera is used and the two-dimensional motion state viewed from this television camera is analyzed has been described. In addition to the analysis of a given two-dimensional motion state, as shown in FIG. 8, for example, a plurality of television cameras TV1, TV2, and TV3 are used to photograph moving bodies from a plurality of directions, and these television cameras are used. Based on the respective television signals VS1, VS2, VS3 obtained by TV1, TV2, TV3, the two-dimensional motion vector seen from each direction is converted into each processing unit U.
1, U2, and U3 are calculated as described above, and a plurality of resulting motion vectors for one moving body are combined by the vector combining unit CU in consideration of the set positions of the respective television cameras. It is also possible to obtain three-dimensional motion vector data. further,
A signal from the television camera may be temporarily recorded using a video deck or an optical disk recording device, and the motion may be analyzed based on the recorded data.

[0033]

According to the present invention, as described above, the correspondence relationship between each moving body image of one field image and each moving body image of the other field image is simplified by the logical product operation of each image data. It is possible to calculate the required motion vector of the moving body based on this correspondence relationship and analyze the motion state of the moving body at that time, so that multiple moving bodies can simultaneously move in their own directions. Even in the case of exercising, the state of motion of these moving bodies at the required timing can be analyzed simultaneously by electrical means. Therefore, based on the analysis result, it is possible to more effectively teach and educate the player after the game ends, and it is also possible to give an appropriate explanation to the spectators during the game.

[Brief description of drawings]

FIG. 1 is a configuration diagram of an embodiment of a motion analysis device according to the present invention.

FIG. 2 is a functional diagram showing a configuration of a main part of the motion analysis device of FIG.

3 is a flowchart showing a motion analysis program executed by the computer shown in FIG.

FIG. 4 is a diagram showing a pair of motion images obtained by the motion analysis device of FIG. 1.

5 is a diagram for explaining a logical product operation and a motion vector calculation between a pair of moving body images, which is executed in the motion analysis device of FIG.

FIG. 6 is an explanatory diagram illustrating a configuration of an interlaced television video signal.

FIG. 7 is an explanatory diagram for explaining a method of creating motion image data for one frame based on an odd field video signal by interpolation.

FIG. 8 is a configuration diagram schematically showing the configuration of another embodiment of the present invention.

[Explanation of symbols]

 10 motion analysis device 11, TV1, TV2, TV3 television camera 12 signal processing device 13 computer 13a first calculation means 13b second calculation means 13c logical product calculation means 13d motion vector calculation means A, B, C, D, E, F, G, H player I ball VS television video signal VS1 odd field video signal VS2 even field video signal

Claims (1)

[Claims] 1. A sports motion analysis apparatus for obtaining motion data of a player or a moving body such as a ball, wherein the number of moving bodies in a required field of view obtained by a television camera of an interlaced scanning tube type without a shutter. In response to a television image signal containing a three-dimensional motion image, the corresponding relation of the image of the moving body between the field image data is determined by the logical product operation of the field image data forming one frame of the television image signal. Kinematic analysis apparatus for sports, comprising: deciding means for calculating the kinematics of the moving body based on the correspondence determined by the deciding means and the field image data. ..