WO2021162313A1

WO2021162313A1 - Golf ball crack determination device and method

Info

Publication number: WO2021162313A1
Application number: PCT/KR2021/001329
Authority: WO
Inventors: 심민경; 장수호; 이승훈; 고태석
Original assignee: 주식회사 골프존
Priority date: 2020-02-14
Filing date: 2021-02-02
Publication date: 2021-08-19
Also published as: KR102258670B1; TW202130393A; TWI747730B

Abstract

The purpose of the present invention is to provide a golf ball crack determination device and method, which present a specific method capable of determining whether a golf ball is cracked by analyzing a captured image of the golf ball to be used in a golf practice system, screen golf system or the like. To this end, a golf ball crack determination method, according to one embodiment of the present invention, comprises the steps of: acquiring a captured image of a stopped golf ball; extracting a ball image from the captured image; extracting a crack candidate from the extracted ball image; and determining whether the extracted crack candidate corresponds to a preset crack determination condition, thereby determining a crack of the golf ball.

Description

Apparatus and method for determining cracks for golf balls

The present invention provides a crack discriminating device for selecting defective golf balls, i.e., crack balls, such as cracks occurring when a user hits a golf ball through a golf practice system or a screen golf system, and a control method of the crack discriminating device is an invention about

In general, a golf practice system or a screen golf system has a ball supply system that prepares many golf balls in advance and supplies the golf balls to the batting table each time the user takes a golf shot because the user repeats the golf shot many times. Often times.

At this time, the ball supply system of the golf practice system or the screen golf system collects the golf balls hit by the user and sequentially transfers them to the user's bat to provide the golf balls when a golf shot is taken.

By the way, the golf balls supplied to the batting table are golf balls that have been hit multiple times by the user. Because crack balls such as the above are supplied to the batting table and the user hits it, it becomes a completely different batted ball than when a normal golf ball is hit. There is a problem in that undesirable results may appear in practice or in a virtual golf game through a screen golf system.

As described above, in the supply of balls in the golf practice system or the screen golf system, it is important to select and exclude crack balls from the supply.

Prior art documents such as Korean Patent Publication No. 10-1358177, Korean Patent Publication No. 10-1632414, and Registered Patent Publication No. 10-1436933 have been published as a technology related to such a conventional crack discrimination device.

The above Registration Patent Publication No. 10-1358177 relates to a technology for determining whether a crack ball exists through sound wave analysis detected by a sound wave sensor that detects a collision sound generated when a golf ball is hit by a striking part, No. 10-1632414 relates to a technique for determining a golf ball having a certain elasticity as a good golf ball by freely falling a golf ball from a predetermined height and determining a bad golf ball for a golf ball that has lost elasticity.

However, if there is a crack in the golf ball, but it is not severe, it is very difficult to determine the crack ball by analyzing the impact sound on the golf ball, and the size of the device is large, so it is difficult to downsize. Even in the case of discrimination, if the crack is not severe, it is impossible to discriminate, and since a space is required for a free fall of the golf ball, the size of the device is quite large, so there is a problem that miniaturization is impossible.

As a method to solve the above problems, in Patent Publication No. 10-1436933 above, a measuring camera is installed on the path of the golf ball, and the image of the golf ball photographed by the measuring camera is analyzed and the golf ball is measured. A technology for determining whether a crack has occurred is provided.

However, the above-mentioned prior art relates to a configuration of a device capable of determining whether a crack has occurred in a golf ball using a measurement camera, and specifically, how to detect a crack from an image photographed for a golf ball is presented. In the case of a golf ball, since a number of dimples exist on the surface and a specific logo or trademark is displayed, there is a problem in that it is quite difficult to detect a crack from an image taken of the golf ball.

The present invention provides an apparatus and method for determining a crack for a golf ball, which presents a specific method for determining whether a crack has occurred in a golf ball by analyzing a photographed image of a golf ball used in a golf practice system or a screen golf system. is to provide

A method for determining a crack for a golf ball according to an embodiment of the present invention includes: acquiring a photographed image of a still golf ball; extracting an image to be viewed from the captured image; extracting a crack candidate from the extracted ball image; and determining a crack for the golf ball by determining whether the extracted crack candidate corresponds to a predetermined crack discrimination condition.

Also preferably, the step of extracting the crack candidate includes: detecting edge pixels related to the surface of the ball with respect to the photographed ball image; and extracting continuously detected edge pixels by examining continuity, and extracting the crack candidate from the continuously detected edge pixels.

Also preferably, the step of extracting as a crack candidate includes: detecting edge pixels related to the surface of the ball with respect to the photographed ball image; It characterized in that it comprises the steps of removing a portion in which a pixel change occurs, and extracting a portion determined as a dense pixel cluster with respect to the edge pixels remaining after the removal as the crack candidate.

Also preferably, in the step of extracting the crack candidate from the continuously detected edge pixels, a pixel cluster in which the degree of cohesion of the pixel cluster is higher than a preset criterion through clustering of the continuously detected edge pixels is selected. and extracting it as a crack candidate.

Also preferably, the step of acquiring the photographed image includes acquiring an image of a continuous frame repeatedly photographed a preset number of times for the still golf ball, and extracting the ball image includes: and extracting the ball image for each of the images of the successive frames taken repeatedly, wherein the extracting of the crack candidate group includes edge pixels on the surface of the ball for each of the ball images of the extracted successive frames. detecting and preparing each edge pixel image; examining the continuity of each corresponding edge pixel of the respective edge pixel images to extract consecutively detected edge pixels to prepare a continuous edge image; and extracting the crack candidate group by analyzing edge pixels on the image.

Also preferably, the step of determining the crack comprises at least one of statistical analysis information about the brightness values of pixels of the crack part, size information of the crack part, and shape information of the crack part. and discriminating the crack by determining whether the crack candidate corresponds to at least one of characteristic information on the crack, which is preset as the crack discrimination condition.

On the other hand, the crack determination apparatus for a golf ball according to an embodiment of the present invention, a camera device for acquiring a photographed image of the still golf ball; and an image processing unit that extracts a ball image from the image photographed by the camera device, and extracts a crack candidate using the extracted ball image, and sets in advance with respect to a crack discrimination condition, the extracted ball image by the image processing unit and an arithmetic device including a crack discriminating unit for discriminating a crack in the golf ball by determining whether a crack candidate corresponds to the preset crack discrimination condition.

Also preferably, the image processing unit detects edge pixels related to the surface of the ball for the ball image photographed by the camera device, and determines the continuity of the detected edge pixels for the image repeatedly photographed a preset number of times. and extracting continuously detected edge pixels by irradiating, and extracting the crack candidate from the continuously detected edge pixels.

The apparatus and method for determining a crack for a golf ball according to the present invention provides a specific method for determining whether a crack has occurred in a golf ball by analyzing a photographed image of a golf ball used in a golf practice system or a screen golf system. By presenting it, the structure is simple and miniaturization is possible, so there is an effect that can be easily applied to an existing golf practice system or a screen golf system.

1 is a view schematically showing a state in which a ball supply system according to an embodiment of the present invention is applied to a screen golf system and installed.

2 is a view showing the configuration of the ball supply system shown in FIG.

FIG. 3 is a view for explaining the configuration of a crack determining apparatus according to an embodiment of the present invention as shown in FIG. 2 .

Figure 4 is a block diagram showing the entire control system of the ball supply system including a crack discrimination device according to an embodiment of the present invention.

5 is a flowchart of a crack discrimination method according to an embodiment of the present invention.

6 and 7 are diagrams for explaining a method of determining a crack of a golf ball according to an embodiment of the present invention.

Specific details of the crack discrimination apparatus and method according to the present invention will be described with reference to the drawings.

First, a ball supply system including a crack discrimination device according to an embodiment of the present invention will be described with reference to FIGS. 1 and 2 .

1 is a view schematically showing a state in which the ball supply system according to an embodiment of the present invention is applied and installed in a screen golf system, and FIG. 2 is a view showing the configuration of the ball supply system shown in FIG. 1 .

1 shows an example in which the ball supply system according to an embodiment of the present invention is applied to a screen golf system, but the present invention is not limited to the screen golf system and requires supply of golf balls for a user's golf shot, such as a golf practice system. Of course, it is applicable to all systems.

As shown in FIG. 1 , a screen 30 on which a virtual golf simulation image is projected is provided on the front surface of the screen golf booth, and on the floor 10 , the batting table 20 and golf where the golfer climbs and swings golf. A golf mat 22 on which the ball 1 is placed is provided, and although not shown in the drawing, a sensing device for sensing that the user hits the golf ball may be provided.

In addition, a simulator S is provided on one side of the booth as a computing device to implement a simulation image to be projected onto the screen 30 by performing predetermined data processing for virtual golf simulation according to the result sensed by the sensing device. can

The ball transfer device 100 of the ball supply system according to an embodiment of the present invention may be installed in the installation space 50 formed under the floor 10 of the booth, and between the screen 30 and the floor 10 . The inclined surface 32 is formed so that the golf ball 1 hit by the user hits the screen 30 and rolls down along the inclined surface 32 to move to the ball transfer device 100 in the installation space 50 . can be configured.

And, it is preferable that the installation space 50 is covered by the cover plate 52 so that the ball transfer device 100 is protected.

On the other hand, a space of a predetermined size is provided under the batting table 20, and a tee-up device 500 for teeing up golf balls is installed in the space so that the golf balls are automatically supplied to the golf mat 22 one by one. can

In this way, the user hits the golf ball 1 multiple times, and each of the hit golf balls 1 is returned to the ball transfer device 100 and supplied to the batting table through the tee-up device 500 again. At this time, golf balls having large and small cracks may occur due to several blows. A defective golf ball having cracks in this way will be referred to as a crack ball.

1 and 2, the ball supply system according to an embodiment of the present invention selects crack balls in the golf ball circulation process as described above and excludes them before being supplied to the user. 200 may be provided between the ball transfer device 100 and the tee-up device 500 .

That is, as shown in FIGS. 1 and 2 , the ball transport device 100 and the crack discrimination device 200 are connected to the ball transport pipe 150 , and the crack discrimination device 200 and the tee-up device 500 are connected to each other. is connected to the ball supply transport pipe 550, and as the golf ball 1 is recovered, the ball transport device 100 transports the golf ball to the crack discrimination device 200 through the ball transport pipe 150 and determines the crack The device 200 determines whether or not crack balls are one by one for the transferred golf balls, selects only normal golf balls, that is, not crack balls, and transfers them to the tee-up device 500 through the ball supply transfer pipe 550. The tee-up device 500 receiving the normal ball supplies the golf ball, which is the normal ball, by using the tee 502 . (In this case, instead of the tee-up device, it is also possible to provide a so-called caddy, which grips the golf ball and supplies it to the batting table).

At this time, as shown in FIG. 2 , the golf ball determined as a normal ball by the crack determination device 200 for the golf ball transferred through the ball transfer pipe 150 is teeed up through the ball supply transfer pipe 550 . On the other hand, the golf ball determined to be a crack ball in the crack determination device 200 may be transferred to a different path than the normal golf ball, so as to be excluded from the circulation of the ball supply, as shown in FIG. As shown in FIG. 1 , only the crack balls may be transported to a location where they are separately collected through the crack ball transport pipe 560 provided separately from the ball supply transport pipe 550 .

The conceptual configuration of the crack discrimination device 200 used in the ball supply system shown in FIG. 2 is shown in FIG. 3 (a).

As shown in (a) of FIG. 3 , the crack determining apparatus according to an embodiment of the present invention can be configured simply by including

camera devices

210 and 220 and an arithmetic unit 230 .

Inside the crack determining device 200, as shown in Fig. 3 (a), connected from the ball transport device 100 to the tee-up device 500 (more specifically, the ball of the ball transport device 100) Connected from the transfer pipe 150 to the ball supply transfer pipe 550) and an observation transfer passage 250 are provided so that the camera device can observe the golf ball on the observation transfer passage 250. .

The observation transfer passage 250 is implemented in the form of a transparent tube as shown in Fig. 3 (a) so that the camera device can sufficiently observe the golf ball 1 inside the observation transfer passage 250. can

The observation transfer passage 250 is not in the form of a transparent tube as described above, but in the form of an opaque tube as shown in FIG. By forming the transparent window 253 only for the region, it is also possible for the camera device to determine cracks through image capturing and analysis only for the golf ball observed through the window 253 .

Such a window 253 is preferably formed on the side facing the first camera device 210 and the side facing the second camera device 220, respectively.

Referring to FIGS. 2 and 3 , a plurality of golf balls transferred from the ball transport device 100 along the ball transport pipe 150 enter the crack discriminating device 200 according to the present invention while the ball transport pipe enters the ball transport pipe. It is transported along the observation transfer passage 250 connected to 150 , and in the process of moving the golf ball along the observation transfer passage 250 , the crack ball is checked by the

camera devices

210 , 220 and the calculating device 230 . If it is determined and is a normal ball, it continues to be transported along the observation transport passage 250 and then goes out to the ball supply transport pipe 550 and is supplied to the tee-up device 500. When it is determined as a crack ball, the crack ball 2 ) may be separated and discharged by bypassing a separate crack ball transfer passage 260 .

The crack ball transfer passage 260 is connected to the crack ball transfer pipe 560, and the crack ball 2 bypassed from the observation transfer passage 250 to the crack ball transfer passage 260 as described above is a crack ball. It may be moved to the transfer pipe 560 and separately discharged.

For example, although not shown in the drawing, a bypass passage connected to the crack ball transfer passage 260 on one side of the observation transfer passage 250 as shown in (a) of FIG. 3 and an opening and closing member for opening and closing it provided, when the arithmetic unit 230 determines that a crack exists in the golf ball at the position photographed by the

camera units

210 and 220, a control means such as the arithmetic unit opens the opening and closing member Thus, it is possible to allow the golf ball determined as a crack ball to be discharged from the observation transfer passage 250 to the crack ball transfer passage 260 through the bypass passage.

In this case, the crack ball transfer passage 260 may be inclined to facilitate discharge, or a separate mechanism may be used to transport the crack ball along the crack ball transfer passage 260 .

On the other hand, the camera device acquires a photographed image of a still golf ball, as shown in FIG. 3 , the first camera device 210 and the second By providing the camera device 220 to simultaneously observe one side and the other side of the golf ball to determine the crack using each photographed image, it is possible to increase the accuracy of the crack determination for the golf ball.

If the angle of view of each of the

camera devices

210 and 220 is widened, a plurality of golf balls may be included in one frame, which may cause a problem of specifying a golf ball to be determined. It is preferable that one golf ball appears completely on the photographed image and only a part of the golf ball around it appears.

When the shape of one golf ball appears completely in the photographed image and only a part of the shape of the golf ball around it appears, after extracting the outline of the object from the image, if the object whose outline has a circular shape is specified as a golf ball, the ball image It is preferable in that it can be easily extracted.

If, as described above, the observation transport path is configured in the form of an opaque tube as shown in (b) of FIG. 3 instead of a transparent tube as shown in (a) of FIG. In the case of the configuration in which the transparent window 253 is formed, a plurality of golf balls appear on the photographed image in that the golf ball 1 observed through the window 253 appears completely in the image photographed by the camera device. It is possible to solve the problem of specifying a golf ball as an object of discrimination.

On the other hand, although not shown, it is preferable that a lighting device for providing a separate visible light as illumination is provided inside the crack determining device 200 or each of the

camera devices

210 and 220 is configured to include a lighting device together.

On the other hand, with reference to Figures 4 and 5 will be described with respect to a crack discrimination device and a control method of a ball supply system including the same according to an embodiment of the present invention.

4 is a block diagram showing the entire control system of a ball supply system including a crack discrimination device according to an embodiment of the present invention, Figure 5 is a crack discrimination method by a crack discrimination device according to an embodiment of the present invention A flowchart showing the process.

As shown in FIG. 4, the crack determination device 200 according to an embodiment of the present invention includes a camera device (a first camera device 210 and a second camera device 220) and an arithmetic device 230, , the arithmetic unit 230 may include an image processing unit 231 and a crack discrimination unit 232 that can be classified according to hardware or software functions.

The image processing unit 231 extracts a ball image from the images captured by the

camera devices

210 and 220, and extracts a crack candidate using the extracted ball image, and the crack discrimination unit 232 ) sets a crack discrimination condition in advance, and serves to discriminate a crack for a golf ball by determining whether a crack candidate extracted by the image processing unit 231 corresponds to the preset crack discrimination condition.

Specific functions of the image processing unit 231 and the crack determining unit 232 will be described later.

As shown in FIG. 4 , the ball supply system is configured so that the crack discrimination device 200 , the ball transport device 100 , and the tee-up device 500 can operate under the control by the ball supply control unit 400 . can be configured.

In addition, according to an embodiment, the arithmetic unit 230 of the ball supply control unit 400 and the crack determination device 200 may be implemented separately, but may be implemented as one and the same controller.

The ball supply control unit 400 may, for example, be provided as a controller of a simulator in a screen golf system or may be provided as a controller provided in the ball supply system itself.

And, the computing device (S) may be a component corresponding to a client computer (simulator) or server of a golf practice system or a virtual golf simulation system using the above-described ball supply system.

A crack determination method according to an embodiment of the present invention according to the configuration as described above will be described with reference to the flowchart of FIG. 5 .

As described above, a plurality of golf balls 1 are transported while in close contact with each other in the observation transfer passage 250 in the crack determination device 200 as shown in FIG. As it is supplied through the device, a plurality of golf balls are transferred while being pushed through a section corresponding to one golf ball.

When the

camera device

210, 220 is in a stationary state as described above, a photographed image of the golf ball is acquired in order to determine the crack.

That is, as shown in FIG. 5 , the camera device acquires an image of a continuous frame repeatedly photographed a preset number of times with respect to a still golf ball ( S100 ).

The image acquired by the camera device is transmitted to the arithmetic unit, and the image processing unit of the arithmetic unit extracts an image to be viewed from each of the images of consecutive frames taken repeatedly as described above (S110).

The image processing unit detects edge pixels related to the surface of the ball for each of the ball images of consecutive frames extracted as described above, and prepares each edge pixel image (S120).

The image processing unit examines the continuity of each corresponding edge pixel of each edge pixel image prepared as described above, extracts the continuously detected edge pixels, and prepares a continuous edge image (S130).

The image processing unit performs clustering of edge pixels on the continuous edge image, that is, a portion in which pixels having similar characteristics (eg, brightness values in a similar range) are clustered into a pixel cluster. They are grouped and classified (S140).

The image processing unit extracts, as a crack candidate, a pixel cluster having a degree of cohesion within the pixel cluster (the degree of density of pixels constituting the pixel cluster) higher than a preset criterion among the pixel clusters classified by the clustering (S150). For example, from among several pixel clusters, only those having the number of pixels greater than or equal to a preset reference value may be selected and extracted as crack candidates.

As such, when a crack candidate or a group of crack candidates are extracted through a predetermined image processing by the image processing unit, the crack discrimination unit of the computing device determines whether the extracted crack candidate corresponds to a predetermined crack discrimination condition (S160). When it is determined that the above-mentioned crack discrimination conditions among the crack candidates are determined, that is, when it is determined that the crack ball is a crack ball (S170), the computing device may make an alarm by video, voice, sound, text, etc. that a crack ball has been detected, The discriminated crack ball may be separated from the ball transport path (eg, the observation transport path as shown in FIG. 3A) to be separately discharged (S180).

For example, in the case of a screen golf system, a method such as displaying a crack ball on a screen or a separate display as an image or giving an alarm by voice is possible.

If none of the crack candidates as described above meet the crack discrimination conditions, the ball is transferred normally and the next golf ball is transferred so that it stops within the camera's field of view, that is, at the shooting position (S190).

The process from the step of extracting the ball image of S110 to the step of extracting the crack candidate of S150 will be described in more detail with reference to FIGS. 6 and 7 .

6 is a diagram illustrating a process of generating a continuous edge image from a ball image through image processing of a crack determination method according to an embodiment of the present invention, and FIG. 7 is a method for extracting crack candidates using the continuous edge image. It is a drawing showing about

First, referring to FIG. 6 , the image captured by each camera device includes not only the golf ball but also the surrounding parts according to the angle of view. Only the part corresponding to the golf ball can be extracted from the first acquired image. have.

That is, it is possible to easily specify a golf ball in the shape of a circle by detecting the contour of the object appearing on the first acquired image and examining the shape, size, aspect ratio, etc. of the contour. By extracting only a portion corresponding to a golf ball, ball images referred to as b1, b2, and b3 of FIG. 6 may be extracted.

As described above, the camera device does not acquire only one image of a golf ball in a stationary state, but obtains images of a continuous frame by repeatedly photographing one stationary golf ball a preset number of times, and the continuous frame image is obtained. A plurality of ball images b1, b2, and b3 may be generated as shown in FIG. 6 by extracting a portion corresponding to a golf ball from each image of the frame.

When the preset number as the number of repeated shots for the golf ball is n, as shown in FIG. 6 , the ball image of b1 can be extracted from the image of the n-2th frame, and the image of the n-1th frame The ball image of b2 can be extracted from , and the ball image of b3 can be extracted from the image of the nth frame.

As described above, the ball image of the image repeatedly photographed for one still golf ball is an image of the same subject, but each ball image is a different image. This is because, in some cases, the pixel values of the same pixel position of each ball image are different from each other, because the light reflected from the golf ball surface by the illumination provided to the golf ball slightly changes over time.

Since dimples always exist on the surface of any golf ball, the light provided to the surface of the golf ball is reflected from the concave portion of each dimple and appears as an image, so ball images (b1, b2, b3) When an image edge is detected with respect to , an image having edge pixels such as e1, e2, and e3 shown in FIG. 6, that is, an edge pixel image is generated, respectively. In order to reveal the edge pixels more clearly, e1, e2, and e3 of FIG. 6 show that edge pixels extracted from the ball image are each generated as an edge pixel image by binarization.

Here, the edge refers to the part in which the brightness value of a pixel on the image changes rapidly. As described above, since a number of dimples are formed on the surface of a golf ball, light according to lighting is reflected in the concave part of the dimple. As a shadow is formed in the adjacent part, the brightness value of the pixel mainly changes rapidly in the dimple part. As described above, when an edge pixel for a ball image is detected, the part corresponding to the dimple is usually detected as an edge pixel. do.

Here, if a crack is formed in the golf ball, the crack part is detected as an edge pixel when the edge pixel is detected because the crack part has a difference in brightness value of the pixel from the peripheral part.

Accordingly, as shown in the edge pixel image of e1 of FIG. 6 , both a portion corresponding to a dimple and a portion corresponding to a crack of a golf ball may be detected as edge pixels.

However, as shown in the edge pixel image, if a large number of portions corresponding to dimples exist on the image, it becomes difficult to extract the crack portion. Therefore, it is preferable to remove the edge pixels of the portion corresponding to the dimples through appropriate image processing. .

Here, looking at the characteristics of the edge pixels for the dimple part, when shooting a golf ball using visible light as illumination, due to a flicker phenomenon caused by a difference between the shutter speed of the visible light illumination and the camera device, As the brightness of the reflected light is changed on the image, the edge pixel images extracted from each of a plurality of repeatedly photographed images have different images.

On the other hand, the crack formed in the golf ball is uniformly displayed on the edge pixel image despite the flicker phenomenon because the difference between the cracked portion and the peripheral portion is clear.

As shown in FIG. 6 , when edge pixel images e1 , e2 , and e3 are generated for each of the ball images b1 , b2 , and b3 on consecutive frames that are repeatedly photographed, the respective edge pixel images e1 and e2 , e3) appear differently depending on changes in the reflected light of the dimple portion, so only edge pixels continuously detected from all edge pixel images can be extracted by examining the continuity of each edge pixel on all the generated edge pixel images.

That is, by examining the brightness values of pixels existing at the same position in all the generated edge pixel images (e1, e2, e3), only pixels included in the same range are extracted, and pixels whose pixel values change rapidly for each image are excluded. By doing , the edge pixels for the dimple part can be removed to some extent.

In FIG. 6, by examining the continuity of each pixel in the edge pixel images (e1, e2, e3) and extracting only the edge pixels continuously detected from all images, a continuous edge image (CE) in which the dimple part is removed to some extent is generated. can

In the continuous edge image CE generated as described above, most of the cracks clearly appearing in all the edge pixel images e1, e2, and e3 are included, and the edge of the dimple part where the change occurred in each edge pixel image Pixels appear in a removed state.

Meanwhile, with reference to FIG. 7 , extraction of a crack candidate using the continuous edge image will be described.

Extracting a crack candidate using the continuous edge image is a process of extracting, as a crack candidate, a pixel cluster in which the degree of cohesion of the pixel cluster is higher than a preset criterion through clustering of the continuously detected edge pixels.

More specifically, as shown in FIG. 7 , by extracting pixels on the ball image bi corresponding to each pixel on the continuous edge image CE, the image corresponding to the continuous edge image referred to as PI in FIG. 7 . (PI) can be obtained.

In order to determine which part corresponds to a crack on the PI image, pixel clustering, ie, clustering, may be performed in which pixels having similar characteristics are grouped together.

Pixels that are once attached to each other on the PI image are made into one pixel cluster, but since the characteristics of the pixels can be different even within it, the pixel clusters can be classified according to the characteristics of the pixels, for example, the average brightness value of the pixel.

Even if the pixel clusters are classified as described above, if the number of pixels in the pixel cluster is very small, the probability of being judged as a crack is very low.

That is, from among the pixel clusters classified by clustering as described above, a pixel cluster having a degree of cohesion higher than a preset criterion may be selected and this may be determined as a crack candidate.

For example, in FIG. 7 , a pixel cluster having pixels that are somewhat dense, such as a G1 part, a G2 part, a G3 part, etc. on the PI image may be determined as a crack candidate.

When a crack candidate or a group of crack candidates is determined as described above, the crack discrimination unit of the arithmetic device of the crack discrimination apparatus of the present invention may determine whether the crack candidate corresponds to preset characteristic information on a crack to determine whether the crack is a crack.

Here, the characteristic information about the crack may include information such as statistical analysis information on the brightness values of pixels of the crack part, size information of the crack part, shape information of the crack part, and the like.

For example, if the average value of the brightness values of the pixels constituting the crack candidate is too high or too low, it may be determined that the crack is not a crack. may be determined, and if the size of the crack candidate is too small, it may be determined that the crack candidate is not a crack.

In this way, characteristic information about a crack that can be determined as a crack is preset as a crack discrimination condition, and it is possible to determine whether a crack exists by judging a crack candidate extracted according to the crack discrimination condition.

As described above, the crack determining apparatus and the crack determining method for a golf ball according to the present invention analyze a photographed image of a golf ball in the process of being transported for a golf shot in a golf practice system or a screen golf system, etc. It presents a specific method to determine whether a crack has occurred, and has the advantage of being able to sort out crack balls by an easy and simple method.

The apparatus and method for determining a crack for a golf ball according to the present invention can be used in the industrial field related to golf practice and the so-called screen golf industry field that allows golf play to be enjoyed based on a virtual golf simulation.

Claims

As a crack discrimination method for a golf ball,

acquiring an image photographed for a still golf ball;

extracting an image to be viewed from the captured image;

extracting a crack candidate from the extracted ball image; and

determining a crack for the golf ball by determining whether the extracted crack candidate corresponds to a predetermined crack discrimination condition;

Crack determination method for a golf ball comprising a.
The method of claim 1, wherein the extracting of the crack candidate comprises:

detecting edge pixels on the surface of the ball with respect to the photographed ball image;

extracting continuously detected edge pixels by examining the continuity of the detected edge pixels with respect to an image repeatedly photographed a preset number of times;

and extracting the crack candidate from the continuously detected edge pixels.
The method of claim 1, wherein the extracting as a crack candidate comprises:

detecting edge pixels related to the surface of the ball with respect to the photographed ball image;

removing a portion in which a pixel change occurs among the detected edge pixels with respect to an image repeatedly photographed a preset number of times;

and extracting, as the crack candidate, a portion determined as a dense pixel cluster with respect to the removed and remaining edge pixels.
The method of claim 2, wherein the extracting of the crack candidate from the continuously detected edge pixels comprises:

and extracting, as the crack candidate, a pixel cluster having a degree of cohesion of the pixel cluster higher than a preset reference through clustering of the continuously detected edge pixels.
According to claim 1,

Acquiring the photographed image includes acquiring an image of a continuous frame repeatedly photographed a preset number of times for the still golf ball,

The step of extracting the ball image includes extracting the ball image for each of the images of the consecutive frames taken repeatedly,

The step of extracting the crack candidate group,

Detecting edge pixels on the surface of the ball for each of the extracted ball images of the consecutive frames to prepare an edge pixel image, respectively;

preparing a continuous edge image by extracting continuously detected edge pixels by examining the continuity of each corresponding edge pixel of the respective edge pixel images;

and extracting the crack candidate group by analyzing edge pixels on the continuous edge image.
According to claim 1, wherein the step of determining the crack,

Statistical analysis information on the brightness values of pixels in the crack portion, crack size information, and crack feature information including at least one of shape information of the crack portion is preset as a criterion for determining the crack, and the crack candidate and determining whether the crack corresponds to at least one of the characteristic information on the crack, thereby determining the crack.
a camera device for acquiring a photographed image of a still golf ball; and

an image processing unit extracting a ball image from the image taken by the camera device and extracting a crack candidate using the extracted ball image;

an arithmetic device including a crack discrimination unit configured to determine a crack in a golf ball by determining whether the crack candidate extracted by the image processing unit corresponds to the preset crack discrimination condition;

Crack determination device for a golf ball comprising a.
The method of claim 7, wherein the image processing unit,

Edge pixels related to the surface of the ball are detected for the ball image taken by the camera device, and the continuously detected edge pixels are extracted by examining the continuity of the detected edge pixels for images repeatedly photographed for a preset number of times. and extracting the crack candidates from the continuously detected edge pixels.