GB2440573A

GB2440573A - Snooker ball setting aided by image correlation

Info

Publication number: GB2440573A
Application number: GB0615022A
Authority: GB
Inventors: Nicholas Richard Jame Opperman
Original assignee: SOUND DECISIONS Ltd
Current assignee: SOUND DECISIONS Ltd
Priority date: 2006-07-28
Filing date: 2006-07-28
Publication date: 2008-02-06
Also published as: GB0615022D0

Abstract

A system for aiding replacement of a snooker ball at a position on a snooker table comprises: means, e.g. video camera 34, for acquiring images of the snooker table and balls; memory means (46: Figure 4) for storing such an image acquired at a particular time or data derived therefrom; means (44,66: Figures 4 and 5) for processing such a stored image data and a currently acquired image or derived data to generate a correlation indication between the previous position of one of the balls and the current ball position; and display means for presenting the indication to the user. The system may indicate the position and direction for ball movement, via graphical image overlay: see Figure 13. Processing means may determine whether balls are moving or static, and store an image when all balls have stopped moving. The system aids a snooker referee in the replacement of a ball at a particular position on a snooker table after a "foul, and a miss" has been declared.

Description

TITLE

Snooker ball setting system

DESCRIPTION

This invention relates to a system for aiding a user in manually positioning snooker balls in their correct positions on a snooker table.

Referring to Figure 1 of the accompanying drawings, the ball-potting game of snooker is played on a table 10 which, if full size, measures 6 feet (1.83 m) by 12 feet (3.66 m) and which is surrounded by cushions 14,20,24. A baulk line 12 is marked on the table 10 spaced 29 inches (737 mm) from and parallel to one of the shorter cushions (the "baulk cushion") 14. A spot 16N is marked on the table 10 at the mid- point of the baulk line 12. A semicircle (the "D") 18 of radius 11 /z inches (292 mm) is marked on the table 10 centred on the spot 16N and lying between the baulk line 12 and the baulk cushion 14. As viewed from the baulk end 14, spots 16G,16Y are marked on the table 10 at the left and right intersections, respectively, of the baulk line 12 and the D 18. A spot 16U is marked on the table 10 at its exact centre. A spot 16P is marked on the table 10 on its longitudinal centreline halfway between the spot 16U and the other shorter cushion (the "top cushion") 20 of the table 10. A spot 16K is marked on the table on its longitudinal centreline 12M inches (324 mm) from the top cushion 20. Pockets 22 are formed at the corners of the table 10 and halfway along the longer cushions (the "side" cushions) 24.

Referring to Figure 2, snooker is played on a full-size table using twenty-two balls of diameter 2'/16 inches (52.4 mm), namely: a white "cue" ball 26; fifteen red balls 28 (shown hatched in Figure 2); and six other "coloured" balls, namely a yellow ball 30Y, green ball 300, brown ball 30N, blue ball 30U, pink ball 30P and a black ball 30K. At the beginning of a frame, the balls are placed on the table 10 in the positions shown in Figure 2, with the coloured balls 30Y,30G,30N,30U,30P,30K (collectively referred to by numeral 30) placed on their respective spots 16Y,16G,16N,16U,16P,16K (collectively referred to by numeral 16) and with the red balls 28 arranged as a closely-packed, equiangularly-triangular array immediately beyond the pink ball 30P, so that the red ball 28X at the apex of the triangle nearest the baulk cushion 14 is as close as possible to the pink ball 30P without touching it. A triangular rack 32 is usually used to assist in arranging the red balls 28. The cue ball 26 is placed anywhere in the D 18 by the player who breaks off the game.

It is a relatively straightforward matter manually to position the balls at the beginning of a frame of snooker. All of the coloured balls 30 are placed on their respective spots 16. The red balls 28 are arranged using the triangular rack 32, and it is easy to arrange them so that the red ball 28X at the apex is as close as possible to the pink ball 30P and by eye so that the longitudinal axis of the triangular array is parallel to the side cushions 24.

During the first phase of a frame of snooker, the coloured balls 30 are occasionally potted in the pockets 22 and score two points for the yellow ball 30Y, three points for the green ball 30G. four points for the brown ball 30N, five points for the blue ball 30U, six points for the pink ball 30P or seven points for the black ball 30K. When a coloured ball 30 is potted during the first phase, it is then manually returned to its original spot 16. If that spot 16 is covered by another ball, the ball 30 is placed on the highest available spot 16. If there is no available spot 16, it is placed as close to its own spot 16 as possible in a direct line between that spot 16 and the top cushion 20, without touching another ball 26,28,30. If there is no room on that side of the spot 16, it is placed as close to the spot 16 as possible in a straight line towards the baulk cushion 14, without touching another ball 26,28,30. Provided that the rules are understood, it is relatively straightforward manually to replace a potted coloured balI 30 at the correct position on the table 10 during the first phase of a frame.

Another occasion when one or more balls may need manually to be positioned on the table is after a "foul, and a miss" is declared. If a foul has been committed by not hitting the ball, or a ball, that is "on" first, or at all, and if certain other circumstances prevail, then "foul, and a miss" is called by the referee. In this case, one option for the other player is that they may request that the balls on the table that have been moved by the shot be returned to their positions before the foul, and that their opponent play the shot again. This rule is not normally applied to amateur matches, but it is applied to most professional matches, where substantial prize money and a title may be at stake. If the moved balls are not precisely returned to their original positions, it might affect the outcome of the match and the award of a title and prize money.

It may be that, when a "foul, and a miss" is declared, only one ball (the cue ball) has been moved and needs to be replaced in its previous position. However, two or more balls may have been moved and need to be replaced, and ultimately every ball on the table may have been moved and need to be replaced. Traditionally, this has required the referee to try to remember the precise position(s) to within, say, 1/z, inch (1 mm) of the moved ball(s) before it or they were moved on a table of area 72 feet2 (6.7 m2), an onerous, if not impossible, task indeed. Of course, a skilled and experienced referee will anticipate when a "foul, and a miss" situation may arise and make a particular mental note of the positions of the particular balls that are likely to be affected. Traditionally, after replacing the balls, the referee checks with the players that they are satisfied with the position(s) in which the ball(s) has or have been replaced. On the whole, snooker is played in a gentlemanly fashion, and controversy rarely arises regarding the position of a replaced ball after a "foul, and a miss." In the case of a dispute, the referee's decision is final, but traditionally all that the referee could really say about the matter is that, as far as they recall, that was the position of that particular ball a minute or more ago before the foul shot was played. Furthermore, as snooker becomes more popular and more matches are televised, the possibility arises that, upon subsequent review of video footage, a referee's decision on a "foul, and a miss" ball replacement may be shown not to have been particularly good, and that may cause disillusionment regarding the referee and the game, and put unjustifiable stress on the referee.

There is therefore a need for something to assist a snooker referee to ensure that a ball is replaced in its proper position after a "foul, and a miss." High level snooker matches are usually officiated at not only by the referee but also by a scorer. The fmal of the snooker World Championship in Sheffield, UK, in April and May 2006 was televised. In order to assist with the "foul, and a miss" ball replacement problem, the BBC made available to the scorer freeze frame shots and a current image from their overhead camera above the snooker table, and an audio link was made available between the referee and the scorer. Accordingly, in the case of a "foul, and a miss" ball replacement, the referee could ask for a second opinion from the scorer. The scorer could then compare the current image of the balls on the table with a freeze frame image of the balls on the table before the foul shot was made, and advise the referee accordingly. Whilst this may alleviate the "foul, and a miss" ball replacement problem to some extent, the aim of the present invention is to provide further improvement.

In accordance with the present invention, there is provided a system for aiding a user (typically a snooker referee) in the replacement of a snooker ball at a particular position on a snooker table. The system comprises: means (such as a video camera) for acquiring images of the snooker table and the balls thereon; means for storing such an image acquired at a particular time or data derived therefrom; means for processing such a stored image or derived data and a currently acquired image or data derived therefrom to generate an indication of correlation between the position of a particular one of the balls at the particular time and the current position of that ball; and means for presenting the indication to the user. By comparison with what was done at the 2006 World Championships, the system of the present invention makes a correlation between the previous and present ball positions, obviating the need for somebody to compare two different images and judge whether or not a ball position is the same in each.

The processing means may be operable to determine the positions of the particular ball at the particular time and at the current time. In this case, the presenting means may be operable to present to the user the direction in which the particular ball needs to be moved in order to assume its same position as at the particular time and/or to present to the user the distance by which the particular ball needs to be moved in order to assume its same position as at the particular time. For example, the presenting means might provide textual or speech-synthesised instructions, such as "3 mm towards baulk cushion" and then "1 mm towards left cushion." In some circumstances, it may be necessary to suspend the acquisition means (camera) by cables from a high ceiling or gantry, in which case there may be a possibility that the camera will move for example as a result of a draught. In order to deal with this, the processing means may be operable to detennine the positions of the particular ball relative to the acquisition means at the particular tune and at the current time, to determine the positions of the table relative to the acquisition means at the particular time and at the current time, and therefrom to determine the positions of the particular ball relative to table at the particular time and at the Current time.

As an alternative or addition to a textual display or speech-synthesis, the presenting means may include a graphical display. In this case, the processing means may be operable to cause the graphical display to display at least a portion of the stored image or a representation of its derived data, for example with an arrow showing the direction in which the particular ball needs to be moved. Alternatively, the processing means may be operable to overlay at least portions of the stored image and the current image or representations of their derived data and to cause the graphical display to display the overlaid images or representations. It will then be possible to tell when the overlaid images or representations of a particular ball coincide.

Again, in order to deal with possible movement of the acquisition means, the processing means may be operable to determine the position of the table in the stored image or derived data and in the current image or derived data, and to overlay the images or representations such that the determined table positions coincide.

The presenting means may be provided by a desktop video monitor. However, it is more preferably provided by a hand-held device, incorporating the graphical display, that the user can place on the table. In this case, the presenting means preferably further includes means (such as an electronic compass) for sensing the horizontal orientation of the graphical display, and the processing means may be operable to cause the graphical display to display the image with a fixed horizontal orientation regardless of the horizontal orientation of the graphical display. The system preferably also includes means for setting a datum orientation when the graphical display is registered with respect to the table, such that the subsequently displayed image is aligned to the table. For example, at the start of a match, the user might be required to place a particular side of a housing of the presenting means against the baulk cushion and press a key.

In the case where a graphical display is provided, the processing means is preferably operable to zoom the displayed image in dependence upon the distance by which the particular ball needs to be moved in order to assume its same position as at the particular time.

The processing means is preferably operable to process such a stored image or derived data and the currently acquired image or derived data to generate a sequence of such indications for a sequence of the balls.

The processing means is also preferably operable to determine whether any of the balls is in motion or whether all of the balls are static and to cause such an acquired image to be so stored each time that it is determined that all of the balls have become static.

A specific embodiment of the present invention will now be described, purely by way of example, with reference to the accompanying drawings, in which: Figure 1 is a plan view of a snooker table; Figure 2 is a plan view of the table, a set of balls in position for the start of a frame, and a rack for the red balls; Figure 3 is an isometric view of the snooker table and a system embodying the invention; Figure 4 is a block diagram of a camera unit forming part of the system; Figure 5 is a block diagram of a handheld unit forming part of the system; Figures 6 to 9 are flow diagrams used to describe the operation of the system; Figure 10 is an example of a display by the handheld unit immediately after a foul and a miss has been declared; Figure 11 illustrates the shot that was taken in the example; and Figures 12 & 13 are examples of displays by the handheld unit during the repositioning of a ball.

Referring to Figures 3 to 5, the system embodying the invention comprises a video camera unit 34 and a handheld unit 36, the construction of each of which will now be described.

The video camera unit 34 comprises a zoom lens 40 focussing onto a colour CCD array 42 that supplies images to a processor 44. The system needs to be able to distinguish between the pink ball and the red ball, and between the green ball and the green baize of the table. A CCD colour depth of 8 bits (256 colours) may be sufficient, but a colour depth of 24 bits (16,777,216 colours) is preferred, The processor 44 has associated memory 46, partly volatile and partly non-volatile, and an associated communications port 48. The zoom lens 40 can be zoomed by a motor 50 controlled by the processor 44. The components of the camera unit 34 and a power supply unit 52 are contained in a housing 54 that is suspended above the snooker table 10 with blue spot 30U of the table 10 on the optical axis 56 of the lens 40. For example, the camera unit 34 may be suspended by cables 58 from a ceiling or lighting gantry, or it may be attached to a lighting canopy for the table 10. For technical reasons, the CCD array 42 may have an aspect ratio of 2:1, but for commercial reasons it may have a more standard ratio such as 16:9, 5:4 or 4:3 in which some pixels will be unused. The CCD array 42 preferably provides a resolution of about 1 mm or less at the table 10. The CCD array 42 may therefore be 4,000 by 2,000 pixels, or 8 million pixels.

The handheld unit 36 has a housing 60 of a size that the referee can conveniently hold it in their hand or place it in their pocket. The housing 60 has a smooth planar base 62 so that the unit 36 can be placed on the table 10, as shown in Figure 3, without damaging the baize. The housing 60 also has at least one smooth planar reference edge 64 so that, with the base 62 resting on the table 10, the reference edge 64 can be placed against the baulk cushion 14 of the table 10 without damaging the cushion 14. The housing 60 contains a processor 66 with associated memory 68, partly volatile and partly non-volatile, and an associated communications port 70. The processor 66 drives a liquid crystal display 72 mounted in the top face of the housing 60 and can receive input from a user-operable push-button switch 74 and an electronic compass module 76. The LCD 72 does not need to have as great a resolution as the CCD array 42 of the camera unit 34, and may have a resolution of, for example, about 800 by 400 pixels, or 0.32 million pixels. The housing 60 also contains a battery 78.

The camera unit 34 and handheld unit 36 can communicate with each other, via their communications ports 48,70, in any suitable, but preferably wire-free, manner, for example by radio or infra red.

The operation and functionality of the system will now be described with reference to the flow diagrams of Figures 6 to 9.

Referring to Figure 6, the handheld unit 36 has an "asleep" mode, a "standby" mode and an active mode. At the beginning of a match or match session, the referee presses the switch 74 for more than three seconds in order wake up the handheld unit 36 from its asleep mode. This is detected by the processor 66 in step 88, and in response, in step 90, the processor 66 causes the LCD 72 to display a textual message such as, "Place unit on table with reference edge against baulk cushion, and then press button." In step 92, the processor 66 waits for the referee to press the switch 74. When they do, in step 94 the processor 66 reads the current bearing of the handheld unit 36 from the compass module 74 and stores it in the memory 46.

The processor 66 can then cause a communications check to be performed with the camera unit 34 in step 96, and assuming that is successful, in step 98 the processor 66 causes the LCD 72 to display a textual message such as, "OK. Press button for foul, and a miss." The handheld unit 36 then assumes its standby state, in which it stays until either the switch 74 is pressed momentarily (in response to which the handheld unit 36 assumes its active state as will be described in detail below), or the switch 74 is pressed for more than three seconds, which is detected by the processor 66 in step 100, as a result of which the handheld unit 36 re-assumes its asleep state.

Referring to Figure 7, when the camera unit 34 is switched on, in step 102 the processor 44 can cause a communications check to be performed with the handheld unit 36, and assuming that is successful, in step 104, the processor 44 causes the zoom motor 50 to zoom out the lens 40 to its maximum field of view. Then, in step 106, the processor 44 captures a frame from the CCD 42 and, in step 108, performs an object recognition process on the captured frame in order to recognise the pixel coordinates of the corners of the snooker table 10 in the captured frame. Object recognition processing of digital images is well known in the art and does not need detailed discussion here, except to say that recognition of the table is relatively straightforward, given that the table will appear as a large rectangle of a single green colour (with some aberrations caused by the balls 26,28,30, spots 16, lines 12,18 and pockets 22) surrounded my by a regular border of a contrasting colour, usually brown, provided by the wooden edge of the table 10. Then, in step 110, the processor calculates a zoom factor that, when applied to the captured frame centred on the centre of the frame, will cause the corners of the table 10 in the zoomed frame to assume positions close to the edges of the frame, and then the processor 44 causes the zoom motor 50 to adjust the lens 40 in accordance with the calculated zoom factor. In step 112, the processor 44 captures another frame from the CCD 42 and, in step 114, performs a further object recognition process on the captured frame in order to recognise the pixel coordinates of the corners of the snooker table 10 in the captured frame. The recognised pixel coordinates are then stored, in step 116, for future use. In step 118, the processor 44 performs an object recognition process on the captured frame in order to recognise the pixel coordinates of the centres of, and the colours of, any balls 26,28,30 on the table 10 in the captured frame. Again, recognition of the balls 26,28,30 is relatively straightforward, given that each ball is of uniform colour, the balls 26,28,30 are particular colours, all of the balls 26,28,30 are the same size and, given that the table 10 substantially fills the zoomed frame, all of the balls 26,28,30 are, within limits, a predetermined size in the captured frame. In step 120, the processor 44 stores, for each recognised ball, its colour and the pixel coordinates of its centre in the captured frame.

It will be appreciated that after the routines described above with reference to Figures 6 and 7 have been carried out, the handheld unit 36 knows the orientation of the table 10 in the horizontal plane and can therefore, upon subsequently reading the bearing provided by the compass module 76, calculate its own instantaneous orientation in the horizontal plane relative to the table 10. Also, the camera unit 34 has adjusted itself so that, as far as possible, the table 10 substantially fills the captured frames. Furthermore, the camera unit 34 has stored the initial positions of the corners of the table 10, so that the camera unit can subsequently determine whether there has been any movement of the camera unit 34 relative to the table 10, for example as a result of a draught moving the camera unit 34 on its suspension cables 58.

Moreover, the camera unit 34 has stored the Initial centre positions and colours of any balls 26,28,30 present on the table 10.

Subsequently, until a foul, and a miss is indicated, the camera unit 34 and handheld unit 36 routinely operate in the manner that will now be described with reference to Figure 8. In step 122, the processor 44 captures a frame from the CCD 42, and in step 124 performs an object recognition process on the captured frame in order to recognise the current pixel coordinates of the corners of the snooker table 10 in the captured frame. If there is any difference between the current corner coordinates and the original corner coordinates, in step 126 the processor 44 transforms the acquired frame by translation and/or rotation so that the corner coordinates match. Then, in step 128, the processor 44 performs an object recognition process on the transformed frame in order to recognise the pixel coordinates of the centres of, and the colours of, any balls 26,28,30 on the table 10 in that frame. In step 130, by comparing the current and previous ball positions, the processor 44 determines whether there has been any movement of the balls 26,28,30 (including the addition of a ball 26,28,30 to, or removal of a ball 26,28,30 from, the table 10). If so, the process loops back to step 122. However, if ball movement is detected in step 130, then in step 132, the processor 44 stores the centre positions and colours of the balls 26,28,30. In step 134, the processor 44 determines, from communication with the handheld unit 36, whether or not the switch 74 is being pressed. If not, then steps 136 to 142 are performed, which are identical to steps 122 to 128 described above.

Then, in step 144, by comparing the current and immediately previous ball positions, the processor 44 determines whether there has been any movement of the balls 26,28,30 (including the addition of a ball 26,28,30 to, or removal of a ball 26,28,30 from, the table 10). If so, the process loops back to step 122, whereas, if not, the process loops back to step 134.

It will be appreciated from the above that the processor 44 stores the ball positions and colours once each time all of the balls have become static. Accordingly, as a snooker game progresses, the processor 44 stores the ball positions and colours after each shot. There is no need for the camera unit 34 to retain all of the stored ball positions and colours in the memory 46, provided that it retains the last two sets of positions and colours.

In the event that the referee calls a "foul, and a miss" and needs the assistance of the system, they press the switch 74, and this is detected in step 134 of Figure 8, whereupon the routine that will now be described with reference to Figure 9 is performed.

When the switch 74 is pressed, those of the balls 26,28,30 on the table will all be at rest following the foul shot, and the ball positions will have been stored after and before the foul shot. In step 146, the processor 44 reads from memory 46 the ball positions before the foul shot and transmits them to the processor 66. In response, in step 148, from those positions the processor 66 generates a colour image of the table 10 and the balls 26,28,30 before the foul shot, and then in step 150 causes that image to be displayed on the LCD 72, for example as shown in Figure 10.

-10 -In the example, all of the coloured balls 30, but none of the red balls 28, are on the table 10, the yellow ball 30Y is "on", and the yellow ball 30Y has been snookered by the brown, blue and pink balls 30N,30U,30P, with the cue ball 26 close behind the brown ball 30N. For the purposes of the example, it is assumed that the player played the shot as indicated by Figure 11, off the side cushion, but with the cue ball 26 missing the "on" yellow ball 36Y and instead hitting the pink ball 36P.

Returning to Figure 9, after step 150 the process performs a "for each" loop for each ball that was on the table before the foul shot. The balls may be processed in a particular order, for example the cue ball 26 first, then the coloured balls 30 in order of value, and then the red balls 28. Each red ball 28 may be given an arbitrary but unique identifier so that each red ball can be distinguished from the others. The first three steps 152,154,156 in the "for each" loop are identical to the steps 122,124,126 or 136,138,140 described above with reference to Figure 8. Then, in step 158, the processor 44 performs an object recognition process on the captured transformed frame in order to recognise the pixel coordinates of the centre of the ball (ball "X") that is the subject of the "for each" loop. In step 160, the processor 44 calculates the misplacement, if any, of that ball position from the position of ball X before the foul shot. In step 162, the processor 44 determines whether the misplacement is zero. If so, the "for each" loop is restarted for the next ball. However, if a misplacement is detected in step 162, the processor 44 sends the coordinates of the misplaced ball to the processor 66, whereupon, in step 164 the processor 66 overlays an image of the misplaced ball on the image of the table and balls before the foul shot, for example as a black and white dotted circular outline. In step 166, the processor 66 reads the current bearing of the handheld unit 36 from its compass module 76, and then in step 168 rotates the image in dependence upon the difference between the current bearing and the bearing that was stored in step 94 (Figure 6) so that the image becomes aligned with the table. Then, in step 170, the processor 66 zooms in on the image, but so that ball X is contained in the zoomed image both in its position before the foul shot and in its current position. Then, in step 172, the processor 66 cause the aligned, zoomed image to be displayed on the LCD 72, together with textual information such as the name of ball X and the amount of misplacement. The process then loops back to step 152, and the process repeats, possibly many times. Meanwhile, it is expected that the referee will inspect the LCD 72 and move ball X to return it, perhaps step by step, to its position before the foul shot. Once that has been achieved, a determination is made in the next step 162 that the misplacement of ball X is zero, and the "for each" loop is restarted for the next ball. After the "for each" loop has been completed for all of the balls that were on the table before the foul shot, in step 174 the processor 66 causes a -11 -suitable message such as "All balls repositioned" on the LCD, and then the process returns to step 122 (Figure 8).

In the example of Figures 10 to 13, the processor 44 will determinethat the cue ball 26 has been moved, and an initial image that might be displayed is shown in Figure 12 assuming that the referee is standing by the right side cushion 24 facing the blue ball 16U and has placed the handheld device 36 on the table 10 in front of them with its reference edge 64 generally parallel to the side cushions 24. The cue ball 26 in its position before the foul shot appears to the left of the displayed image, and the current position of the cue ball is indicated by a black and white dotted circle 176 to the right of the displayed image. In the example, the processor 66 has also caused an arrowed line 178 to point from the current position to the position before the foul shot. Towards, the bottom of the LCD, the identity of the ball and the distance that it needs to be moved is displayed. An image that might be displayed once the cue ball 26 has almost been returned to its former position is shown in Figure 13, assuming that the referee is has moved to stand at the baulk end of the table 10 and has placed the handlield device 36 on the table 10 in front of them with its reference edge 64 generally parallel to the baulk cushion 14.

After the cue ball 26 has been returned to its position before the foul shot, the processor 44 will determine that the yellow, green, brown and blue balls 30Y,30G,30N,30U have not been moved as a result of the foul shot. It will then determine that the pink ball 30P has been moved and will assist the referee in repositioning it. The processor will then determine that the black ball 30K has not been moved.

It will be appreciated that many modification and developments may be made to the embodiment of the invention described above. For example, rather than displaying generated images on the LCD 72, the system may be arranged to display mere textual information, to utter synthesised speech or to display images as actually captured by the camera, or overlays of two such images, an overlay of a representation of the current ball position on an image as actually captured by the camera.

Rather than implementing the system as two units 34,36, it may be implemented as three units, namely a simpler camera unit, the handheld unit 36 and a specially programmed personal computer in communication with the camera unit and the handheld unit.

Referring to step 158 (Figure 9), in the event that the processor 44 is unable to recognise a particular ball in the acquired image, for example because it has been potted and not replaced on the table, or because it is being covered by the referee's hand, the system may be arranged to display an appropriate message, such as "Cue ball not visible. Press button to continue." In the event that the referee may decide that the system may be indicating an incorrect position, a facility may be provided for the referee to override the misplacement determination of step 162 (Figure 9) so that the process moves on to the next ball in the "for each" loop.

It should be noted that the embodiment of the invention has been described above purely by way of example and that many other modifications and developments may be made thereto within the scope of the present invention.

Claims

-13 -

CLAIMS

1. A system for aiding a user in the replacement of a snooker bail at a particular position on a snooker table, the system comprising: means for acquiring images of the snooker table and the balls thereon; means for storing such an image acquired at a particular time or data derived therefrom; means for processing such a stored image or derived data and a currently acquired image or data derived therefrom to generate an indication of correlation between the position of a particular one of the balls at the particular time and the current position of that ball; and means for presenting the indication to the user.

2. A system as claimed in claim 1, wherein the processing means is operable to determine the positions of the particular ball at the particular time and at the current time.

3. A system as claimed in claim 2, wherein the presenting means is operable to present to the user the direction in which the particular ball needs to be moved in order to assume its same position as at the particular time.

4. A system as claimed in claim 2 or 3, wherein the presenting means is operable to present to the user the distance by which the particular ball needs to be moved in order to assume its same position as at the particular time.

5. A system as claimed in any of claims 2 to 4, wherein the processing means is operable to determine the positions of the particular ball relative to the acquisition means at the particular time and at the current time, to determine the positions of the table relative to the acquisition means at the particular time and at the current time, and therefrom to determine the positions of the particular ball relative to table at the particular time and at the current time.

6. A system as claimed in any preceding claim, wherein the presenting means includes a graphical display.

7. A system as claimed in claim 6, wherein the processing means is operable to cause the graphical display to display at least a portion of the stored image or a representation of its derived data.

8. A system as claimed in claim 6, wherein the processing means is operable to overlay at least portions of the stored image and the current image or representations of the derived data and to cause the graphical display to display the overlaid images or representations.

-14 - 9. A system as claimed in claim 8, wherein the processing means is operable to determine the position of the table in the stored image or derived data and in the current image or derived data, and to overlay the images or representations such that the determined table positions coincide.

10. A system as claimed in any of claims 6 to 9, further including means for sensing the horizontal orientation of the graphical display, and wherein the processing means is operable to cause the graphical display to display the image with a fixed horizontal orientation regardless of the horizontal orientation of the graphical display.

11. A system as claimed in claim 10, further including means for setting a datum orientation when the graphical display is registered with respect to the table, such that the subsequently displayed image is aligned to the table.

12. A system as claimed in any of claims 6 to 11, wherein the processing means is operable to zoom the displayed image in dependence upon the distance by which the particular ball needs to be moved in order to assume its same position as at the particular time.

13. A system as claimed in any preceding claim, wherein the processing means is operable to process such a stored image or derived data and the currently acquired image or derived data to generate a sequence of such indications for a sequence of the balls.

14. A system as claimed in any preceding claim, wherein the processing means is operable to determine whether any of the balls is in motion or whether all of the balls are static and to cause such an acquired image to be so stored each time that it is determined that all of the balls have become static.

15. A system for aiding a user in the replacement of a snooker ball at a particular position on a snooker table, substantially as described with reference to the drawings.

16. A snooker table in combination with a system as claimed in any preceding claim.