JP2012002659A - Analytic parameter output method, analytic parameter output device and analytic parameter output program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To output parameters for analyzing image data.SOLUTION: The analytic parameter output device 1 includes an object position data 12a that records corresponding positions of objects included in image data, a crowding degree calculation part 23 which reads the object position data 12a out and calculates the crowding degree of the objects, and a parameter calculation part 24 which calculates averages of crowding degrees of objects of a plurality of image data while calculating the parameters from the rates of the crowding degrees and averages of the image data by the image data.

Description

  The present invention relates to an analysis parameter output method, an analysis parameter output device, and an analysis parameter output program for outputting parameters for analyzing image data.

  With the recent development of information communication, content such as video data has increased explosively. In connection with this, the scene where a user uses a content is also various.

  Therefore, a method of performing a strategy analysis from a video of soccer, which is one of team sports, has been proposed (see, for example, Non-Patent Document 1). Here, as one of the strategies analysis of soccer, analysis of pass success or failure can be considered. It is known that the soccer ball pass course is represented by a gentle curve from the ball holder to the teammate. Therefore, in the method described in Non-Patent Document 1, a region where such a curve exists is extracted as a region where the path can be successful.

  Here, the soccer pass includes a pass to an attacking player who is a friend and a pass to an area where no player exists. Therefore, the passable area is an area connecting the ball holder and the teammate player, or an area connecting the ballholder and the player's existence area, and further passing through a position far from the opponent team player. This is a region where there is a curved line. In Non-Patent Document 1, such a passable area is extracted by the following method using an active net model.

  First, based on the player's position, an energy image is generated that represents the potential of the pass at each location on the soccer field. Next, by repeating the deformation so that the energy of the active net is minimized, a passable area in the image is extracted using the active net. At this time, the energy of the active net is defined as the sum of internal strain energy and compatibility energy. The internal strain energy is the energy of the internal strain of the active net itself corresponding to the force that shrinks the active net and keeps the shape smooth. The compatibility energy is the energy of the compatibility between the active net and the image that corresponds to the force that attracts the active net to a characteristic region in the image. Finally, paying attention to the density of the lattice points of the active net, the possibility of a successful path passing through the passable area is estimated.

In Non-Patent Document 1, an energy image generation method for analyzing a passable region will be described. The energy image is generated based on the position of the player on the soccer field. In the energy image, a region where there is an attacking player who is a pass target from a player holding the ball and a region where no player exists are represented by low gradation values. In addition, the area where the opponent team's defensive players are present is represented by a high gradation value. Since the player can move until the pass arrives, the energy given by the gradation value is made circular with respect to the position of the player. Here, when setting the energy of the area where the attacking player who is the target of the pass is set, it must be considered that the time until the pass reaches depends on the distance to the ball holder. Therefore, the radius r of the energy circle is defined as in equation (1) using the distance d1 from the ball holder.

  Here, vh is a parameter representing the speed at which the player can move the fastest. vb is a parameter representing the speed of the ball during the pass. In the method described in Non-Patent Document 1, an area in which a player can move until the ball reaches is calculated by Expression (1), and an energy image for analyzing a passable area to the player is generated.

  As described above, in order to generate an energy image that is analysis image data, the speed vh at which the player can move the earliest and the speed vb of the ball during the pass are required. The speed vh at which the player can move the earliest can be estimated from the player's level. For example, the speed vh at which the player can move the fastest varies depending on the attributes of athletes such as top professionals, college students, high school students, and junior high students, and the competition level. Accordingly, an average value can be set in advance for the speed vh at which the player can move the fastest for each attribute and competition level. On the other hand, a fixed value is set for the ball speed vb during the pass in the method described in Non-Patent Document 1.

Sho Takahashi, Hiroshi Ima, Miki Haseyama, “Estimation of Passable Area in Soccer Video Using Active Net”, IEICE Transactions, Vol. J92-D, No. 4, pp. 501-510, 2009

  However, the ball speed vb during the pass varies depending on the situation, so it is necessary to set it for each scene to be analyzed. However, since it is currently set as a fixed value, there is a problem that it cannot be analyzed properly. For example, when a player passes at a high speed in a crowded scene, it is impossible to set the ball speed vb suitable for the scene. As a result, analysis image data with high accuracy could not be output.

  Accordingly, an object of the present invention is to provide an analysis parameter output method, an analysis parameter output device, and an analysis parameter output program for outputting parameters for analyzing image data with high accuracy.

  In order to solve the above problems, a first feature of the present invention relates to an analysis parameter output method for outputting a parameter for analyzing image data. That is, the analysis parameter output method according to the first aspect of the present invention includes a density calculation step of reading object position data storing the position of an object included in image data from a storage device, and calculating the density of the object. A parameter calculating step of calculating an average density of objects of a plurality of image data and calculating a parameter for each image data from a ratio between the density of the image data and the average.

  Here, the image data is frame data of video data, and may be extracted based on the position of the object.

  Further, the image data may be image data in which the player passes the ball, and the object position data may be position data of the player's object. In this case, the parameter is the speed of the ball for analyzing the passable area of the ball in the image data.

  A second feature of the present invention relates to an analysis parameter output device that outputs a parameter for analyzing image data. That is, the analysis parameter output device according to the second feature of the present invention is a storage device that stores object position data stored in association with the position of an object included in image data, and reads the object position data from the storage device, A density calculation unit for calculating the density of the object and an average of the density of the objects of the plurality of image data, and for each image data, a parameter is calculated from a ratio between the density and the average of the image data. A parameter calculation unit.

  Here, the image data is frame data of video data, and may be extracted based on the position of the object.

  Further, the image data may be data of an image in which the player passes the ball, and the object position data may be position data of the player's object. In this case, the parameter is the speed of the ball for analyzing the passable area of the ball in the image data.

  A third feature of the present invention relates to an analysis parameter output program for causing a computer to execute the steps described in the first feature of the present invention.

  According to the present invention, it is possible to provide an analysis parameter output method, an analysis parameter output device, and an analysis parameter output program that output parameters for analyzing image data with high accuracy.

It is a figure explaining the functional block of the analysis parameter output device concerning an embodiment of the invention. It is a figure explaining the hardware constitutions of the analysis parameter output device concerning an embodiment of the invention. It is the figure which showed the position of the player detected from the analysis object image of the analysis parameter output device which concerns on embodiment of this invention, and an analysis object image. It is a figure explaining an example of data structure and data of object position data of an analysis parameter output device concerning an embodiment of the invention. It is a figure explaining an example of the data structure and data of object group density data of the analysis parameter output device concerning an embodiment of the invention. It is a figure explaining an example of data structure and data of parameter data of an analysis parameter output device concerning an embodiment of the invention. It is a flowchart explaining the process of the analysis parameter output device which concerns on embodiment of this invention. It is a figure explaining an example of the image which an analysis image generation apparatus outputs. It is a figure explaining the energy image produced | generated using the parameter output by the analysis parameter output device which concerns on embodiment of this invention, and the energy image produced | generated by the conventional method.

  Next, embodiments of the present invention will be described with reference to the drawings. In the following description of the drawings, the same or similar parts are denoted by the same or similar reference numerals.

  The analysis parameter output device 1 according to the embodiment of the present invention outputs parameters for analyzing image data to the analysis image generation device 2 as shown in FIG. The analysis image generation device 2 analyzes the semantic content of the image data to be analyzed by the method described in Non-Patent Document 1 described above. In particular, the analysis image generating apparatus 2 according to the embodiment of the present invention generates an image showing a passable area for image data in a soccer game. The analysis parameter output device 1 acquires image data from the analysis image generation device 2, calculates parameters necessary for analysis, and outputs them to the analysis image generation device 2.

  Here, a case where the analysis parameter output device 1 and the analysis image generation device 2 are mounted on different computers and can transmit and receive data via a communication network will be described. As another example, each data and each processing function of the analysis parameter output device 1 and each data and each processing function of the analysis image generating device 2 may be mounted on one computer.

(Analysis parameter output device)
As shown in FIG. 2, the analysis parameter output device 1 according to the preferred embodiment of the present invention includes a central processing control device 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, and an input / output interface 109. Are connected via the bus 110. An input device 104, a display device 105, a communication control device 106, a storage device 107, and a removable disk 108 are connected to the input / output interface 109.

  The central processing control device 101 reads out and executes a boot program for starting the analysis parameter output device 1 from the ROM 102 based on an input signal from the input device 104, and further reads an operating system stored in the storage device 107. Further, the central processing control device 101 controls various devices based on input signals from the input device 104, the communication control device 106, etc., and reads programs and data stored in the RAM 103, the storage device 107, etc. into the RAM 103. A processing device that loads and implements a series of processing described later, such as data calculation or processing, based on a program command read from the RAM 103.

  The input device 104 includes input devices such as a keyboard and a mouse through which an operator inputs various operations. The input device 104 generates an input signal based on the operation of the operator, and inputs via the input / output interface 109 and the bus 110. It is transmitted to the central processing control apparatus 101. The display device 105 is a CRT (Cathode Ray Tube) display, a liquid crystal display, or the like, and receives an output signal to be displayed on the display device 105 from the central processing control device 101 via the bus 110 and the input / output interface 109. It is a device that displays the processing result of the control device 101 and the like. The communication control device 106 is a device such as a LAN card or a modem, and is a device that connects the analysis parameter output device 1 to a communication network such as the Internet or a LAN. Data transmitted / received to / from the communication network via the communication control device 106 is transmitted / received to / from the central processing control device 101 via the input / output interface 109 and the bus 110 as an input signal or an output signal.

  The storage device 107 is a semiconductor storage device or a magnetic disk device, and stores programs and data executed by the central processing control device 101. The removable disk 108 is an optical disk or a flexible disk, and signals read / written by the disk drive are transmitted / received to / from the central processing control apparatus 101 via the input / output interface 109 and the bus 110.

  An analysis parameter output program is stored in the storage device 107 of the analysis parameter output apparatus 1 according to the preferred embodiment of the present invention. The storage device 107 further includes a video data storage unit 11, an object position storage unit 12, an analysis target image storage unit 13, an object group density storage unit 14, and a parameter storage unit 15. Further, the analysis parameter output program is read and executed by the central processing control device 101 of the analysis parameter output device 1, whereby the video acquisition unit 21, the object position detection unit 22, the analysis target image detection unit 23, and the object group density calculation. The unit 24, the parameter calculation unit 25, and the parameter output unit 26 are mounted on the analysis parameter output device 1.

  The video acquisition unit 21 acquires the video data 11 a from the analysis image generation device 2 and stores it in the video data storage unit 11. The video data storage unit 11 is a storage area in the storage device 107 in which the video data 11a is stored. The video data 11a is data including image data to be analyzed according to the embodiment of the present invention as a frame.

  The object position detection unit 22 reads the video data 11a from the storage device 107, detects the position of the object for each frame of the video data 11a, generates object position data 12a, and stores the object position data 12a in the object position storage unit 12.

  In the embodiment of the present invention, since the video data 11a is data during a soccer game, the objects are a player and a ball. The object position detection unit 22 first generates a background image of the soccer field. Furthermore, the object position detection unit 22 detects the player and ball areas based on the difference between the generated background image and the image data of each frame. Moreover, the object position detection part 22 discriminate | determines whether it is a player or a ball according to the size or shape of a difference area. Here, the shape and the line of the soccer field are known. By estimating the camera parameters representing the position and direction of the camera that shot the video, it is possible to detect the position of the player and the ball on the soccer field from the area of the player and the ball obtained from the background difference.

  With reference to FIG. 3, the processing of the object position detection unit 22 will be described. FIG. 3A shows image data of one frame of the video data 11a. The object position detection unit 22 estimates camera parameters from the image data shown in FIG. 3A and determines the position of the object and the type of the object. The image data obtained as a result is shown in FIG. The image data shown in FIG. 3B is a visual representation of the position of the object in the soccer field. In FIG. 3B, a circle mark indicates a ball position, and a square mark indicates a player position.

  The object position storage unit 12 is a storage area in the storage device 107 in which the object position data 12a is stored. The object position data 12a is data in which the identifier of the image data that is each frame constituting the video data 11a is associated with the position of the object included in the image data. In the embodiment of the present invention, the object is a player playing soccer and a ball, and the object position data 12a includes at least position information of the object of the player.

  The object position data 12a has a data structure as shown in FIG. 4, for example. The object position data 12a shown in FIG. 4 associates the position information of the object with the type using the image identifier and the object identifier as keys. The “position information” of the object is data of coordinates where the object is located in the frame. In the example shown in FIG. 4, the position information is indicated by coordinates in the X-axis direction and coordinates in the Y-axis direction. The “type” of the object indicates whether the object is “player” or “ball”.

  The analysis target image detection unit 23 reads the video data 11a from the storage device 107, and extracts a plurality of image data to be analyzed based on the position of the object of the frame data of the video data 11a. The analysis target image detection unit 23 stores the extracted image data in the analysis target image storage unit 13.

  For example, when the video data 11a is a video during soccer play, the analysis target image detection unit 23 is based on the movement of the position of the object in each frame, and switches the player who first held the ball in the team. After that, a frame at which a path is issued is detected as an analysis target. Further, the analysis target image detection unit 23 may preset a position where the position where the pass is made is important in the soccer field, and detect a frame at which the pass is issued from the set important position as the analysis target. . Further, the analysis target image detection unit 23 may detect frames at a predetermined time interval such as once per second as image data to be analyzed.

  The analysis target image storage unit 13 is a storage area in which a plurality of analysis target image data 13a, 13b, 13c,. The analysis target image storage unit 13 stores image data of frames extracted from the video data 11a. Here, the identifier data of the frame extracted from the video data 11a may be stored. In the embodiment of the present invention, the image data 13a, 13b,... To be analyzed is image data for the player to pass the ball.

  The object group density calculation unit 24 reads the object position data 12a from the storage device 107, and calculates the object density for each image data to be analyzed. Here, the object position data 12a read by the object group density calculation unit 24 is position information of an object associated with each identifier of the image data 13a, 13b, 13c,. In the embodiment of the present invention, the object group density calculation unit 24 calculates the density of the players from the position information of the players. The object group density calculation unit 24 stores the calculated player density for each image data in the object group density storage unit 14.

  The object group density calculation unit 24 first detects an area for calculating the density for each image data. This area is, for example, an area photographed by a camera and indicated by image data. Further, the object group density calculation unit 24 acquires the ball position information and the player position information of each image data from the object position data 12a, and detects the area where the players are crowded around the ball. good. Furthermore, the object group density calculation unit 24 calculates the density of players for each image data from the number of players located in the detected area and the ratio of the areas of the detected areas. As another method, after the area is detected, the object group density calculation unit 24 calculates, for each player located in the detected area, the shortest distance on the soccer field with other players. The average of the distances is the density. The object group density calculation unit 24 further repeats this process for each of the image data 13b, 13c,... To calculate the object group density of each image data.

  The object group density storage unit 14 is a storage area in the storage device 107 in which the object group density data 14a is stored. The object group density data 14a is data on the object density calculated for each image data to be analyzed. The object group density data 14a according to the embodiment of the present invention is data on the density of the player's object group.

  The object group density data 14a has a data structure shown in FIG. 5, for example. The object group density data 14a shown in FIG. 5 is associated with the density data of the players using the image identifier of the image data to be analyzed as a key.

  The parameter calculation unit 25 reads the object group density data 14 a from the storage device 107, calculates a parameter for each image data, and outputs it to the parameter storage unit 15. The parameter calculation unit 25 calculates the average density of objects of a plurality of analysis target image data, and calculates a parameter from the ratio between the density of the image data and the average for each analysis target image data. The average density of objects calculated here is, for example, the average density of objects in a plurality of frames detected by the analysis target image detection unit 23. Further, the average density of objects may be the average density of objects of a plurality of frames at regular intervals in the video data 11a.

In the embodiment of the present invention, the parameter is a speed vb when a path is issued. Here, the speed of the ball during the pass is proportional to the density of the players, and the speed of the pass is high. Therefore, when the player density of the image data to be analyzed is Nd and the average value of the player density of all the image data to be analyzed is Na, the ball speed vb during the pass is calculated from the equation (2). The

In Expression (1), V is set in advance at an average speed of a general path. This V may be set for each attribute or competition level of players such as top professionals, university students, high school students, and junior high school students.

  The parameter storage unit 15 is a storage area in the storage device 107 that stores parameter data 15a. The parameter data 15a according to the embodiment of the present invention is the velocity of the ball in each image data to be analyzed. As described above, the speed of the ball is calculated based on the density of players of each image data.

  The parameter data 15a has a data structure shown in FIG. 6, for example. The parameter data 15a shown in FIG. 6 is associated with ball speed data, which is a parameter, using the image identifier of the image data to be analyzed as a key. This ball velocity data is used to analyze the passable area of the ball in each image data.

  The parameter output unit 26 reads the parameter data 15 a from the storage device 107 and outputs it to the analysis image generation device 2. In the analysis image generation device 2, the image data is analyzed based on the parameter data 15a. In the embodiment of the present invention, the parameter output unit 26 outputs the ball speed data at the time of pass corresponding to each image data to be analyzed to the analysis image generating apparatus 2. The analysis image generation device 2 calculates a passable area for each image data to be analyzed, using the speed of the ball at the time of pass output from the analysis parameter output device 1.

  Next, an analysis parameter output method in the analysis parameter output device 1 according to the embodiment of the present invention will be described with reference to FIG.

  First, in step S <b> 101, the video acquisition unit 21 of the analysis parameter output device 1 acquires the video data 11 a from the analysis image generation device 2 and stores it in the video data storage unit 11. The video data 11a includes image data to be analyzed as frame data.

  Next, in step S102 to step S104, the object position detection unit 22 calculates the object position for each frame of the video data 11a. In step S102, the object position detection unit 22 detects the position of the player in each frame and acquires position information. In step S103, the object position detection unit 22 detects the position of the ball in each frame and acquires position information. In step S104, the object position detection unit 22 records the frame identifier, the object identifier, and the position information of each object in association with each other in the object position data 12a. When the processing from step S102 to step S104 is completed for all the frames of the video data 11a, the process proceeds to step S105. At this time, the object position data 12a includes object position information regarding all the frames of the video data 11a.

  In step S105, the analysis target image detection unit 23 detects a plurality of image data to be analyzed among the image data of each frame of the video data 11a based on the object position data 12a. In step S <b> 106, the object group density calculation unit 24 calculates the player density for each image stored in the analysis target image storage unit 13.

  In step S107, the parameter calculation unit 25 calculates the average value of the density of the players calculated in step S106. Further, in step S108, the parameter calculation unit 25 calculates parameters for each image. Specifically, the parameter calculation unit 25 determines the speed of the ball at the time of passing for each image data to be analyzed from the ratio between the density of players in each image and the average value of the density calculated in step S107. calculate.

  Finally, in step S109, the parameter output unit 26 outputs the velocity of the ball at the time of each image analysis target calculated in step S108 to the analysis image generating apparatus 2.

  As described above, according to the analysis parameter output apparatus 1 according to the embodiment of the present invention, parameters necessary for analyzing a passable area using an active net can be calculated from image data to be analyzed. Thereby, the passable region can be analyzed with high accuracy.

(Analytical image generator)
Processing of the analysis image generation device 2 will be described. In the embodiment of the present invention, the analysis image generation device 2 outputs the analysis target video data 211a to the analysis parameter output device 1 and uses the parameters output from the analysis parameter output device 1 to generate the analysis image data 213a. Generate.

  The analysis image generation apparatus 2 according to the embodiment of the present invention is realized by installing a program for executing a predetermined process in a general computer as shown in FIG.

  As shown in FIG. 1, the analysis image generation apparatus 2 according to the embodiment of the present invention includes a video data storage unit 211, a parameter storage unit 212, an analysis image storage unit 213, and an analysis image generation unit 221. The video data storage unit 211, the parameter storage unit 212, and the analysis image storage unit 213 are mounted on the storage device of the analysis image generation device 2. The analysis image generation unit 221 is mounted on the central processing control device of the analysis image generation device 2.

  In the embodiment of the present invention, the video data storage unit 211 is a storage area in which video data 211a including an image to be analyzed is stored. In the embodiment of the present invention, the video data 211a is video data during a soccer game. The analysis image generation device 2 analyzes a passable area for a predetermined frame of the video data 211a. The analysis target frame by the analysis image generation device 2 is specified in the analysis target image data 13 a of the analysis parameter output device 1.

  The parameter storage unit 212 is a storage area in which parameter data 212a necessary for analyzing the passable area by the analysis image generating apparatus 2 according to the embodiment of the present invention is stored. The parameter data 212a is output from the analysis parameter output device 1. In the embodiment of the present invention, the parameter data 212a is the speed vb of the ball when the player makes a pass.

  The analysis image storage unit 213 is a storage area in which analysis image data 213a output from the analysis image generation device 2 according to the embodiment of the present invention is stored. In the embodiment of the present invention, the analysis image data 213a is image data indicating a passable area for image data during a soccer game.

  The analysis image generation unit 221 generates analysis image data 213a for image data that is an analysis target frame of the video data 211a in consideration of the parameter data 212a. The analysis image data 213a is image data shown in FIGS. 8A to 8H.

  Hereinafter, processing of the analysis image generation unit 221 will be described.

  The analysis image generation unit 221 detects the object position for the image data to be analyzed. At this time, the object position is detected by the same process as the object position detection unit 22 described above. Here, the analysis image generating apparatus 2 analyzes each of the pass to the player of the teammate and the pass to the space. Therefore, the analysis image generation unit 221 determines whether the team player's uniform or the opponent team player's uniform is based on the color distribution of the player's uniform of each team and the color distribution of the player's object stored in advance. It is determined, and it is determined whether the player object is a player of a teammate or an opponent team. Thereby, the analysis image generation unit 221 can generate an image showing the position of the player of the teammate team, the position of the player of the opponent team, and the position of the ball on the soccer field.

  The analysis image generation unit 221 estimates the trajectory of the ball when passing based on the image data indicating the position of each object. Here, the pass may be a pass from a ball holder to a player of a teammate or a pass to an area (space) where no player exists. In order to estimate the passable area, an area where a curve passing through a position far from the player of the opponent team is extracted by connecting the ball holder and the player or space of the team member.

  Here, the analysis image generation unit 221 according to the embodiment of the present invention uses an active net. The active net can extract a region by repeatedly deforming into a shape that minimizes the sum of the internal strain energy and the image compatibility energy. By applying the active net, the lattice points are attracted to the edge to be extracted, and region extraction is realized. At this time, since the lattice points are parallel due to the internal strain energy, a sharp change in curvature does not occur at the edge of the extracted region. A region having a high density of lattice points in the final result obtained by such energy convergence shows a gentle curvature. Therefore, the analysis image generation unit 221 can estimate the area where the curve exists as a passable area assuming that the curve is a trajectory of the ball.

  Explain the path to a teammate's player. Players can move before the pass reaches the teammates. Therefore, for the pass to the player of the teammate, the energy is a circular shape with respect to the player position. As described above, the radius r of the circle of energy until the ball reaches is expressed by the above formula (1) using the distance d1 with the ball holder. In the embodiment of the present invention, the velocity vb of the ball during the pass used in the equation (1) is acquired from the analysis parameter output device 1. Thus, the image which showed the energy of the player of a teammate's team and the energy of the player of a partner team about the pass to the player of a teammate's team is shown in Drawing 8 (a). In FIG. 8A, a circle with a low gradation value indicates the energy of the player of the team member, and a circle with a high gradation value indicates the energy of the player of the opponent team. Of these, the region with the lower gradation value is estimated as the space that is the target of the pass to the teammate player.

  Describe the path to the space. In the path to the space, the analysis image generation unit 221 defines the degree of space that the teammate's players can reach, and regards this as the player position in the path to the player, thereby passing the path in the same way as the path to the player. Possible areas can be estimated. The analysis image generation unit 221 defines the degree of space at each point in the field shape from the position of the player of the opponent team, and estimates the possibility that the player of the team of friends will reach that point earlier than the pass. It is possible to estimate the space that is the target of the path. FIG. 8B shows an image showing the space that is the target of the path to the space in this way. In FIG. 8B, a region having a low gradation value is estimated as a space that is a target of a path to the space.

  The analysis image generation unit 221 applies an active net to each image created with respect to FIGS. 8A and 8B. Here, the properties of the active net are determined by defining the initial shape, the internal strain energy, and the conformity energy of the image from the properties of the object to be extracted. The diagram shown in FIG. 8C is a diagram in which the active net defined in this way is applied to the image shown in FIG. 8A, and the diagram shown in FIG. 8D is the active net applied in this way. It is the figure which applied the net to the image shown in FIG.8 (b). Moreover, FIG.8 (e) and FIG.8 (f) are the figures which expanded a part of FIG.8 (c) and FIG.8 (d), respectively.

  Further, the analysis image generation unit 221 pays attention to the density of the lattice points of the active net in the diagrams shown in FIGS. 8C and 8D, and the possibility that the path passing through the passable area is successful (path course). (Possibility). The analysis image generation unit 221 divides the soccer field into small areas in advance. Here, if the number of divisions is made fine, the width that allows the path deviation is reduced, and if it is increased, the width that allows the path deviation is increased.

  The passable area is expressed by a curve that exists radially from the ball holder in the active net, and the possibility of a pass course is determined according to the number per unit area of the curve. Therefore, in order to estimate the possibility of pass course for the divided areas, the number of curves passing through each area is counted. If the number of curves is large for each region, it is determined that there is a high possibility of a path course to that region. Furthermore, by binarizing the pass course possibility of each area with a threshold value, a passable area can be determined as shown in FIG. FIG. 8 (h) is an enlarged view of FIG. 8 (g).

  FIG. 9 is an image output from the analysis image generation apparatus 2 according to the embodiment of the present invention, and is an energy image of a passable region. The diagram shown in FIG. 9 corresponds to FIG. In FIG. 9, the position where the teammate player can receive the pass is shown in black with a low gradation value, and the position where the opponent team player is present and difficult to pass is high in the gradation value. It is shown in white.

  FIG. 9A is a diagram in the case where a passable area is obtained by a conventional method with a fixed ball speed. FIG. 9B is a diagram when the passable area is obtained based on the velocity of the ball obtained by analyzing the image data by the analysis parameter output device 1 according to the embodiment of the present invention. In the diagram shown in FIG. 9 (a), the position where the teammate's player can receive a pass overlaps the position where the opponent team's player is present and the pass is difficult to pass. It cannot be expressed. On the other hand, in the diagram shown in FIG. 9B, it can be seen that these areas do not overlap and the passable areas can be appropriately expressed.

  As described above, the analysis parameter output device 1 according to the embodiment of the present invention can output appropriate parameters for analyzing the image data from the object position of the image data.

(Other embodiments)
As described above, the embodiments of the present invention have been described. However, it should not be understood that the descriptions and drawings constituting a part of this disclosure limit the present invention. From this disclosure, various alternative embodiments, examples, and operational techniques will be apparent to those skilled in the art.

  For example, in the embodiment of the present invention, the case where the processes necessary for image analysis are implemented by the analysis parameter output device 1, the analysis image generation device 2, and separate computers has been described. Also good.

  It goes without saying that the present invention includes various embodiments not described herein. Therefore, the technical scope of the present invention is defined only by the invention specifying matters according to the scope of claims reasonable from the above description.

DESCRIPTION OF SYMBOLS 1 Analysis parameter output apparatus 2 Analysis image generation apparatus 11, 211 Image | video data storage part 12 Object position storage part 13 Analysis object image storage part 14 Object group density storage part 15, 212 Parameter storage part 21 Image | video acquisition part 22 Object position detection part 23 Analysis Object Image Detection Unit 24 Object Group Concentration Calculation Unit 25 Parameter Calculation Unit 26 Parameter Output Unit 101 Central Processing Controller 102 ROM
103 RAM
104 Input Device 105 Display Device 106 Communication Control Device 107 Storage Device 108 Removable Disk 109 Input / Output Interface 110 Bus 213 Analysis Image Storage Unit 221 Analysis Image Generation Unit

Claims (7)

An analysis parameter output method for outputting parameters for analyzing image data,
A density calculation step of reading object position data storing the positions of the objects included in the image data from the storage device, and calculating the density of the objects;
A parameter calculating step of calculating an average density of the objects of the plurality of image data, and calculating a parameter for each of the image data from a ratio between the density of the image data and the average;
An analysis parameter output method comprising: The analysis parameter output method according to claim 1, wherein the image data is frame data of video data and is extracted based on a position of the object. The image data is image data in which a player passes the ball,
The object position data is position data of the object of the player,
The analysis parameter output method according to claim 1, wherein the parameter is a velocity of the ball for analyzing a passable region of the ball in the image data. An analysis parameter output device for outputting parameters for analyzing image data,
A storage device for storing object position data stored in association with the position of an object included in image data;
A density calculation unit that reads the object position data from the storage device and calculates a density of the object;
A parameter calculating unit that calculates an average of the density of the objects of the plurality of image data and calculates a parameter for each of the image data from a ratio between the density of the image data and the average;
An analysis parameter output device comprising: The analysis parameter output device according to claim 4, wherein the image data is frame data of video data, and is extracted based on a position of the object. The image data is image data in which a player passes the ball,
The object position data is position data of the object of the player,
The analysis parameter output device according to claim 4, wherein the parameter is a velocity of the ball for analyzing a passable region of the ball in the image data.   The analysis parameter output program for making a computer perform the step of any one of Claim 1 thru | or 3.