Billiard ball color identification method and system
Through high-speed rotating motion target tracking camera group and mobile target focus technology, real-time tracking and identification of the styles of high-speed rotating billiards in billiards competitions is achieved, the problem of difficulty in identifying styles under high-speed motion is solved, and the intelligence and automation of the billiards competition system is improved. Control ability.
Patent Information
- Application Number
- CN202411121371.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-08-15
AI Technical Summary
In billiards, high-speed moving and high-speed rotating billiards make it difficult to identify suits. The existing technology is difficult to track and identify billiards on the table in real time, affecting the intelligent development of automatic game control.
The high-speed rotating motion target tracking camera team is used to capture the status image of each billiard ball during the billiard game, and the moving target focus tracking technology is used to automatically generate real-time tracking rules. The intelligent control camera team automatically tracks and shoots the high-speed rotating pool image after hitting the ball, and recognizes The remaining table and the pocket-snatch billiards style.
It significantly improves the accuracy and complex sports tracking ability of various billiard styles in billiards competitions, solves the difficulty of confusion and identification of patterns caused by high-speed movements, and improves the accuracy of the remaining pocket-stopping billiard styles and changes of the table.
Smart Images

Figure CN119027853A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of image recognition motion intelligent control, and more specifically, to a billiards suit recognition method and system. Background Art
[0002] Billiard ball suit recognition is very critical to the development of intelligent and automatic control technology of billiard game system; various standard billiard balls have various suits, and during billiard games, the ball moves after hitting the ball, especially high-speed movement and high-speed rotation, which can easily cause difficulty in suit recognition or even recognition errors. During billiard games, various standard billiard balls and multiple billiard ball targets often have different dynamic and static states, different movement speeds, continuous collision dynamic and static conversion, edge collision and rebound movement, etc., which makes suit recognition more complicated; specific problems include: how to shoot the state image of each billiard game process during the billiard game and track the high-speed rotating billiard ball after hitting the ball, how to generate real-time tracking rules for each billiard game process corresponding to the billiard standard game, how to track and shoot the image of the high-speed rotating billiard ball after hitting the ball, how to identify the suit of the remaining billiard balls in the table and the changing billiard suit, etc., which need to be solved; therefore, it is necessary to propose a billiard suit recognition method and system to at least partially solve the problems existing in the prior art. Summary of the invention
[0003] A series of simplified concepts are introduced in the content of the invention, which will be further described in detail in the specific implementation method section; the content of the invention of the present invention does not mean to attempt to limit the key features and essential technical features of the technical solution claimed for protection, nor does it mean to attempt to determine the scope of protection of the technical solution claimed for protection.
[0004] In order to at least partially solve the above problems, the present invention provides a billiard ball suit recognition method, comprising:
[0005] S100, using a high-speed rotating moving target tracking camera group to capture each billiards game process state image during the billiards game, and using a moving target focus tracking to track and capture the high-speed rotating billiards ball after hitting the ball;
[0006] S200, according to the correspondence between billiard ball suits and billiard ball formats, during the corresponding billiard ball format competition process, automatically generate real-time tracking rules for each billiard ball competition process;
[0007] S300, according to the real-time tracking rules of each billiards game process, intelligently controls the high-speed rotating target tracking camera group, automatically tracks and shoots the image of the high-speed rotating billiard ball after hitting the ball, and obtains multiple continuous images of the high-speed rotating billiard ball after hitting the ball;
[0008] S400, identifying the billiard suits in each billiard game process state image and the billiard suits in multiple continuous images of high-speed rotating billiard balls, and intelligently determining the suits of remaining billiard balls on the table, the suits of pocketed billiard balls, and the billiard suits that cause changes in game scores.
[0009] Preferably, S100 includes:
[0010] S101, setting a high-speed rotating moving target tracking camera group at an unobstructed position above the billiard table;
[0011] S102, capturing each billiards game process state image during the billiards game by a high-speed rotating moving target tracking camera group;
[0012] S103, using moving target focus tracking, tracking and photographing the billiard ball that rotates at high speed after being hit.
[0013] Preferably, S200 includes:
[0014] S201, according to the correspondence between billiard ball suits and billiard ball standards, select all billiard ball suit information of the corresponding standard;
[0015] S202, according to the information of all billiard balls of the corresponding standard, during the billiards standard game process, automatically generate real-time tracking rules for each billiards game process; automatically generate real-time tracking rules for each billiards game process including: according to the information of all billiard balls of the corresponding standard, one by one correspond to all rules of the billiards game process from serving to the ball falling into the pocket or the change of the game score until the end of the game, automatically track each billiard ball according to each billiard ball suit of the cue ball, the scoring ball and the deciding ball, and automatically generate real-time tracking rules for each billiards game process.
[0016] Preferably, S300 includes:
[0017] S301, generating high-speed rotating motion target tracking rule information according to the real-time tracking rule of each billiards game process;
[0018] S302, intelligently controlling the high-speed rotating target tracking camera group according to the high-speed rotating target tracking rule information, automatically tracking and photographing the image of the high-speed rotating billiard ball after hitting the ball, and obtaining multiple continuous images of the high-speed rotating billiard ball after hitting the ball.
[0019] Preferably, S400 includes:
[0020] S401, identifying the billiard ball suit in each billiard game process state image and the billiard ball suit in the multiple continuous images of the billiard ball moving and rotating at high speed according to each billiard game process state image and the multiple continuous images of the billiard ball moving and rotating at high speed;
[0021] S402, based on the billiard suits in each billiard game process status image and the billiard suits in multiple continuous images of high-speed rotating billiard balls, and based on the real-time tracking rules of each billiard game process, intelligently determine the remaining billiard suits on the table, the billiard suits in the pockets, and the billiard suits that cause changes in the game scores.
[0022] The present invention provides a billiard ball suit recognition system, comprising:
[0023] The high-speed moving target tracking module uses a high-speed rotating moving target tracking camera group to capture the state image of each billiards game process during the billiards game, and uses mobile target focus tracking to track and capture the high-speed moving rotating billiards after hitting the ball;
[0024] The billiards standard color and rule information module automatically generates real-time tracking rules for each billiards game process during the corresponding billiards standard game process according to the corresponding relationship between billiards color and billiards standard;
[0025] The moving target tracking control module intelligently controls the high-speed rotating moving target tracking camera group according to the real-time tracking rules of each billiards game process, automatically tracks and shoots the image of the high-speed rotating billiard ball after hitting the ball, and obtains multiple continuous images of the high-speed rotating billiard ball after hitting the ball;
[0026] The billiard suit intelligent recognition module identifies the billiard suit in each billiard game process status image and the billiard suit in multiple continuous images of high-speed rotating billiards, and intelligently determines the billiard suits remaining on the table, the billiard suits in the pocket, and the billiard suits that cause changes in the game score.
[0027] Preferably, the high-speed moving target tracking module includes:
[0028] A high-speed rotating motion tracking camera unit is set at an unobstructed position above the billiard table.
[0029] The billiards process status image shooting unit shoots each billiards game process status image during the billiards game through a high-speed rotating moving target tracking camera group;
[0030] The high-speed rotating target tracking unit utilizes the moving target focusing tracking to track and shoot the high-speed rotating billiard ball after hitting the ball.
[0031] Preferably, the billiards standard suit and rule information module includes:
[0032] The billiard suit and standard corresponding unit selects all billiard suit information of the corresponding standard according to the corresponding relationship between billiard suit and billiard standard;
[0033] The game process tracking rule unit automatically generates real-time tracking rules for each billiard game process according to all billiard ball suit information of the corresponding standard and the corresponding billiard game process; the automatic generation of real-time tracking rules for each billiard game process includes: according to all billiard ball suit information of the corresponding standard, one by one corresponding to all rules of the billiard game process from serving to the billiard ball falling into the pocket or the change of the game score until the end of the game, each billiard ball is automatically tracked according to each billiard ball suit of the cue ball, the scoring ball and the deciding ball, and the real-time tracking rules for each billiard game process are automatically generated.
[0034] Preferably, the moving target tracking control module includes:
[0035] The high-speed rotation motion tracking information unit automatically generates high-speed rotation motion target tracking rule information according to the real-time tracking rules of each billiards game process;
[0036] The high-speed billiards camera control unit intelligently controls the high-speed rotating target tracking camera group according to the high-speed rotating target tracking rule information, automatically tracks and shoots the image of the high-speed rotating billiards after hitting the ball, and obtains multiple continuous images of the high-speed rotating billiards after hitting the ball.
[0037] Preferably, the billiard suit intelligent recognition module includes:
[0038] A billiard ball suit recognition unit for hitting a ball in a match, which recognizes the billiard ball suit in each billiard ball match process state image and the billiard ball suit in the multiple continuous images of the billiard ball in high-speed motion and rotation according to each billiard ball match process state image and the multiple continuous images of the billiard ball in high-speed motion and rotation;
[0039] The billiard suit intelligent judgment unit can intelligently judge the remaining billiard suits on the table, the suits of the pocketed billiard balls and the suits that cause changes in the game scores according to the billiard suits in the status image of each billiard game process and the billiard suits in multiple continuous images of high-speed rotating billiard balls and the real-time tracking rules of each billiard game process.
[0040] The beneficial effects of the above technical solution include:
[0041] The present invention provides a billiard suit recognition method and system. The method uses a high-speed rotating target tracking camera group to shoot each billiard match process state image during a billiard match, and uses mobile target focus tracking to track and shoot the billiard ball that rotates at high speed after hitting the ball; according to the corresponding relationship between the billiard suit and the billiard standard, in the corresponding billiard standard match process, the real-time tracking rule of each billiard match process is automatically generated; according to the real-time tracking rule of each billiard match process, the high-speed rotating target tracking camera group is intelligently controlled to automatically track and shoot the image of the billiard ball that rotates at high speed after hitting the ball, and multiple continuous images of the billiard ball that rotates at high speed after hitting the ball are obtained; the billiard suit in each billiard match process state image and the billiard suit in multiple continuous images of the high-speed rotating billiard are identified, and the billiard suit remaining on the table, the suit of the billiard ball that falls into the pocket, and the billiard suit that causes the change of the match score are intelligently determined; the present invention provides a billiard suit recognition method and system. The invention is beneficial to the improvement of the intelligence and automatic control of the billiard game system; it can identify the billiard game process of various standards and colors of billiard balls; it can solve the problem of the movement of the ball after hitting the ball, and can solve the problem that high-speed movement and high-speed rotation easily cause confusion and difficulty in identification of colors or even identification errors; it can significantly improve the recognition of different states of multiple billiard balls of various standards in the billiard game, tracking of different movement speeds, and more complex color recognition accuracy such as continuous impact movement and static conversion and edge impact rebound movement tracking; it can shoot the state image of each billiard game process in the billiard game and track the high-speed moving and rotating billiard balls after hitting the ball; it can generate real-time tracking rules for each billiard game process corresponding to the billiard standard game; it can track and shoot the image of the high-speed moving and rotating billiard balls after hitting the ball; it can significantly improve the accuracy of identifying the remaining pocket billiard balls on the table and the changing billiard balls.
[0042] The present invention provides a billiard ball suit identification method and system. Other advantages, objectives and features of the present invention will be partially reflected in the following description, and will also be partially understood by those skilled in the art through research and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0044] Figure 1 The figure is a diagram of an embodiment of a billiard suit recognition system of the present invention.
[0045] Figure 2 This is another embodiment of a billiard ball suit recognition system according to the present invention.
[0046] Figure 3 The present invention is a diagram of an embodiment of a billiards suit recognition method and system application. DETAILED DESCRIPTION
[0047] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments so that those skilled in the art can implement the invention according to the instructions. Figure 1-Figure 3 As shown, the present invention provides a billiard ball suit recognition method, comprising:
[0048] S100, using a high-speed rotating moving target tracking camera group to capture each billiards game process state image during the billiards game, and using a moving target focus tracking to track and capture the high-speed rotating billiards ball after hitting the ball;
[0049] S200, according to the correspondence between billiard ball suits and billiard ball formats, during the corresponding billiard ball format competition process, automatically generate real-time tracking rules for each billiard ball competition process;
[0050] S300, according to the real-time tracking rules of each billiards game process, intelligently controls the high-speed rotating target tracking camera group, automatically tracks and shoots the image of the high-speed rotating billiard ball after hitting the ball, and obtains multiple continuous images of the high-speed rotating billiard ball after hitting the ball;
[0051] S400, identifying the billiard suits in each billiard game process state image and the billiard suits in multiple continuous images of high-speed rotating billiard balls, and intelligently determining the suits of remaining billiard balls on the table, the suits of pocketed billiard balls, and the billiard suits that cause changes in game scores.
[0052] The principle and effect of the above technical solution are as follows: the present invention provides a billiard suit recognition method, comprising: using a high-speed rotating moving target tracking camera group to shoot each billiard game process state image during the billiard game, and using a moving target focus tracking to track and shoot the high-speed rotating billiard ball after hitting the ball; according to the corresponding relationship between the billiard suit and the billiard standard, in the corresponding billiard standard game process, automatically generate a real-time tracking rule for each billiard game process; according to the real-time tracking rule for each billiard game process, intelligently control the high-speed rotating moving target tracking camera group, automatically track and shoot the image of the high-speed rotating billiard ball after hitting the ball, and obtain multiple continuous images of the high-speed rotating billiard ball after hitting the ball; identify the billiard suit in each billiard game process state image and the billiard suit in multiple continuous images of the high-speed rotating billiard ball, and intelligently determine the remaining billiard suits on the table, the suits of the pocketed billiard balls, and the changes in the game scores caused Billiard ball suits; the present invention is conducive to the improvement of the intelligence and automatic control of billiard game systems; it can identify billiard game processes with various suits of various standard billiard balls; it can solve the problem of the movement of the ball after hitting the ball, and can solve the problem that high-speed movement and high-speed rotation easily cause confusion and difficulty in identifying suits or even wrong identification; it can significantly improve the recognition of different states of motion and static of multiple billiard ball targets of various standard billiard balls during billiard games, tracking of different movement speeds, and more complex suit recognition accuracy such as continuous impact motion and static conversion and edge impact rebound movement tracking; it can shoot the state image of each billiard game process during the billiard game and track the high-speed moving and rotating billiard balls after hitting the ball; it can generate real-time tracking rules for each billiard game process corresponding to the billiard standard game; it can track and shoot the image of the high-speed moving and rotating billiard balls after hitting the ball; it can significantly improve the accuracy of identifying the suits of the remaining billiard balls in the table and the changing billiard suits;
[0053] The logic of the billiards suit recognition algorithm is as follows:
[0054] Step 1. Set the image cropping size: to speed up recognition and reduce the amount of calculation;
[0055] Step 2. Scale to keep a fixed size: to facilitate the modification of the cropping size and reduce the impact on subsequent algorithm parameters;
[0056] Step 3. Gaussian blur processing: Select appropriate Gaussian blur kernel parameters to reduce the noise and brightness changes of the image;
[0057] Step 4. Ball outline recognition: a. Add mask: set the shape and size of the mask to cover irrelevant areas in the image, reduce the amount of calculation and eliminate the image information in irrelevant areas more thoroughly; b. Color gamut conversion: convert the original BGR color gamut into LAB color gamut and HSV color gamut; LAB color gamut and HSV color gamut can better binarize the image to distinguish the ball outline; by fusing and comparing the results of LAB color gamut and HSV color gamut, the outline can be better obtained; c. Binarized image: empirically set the LAB color gamut and HSV color gamut to correspond to the upper and lower limits of the first optical definition, and use the upper and lower limits corresponding to the first optical definition The image is binarized by using the value, and the color image is converted into a black and white image to obtain a first optical binary image; d. Edge detection: edge detection is performed on the first optical binary image to find all contours in the image; e. Contour filtering: all contours in the image obtained based on edge detection are filtered and screened, and the corresponding first minimum threshold is set according to the contour closure, contour length and contour area; contours smaller than the corresponding first minimum threshold are judged as noise points and deleted, so as to obtain the final contour containing only the ball; f. Image fusion: the final ball contour after filtering is fused with the original image to obtain an image containing only pixels in the ball area;
[0058] Step 5. White outline recognition: Identify the white outline on the ball and obtain a binary image for size determination: a. Color gamut conversion: Convert the BGR image in step 4.f to LAB color gamut and HSV color gamut again; LAB color gamut and HSV color gamut better binarize the image to distinguish the white area of the ball; b. Binarized image: Experience setting new LAB color gamut and HSV color gamut corresponding to the upper and lower limits of the second optical definition, use the corresponding upper and lower limits of the second optical definition to binarize the image, and convert the color image into a black and white image to obtain the second optical binary image; c. Edge detection : Perform edge detection on the second optical binary image to find all white contours in the image; d. Contour filtering: Perform filtering and screening on all white contours obtained based on edge detection, and set the corresponding second minimum threshold by contour closure, contour length, and contour area; contours smaller than the corresponding second minimum threshold are judged as noise points and deleted; thus, the final white area contour of the ball is obtained; e. Identify the size: judge the size of the ball by the white area parameters; the white area parameters include: the number of white areas, area, arc length, approximate rectangular aspect ratio, and perimeter area ratio;
[0059] Step 6. Recognition of color contours, identification of color contours on the ball, used to determine which ball it is: a. Color gamut conversion: convert the BGR image in step 4.f to LAB color gamut and HSV color gamut again; b. Binarization image: empirically set new LAB color gamut and HSV color gamut corresponding to the upper and lower limits of the third optical definition, use the corresponding upper and lower limits of the third optical definition to binarize the image, and convert the color image into a black and white image to obtain the third optical binary image; c. Edge detection: perform edge detection on the third optical binary image to find all color contours in the image; d. Contour filtering: filter and screen all contours obtained based on edge detection, set the corresponding third minimum threshold according to the contour closure shape, contour length, and contour area; contours smaller than the corresponding third minimum threshold are judged as noise points and deleted; thus, the contour of the final ball with color area is obtained;
[0060] Step 7. Image fusion: Based on the first optical binary image, the second optical binary image, the third optical binary image and the final image containing only the ball outline, the final ball white outline and the final ball colored area outline obtained in steps 4, 5 and 6, perform debinarization processing and image bit operation to obtain an image in which only the color area retains the original color and the other areas are all black.
[0061] Step 8. Color calculation: Perform edge detection on the image in step 7 to obtain color contour position information; perform mean calculation on the color within the contour range to obtain the average color of the identified target ball.
[0062] Step 9. Color prediction: The color average value of the target ball identified in step 8 is elastically increased in size; the recognition requirements are weakened to reduce the inability to judge the color due to the color average value not being in the database due to changes in light and angle; the color average value corresponds to the LAB value and HSV value; then the color range value of the ball learned in advance in the database is compared. As long as the average value falls into the color range value of the ball learned in advance in the database after the elastic increase, it indicates that the ball may be a certain number of the ball in the database;
[0063] Step 10. Ball number judgment: Based on the result of the size, the color is predicted; the final result of the ball is logically judged; if it is not possible to make a clear judgment, the most likely ball number and the truth of the result are returned; true indicates that the result is correct, and false indicates that the result is uncertain;
[0064] Step 11. Final ball number determination: Since at least two cameras are used, each ball will provide two recognition results. Based on the above results, the most likely ball number is finally determined by logical judgment. The two recognition results include: the ball number and the truth of the result. If the recognition results of the two pictures are consistent, the result is output. If they are inconsistent, the final result is determined by logic. The maximum truth of the result is selected through cyclic determination.
[0065] Step 12. Learning about the color range of the ball in the database: Using the above algorithm, the color mean is obtained. The maximum and minimum values of the mean values obtained from multiple angles of the ball are taken as the color range;
[0066] The camera is installed in different positions and in different light environments, so two sets of algorithm parameters are used. This significantly improves the accuracy and stability of recognition, increases the threshold of recognition fault tolerance, and significantly improves the fault tolerance of billiards suit recognition.
[0067] In one embodiment, S100 includes:
[0068] S101, setting a high-speed rotating moving target tracking camera group at an unobstructed position above the billiard table;
[0069] S102, capturing each billiards game process state image during the billiards game by a high-speed rotating moving target tracking camera group;
[0070] S103, using moving target focus tracking, tracking and photographing the billiard ball that rotates at high speed after being hit.
[0071] The principle and effect of the above technical solution are as follows: a high-speed rotating target tracking camera group is set at an unobstructed position above the billiard table; the high-speed rotating target tracking camera group is used to shoot the status image of each billiard game process during the billiard game; the moving target is used for focusing and tracking to track and shoot the high-speed rotating billiard ball after hitting the ball; the high-speed rotating target tracking camera group includes: a target tracking compound camera, a target tracking motion control unit, a high-speed rotating motion motor unit and a target tracking high-speed camera shadowless lamp; the target tracking motion control unit controls the high-speed rotating motion motor unit, and the high-speed rotating motion motor unit drives the target tracking compound camera to shoot the status image of each billiard game process during the billiard game; the high-speed camera shadowless lamp eliminates the shadow of the billiard ball and emits a set wavelength of light; the target tracking compound camera monitors the reflected light of the set wavelength emitted by the target tracking high-speed camera shadowless lamp, keeps focusing and tracking the center point of the billiard ball surface, and uses the moving target to focus and track and shoot the high-speed rotating billiard ball after hitting the ball.
[0072] In one embodiment, S200 includes:
[0073] S201, according to the correspondence between billiard ball suits and billiard ball standards, select all billiard ball suit information of the corresponding standard;
[0074] S202, according to the information of all billiard balls of the corresponding standard, during the billiards standard game process, automatically generate real-time tracking rules for each billiards game process; automatically generate real-time tracking rules for each billiards game process including: according to the information of all billiard balls of the corresponding standard, one by one correspond to all rules of the billiards game process from serving to the ball falling into the pocket or the change of the game score until the end of the game, automatically track each billiard ball according to each billiard ball suit of the cue ball, the scoring ball and the deciding ball, and automatically generate real-time tracking rules for each billiards game process.
[0075] The principle and effect of the above technical solution are: according to the correspondence between billiard suits and billiard standards, all billiard suit information of the corresponding standard is selected; according to all billiard suit information of the corresponding standard, during the corresponding billiard standard competition process, real-time tracking rules for each billiard game process are automatically generated; the automatic generation of real-time tracking rules for each billiard game process includes: according to all billiard suit information of the corresponding standard, all rules of the billiard game process from serving to the billiard ball falling into the pocket or the change of the game score until the end of the game are matched one by one, each billiard ball is automatically tracked according to each billiard suit of the cue ball, the scoring ball and the deciding ball, and the real-time tracking rules for each billiard game process are automatically generated.
[0076] In one embodiment, S300 includes:
[0077] S301, generating high-speed rotating motion target tracking rule information according to the real-time tracking rule of each billiards game process;
[0078] S302, intelligently controlling the high-speed rotating target tracking camera group according to the high-speed rotating target tracking rule information, automatically tracking and photographing the image of the high-speed rotating billiard ball after hitting the ball, and obtaining multiple continuous images of the high-speed rotating billiard ball after hitting the ball.
[0079] The principle and effect of the above technical solution are as follows: according to the real-time tracking rules of each billiard game process, high-speed rotating motion target tracking rule information is generated; according to the high-speed rotating motion target tracking rule information, the high-speed rotating motion target tracking camera group is intelligently controlled to automatically track and shoot the image of the high-speed rotating billiard ball after hitting the ball, and multiple continuous images of the high-speed rotating billiard ball after hitting the ball are obtained; the high-speed rotating motion target tracking camera group is intelligently controlled to automatically track and shoot the image of the high-speed rotating billiard ball after hitting the ball, including: according to the high-speed rotating motion target tracking rule information, the focus point of the high-speed rotating motion target tracking camera always follows the rapid movement of the high-speed rotating billiard ball after hitting the ball; when the tracking target billiard ball falls into the pocket, the adjacent billiard target is automatically tracked until the billiard target tracking ends at the end of the game; before the start of each billiard game, all billiard balls on the table are scanned and photographed to obtain the first image of all billiard balls on the table; when the first ball is hit, multiple images of the billiard balls during the first hitting process are continuously taken. The billiard balls on the table are captured by the camera, and the images are taken until all the billiard balls on the table are stationary; the moving billiard balls during the first shot are tracked, and the billiard balls that fall into the pockets during the first shot or the billiard balls that cause the change in the match score are identified; before the second shot, when all the remaining billiard balls on the table are stationary, a second image of the remaining billiard balls on the table is captured; after the second shot, multiple images of the billiard balls during the second shot are captured continuously until all the billiard balls on the table are stationary; the moving billiard balls during the second shot are tracked, and the billiard balls that fall into the pockets during the second shot or the billiard balls that cause the change in the match score are identified; after the billiard balls are stationary again during the second shot, a third image of the remaining billiard balls on the table is captured; when a billiard ball falls into the pockets or the match score changes in each shot, the billiard balls that fall into the pockets or the billiard balls that cause the change in the match score are marked according to the last image of the billiard balls that fall into the pockets or the billiard balls that cause the change in the match score; reversely mark multiple images of the billiard balls during the shot, and reversely mark the images of the billiard balls that fall into the pockets or the billiard balls that cause the change in the match score in turn.
[0080] In one embodiment, S400 includes:
[0081] S401, identifying the billiard ball suit in each billiard game process state image and the billiard ball suit in the multiple continuous images of the billiard ball moving and rotating at high speed according to each billiard game process state image and the multiple continuous images of the billiard ball moving and rotating at high speed;
[0082] S402, based on the billiard suits in each billiard game process status image and the billiard suits in multiple continuous images of high-speed rotating billiard balls, and based on the real-time tracking rules of each billiard game process, intelligently determine the remaining billiard suits on the table, the billiard suits in the pockets, and the billiard suits that cause changes in the game scores.
[0083] The principle and effect of the above technical solution are as follows: according to each billiard game process state image and multiple continuous images of high-speed rotating billiard balls, the billiard suits in each billiard game process state image and the billiard suits in multiple continuous images of high-speed rotating billiard balls are identified; according to the billiard suits in each billiard game process state image and the billiard suits in multiple continuous images of high-speed rotating billiard balls, according to the real-time tracking rules of each billiard game process, the remaining billiard suits on the table, the suits of pocket billiard balls, and the billiard suits that cause changes in game scores are intelligently determined; the intelligent determination of the remaining billiard suits on the table, the suits of pocket billiard balls, and the billiard suits that cause changes in game scores includes: According to each billiard game process status image and multiple continuous images of high-speed rotating billiards, the billiard suits in each billiard game process status image and the billiard suits in multiple continuous images of high-speed rotating billiards are respectively identified to obtain the remaining billiard suit information on the table, and obtain the suit information of pocketed billiard balls or the billiard suit information that causes the change in the game score; according to the real-time tracking rules of each billiard game process, determine the remaining billiard suit information on the table, and obtain the suit information of pocketed billiard balls and the billiard suit information that causes the change in the game score. Whether it complies with the game rules; intelligently determine the remaining billiard suits on the table, the suits of pocketed billiard balls, and the billiard suits that cause the change in the game score.
[0084] The present invention provides a billiard ball suit recognition system, comprising:
[0085] The high-speed moving target tracking module uses a high-speed rotating moving target tracking camera group to capture the state image of each billiards game process during the billiards game, and uses mobile target focus tracking to track and capture the high-speed moving rotating billiards after hitting the ball;
[0086] The billiards standard color and rule information module automatically generates real-time tracking rules for each billiards game process during the corresponding billiards standard game process according to the corresponding relationship between billiards color and billiards standard;
[0087] The moving target tracking control module intelligently controls the high-speed rotating moving target tracking camera group according to the real-time tracking rules of each billiards game process, automatically tracks and shoots the image of the high-speed rotating billiard ball after hitting the ball, and obtains multiple continuous images of the high-speed rotating billiard ball after hitting the ball;
[0088] The billiard suit intelligent recognition module identifies the billiard suit in each billiard game process status image and the billiard suit in multiple continuous images of high-speed rotating billiards, and intelligently determines the billiard suits remaining on the table, the billiard suits in the pocket, and the billiard suits that cause changes in the game score.
[0089] The principle and effect of the above technical solution are as follows: the present invention provides a billiard suit recognition system, including: a high-speed moving target tracking module 1, which uses a high-speed rotating moving target tracking camera group to shoot each billiard game process state image during the billiard game, and uses mobile target focusing tracking to track and shoot the high-speed rotating billiard ball after hitting the ball; a billiard standard suit rule information module 2, which automatically generates a real-time tracking rule for each billiard game process in the corresponding billiard standard game process according to the corresponding relationship between the billiard suit and the billiard standard; a moving target tracking control module 3, which intelligently controls the high-speed rotating moving target tracking camera group according to the real-time tracking rule of each billiard game process, automatically tracks and shoots the image of the high-speed rotating billiard ball after hitting the ball, and obtains multiple continuous images of the high-speed rotating billiard ball after hitting the ball; a billiard suit intelligent recognition module 4, which identifies the billiard suit in each billiard game process state image and the billiard suit in multiple continuous images of the high-speed rotating billiard ball, and intelligently judges the billiard suit. The present invention can identify the remaining billiard balls on the table, the billiard balls in pockets and the billiard balls that cause changes in the game scores; the present invention is conducive to the improvement of the intelligence and automatic control of the billiard game system; it can identify the billiard game process with various billiard balls of various standards and colors; it can solve the problem that the ball moves after hitting the ball, and it can solve the problem that high-speed movement and high-speed rotation easily cause confusion and difficulty in identifying the colors or even wrong identification; it can significantly improve the recognition accuracy of more complex colors such as the recognition of different states of motion and static of multiple billiard balls of various standards in the billiard game, the tracking of different movement speeds, and the continuous impact motion and static conversion and edge impact rebound movement tracking; it can shoot the state image of each billiard game process in the billiard game and track the high-speed moving and rotating billiard balls after hitting the ball; it can generate real-time tracking rules for each billiard game process corresponding to the billiard standard game; it can track and shoot the image of the high-speed moving and rotating billiard balls after hitting the ball; it can significantly improve the recognition accuracy of the remaining billiard balls in pockets and the changing billiard balls;
[0090] The logic of the billiards suit recognition algorithm is as follows:
[0091] Step 1. Set the image cropping size: to speed up recognition and reduce the amount of calculation;
[0092] Step 2. Scale to keep a fixed size: to facilitate the modification of the cropping size and reduce the impact on subsequent algorithm parameters;
[0093] Step 3. Gaussian blur processing: Select appropriate Gaussian blur kernel parameters to reduce the noise and brightness changes of the image;
[0094] Step 4. Ball outline recognition: a. Add mask: set the shape and size of the mask to cover irrelevant areas in the image, reduce the amount of calculation and eliminate the image information in irrelevant areas more thoroughly; b. Color gamut conversion: convert the original BGR color gamut into LAB color gamut and HSV color gamut; LAB color gamut and HSV color gamut can better binarize the image to distinguish the ball outline; by fusing and comparing the results of LAB color gamut and HSV color gamut, the outline can be better obtained; c. Binarized image: empirically set the LAB color gamut and HSV color gamut to correspond to the upper and lower limits of the first optical definition, and use the upper and lower limits corresponding to the first optical definition The image is binarized by using the value, and the color image is converted into a black and white image to obtain a first optical binary image; d. Edge detection: edge detection is performed on the first optical binary image to find all contours in the image; e. Contour filtering: all contours in the image obtained based on edge detection are filtered and screened, and the corresponding first minimum threshold is set according to the contour closure, contour length and contour area; contours smaller than the corresponding first minimum threshold are judged as noise points and deleted, so as to obtain the final contour containing only the ball; f. Image fusion: the final ball contour after filtering is fused with the original image to obtain an image containing only pixels in the ball area;
[0095] Step 5. White outline recognition: Identify the white outline on the ball and obtain a binary image for size determination: a. Color gamut conversion: Convert the BGR image in step 4.f to LAB color gamut and HSV color gamut again; LAB color gamut and HSV color gamut better binarize the image to distinguish the white area of the ball; b. Binarized image: Experience setting new LAB color gamut and HSV color gamut corresponding to the upper and lower limits of the second optical definition, use the corresponding upper and lower limits of the second optical definition to binarize the image, and convert the color image into a black and white image to obtain the second optical binary image; c. Edge detection : Perform edge detection on the second optical binary image to find all white contours in the image; d. Contour filtering: Perform filtering and screening on all white contours obtained based on edge detection, and set the corresponding second minimum threshold by contour closure, contour length, and contour area; contours smaller than the corresponding second minimum threshold are judged as noise points and deleted; thus, the final white area contour of the ball is obtained; e. Identify the size: judge the size of the ball by the white area parameters; the white area parameters include: the number of white areas, area, arc length, approximate rectangular aspect ratio, and perimeter area ratio;
[0096] Step 6. Recognition of color contours, identification of color contours on the ball, used to determine which ball it is: a. Color gamut conversion: convert the BGR image in step 4.f to LAB color gamut and HSV color gamut again; b. Binarization image: empirically set new LAB color gamut and HSV color gamut corresponding to the upper and lower limits of the third optical definition, use the corresponding upper and lower limits of the third optical definition to binarize the image, and convert the color image into a black and white image to obtain the third optical binary image; c. Edge detection: perform edge detection on the third optical binary image to find all color contours in the image; d. Contour filtering: filter and screen all contours obtained based on edge detection, set the corresponding third minimum threshold according to the contour closure shape, contour length, and contour area; contours smaller than the corresponding third minimum threshold are judged as noise points and deleted; thus, the contour of the final ball with color area is obtained;
[0097] Step 7. Image fusion: Based on the first optical binary image, the second optical binary image, the third optical binary image and the final image containing only the ball outline, the final ball white outline and the final ball colored area outline obtained in steps 4, 5 and 6, perform debinarization processing and image bit operation to obtain an image in which only the color area retains the original color and the other areas are all black.
[0098] Step 8. Color calculation: Perform edge detection on the image in step 7 to obtain color contour position information; perform mean calculation on the color within the contour range to obtain the average color of the identified target ball.
[0099] Step 9. Color prediction: The color average value of the target ball identified in step 8 is elastically increased in size; the recognition requirements are weakened to reduce the inability to judge the color due to the color average value not being in the database due to changes in light and angle; the color average value corresponds to the LAB value and HSV value; then the color range value of the ball learned in advance in the database is compared. As long as the average value falls into the color range value of the ball learned in advance in the database after the elastic increase, it indicates that the ball may be a certain number of the ball in the database;
[0100] Step 10. Ball number judgment: Based on the result of the size, the color is predicted; the final result of the ball is logically judged; if it is not possible to make a clear judgment, the most likely ball number and the truth of the result are returned; true indicates that the result is correct, and false indicates that the result is uncertain;
[0101] Step 11. Final ball number determination: Since at least two cameras are used, each ball will provide two recognition results. Based on the above results, the most likely ball number is finally determined by logical judgment. The two recognition results include: the ball number and the truth of the result. If the recognition results of the two pictures are consistent, the result is output. If they are inconsistent, the final result is determined by logic. The maximum truth of the result is selected through cyclic determination.
[0102] Step 12. Learning about the color range of the ball in the database: Using the above algorithm, the color mean is obtained. The maximum and minimum values of the mean values obtained from multiple angles of the ball are taken as the color range;
[0103] The camera is installed in different positions and in different light environments, so two sets of algorithm parameters are used. This significantly improves the accuracy and stability of recognition, increases the threshold of recognition fault tolerance, and significantly improves the fault tolerance of billiards suit recognition.
[0104] In one embodiment, the high-speed moving target tracking module includes:
[0105] A high-speed rotating motion tracking camera unit is set at an unobstructed position above the billiard table.
[0106] The billiards process status image shooting unit shoots each billiards game process status image during the billiards game through a high-speed rotating moving target tracking camera group;
[0107] The high-speed rotating target tracking unit utilizes the moving target focusing tracking to track and shoot the high-speed rotating billiard ball after hitting the ball.
[0108] The principle and effect of the above technical solution are as follows: a high-speed moving target tracking module comprises: a high-speed rotating motion tracking camera unit, a high-speed rotating motion target tracking camera group is arranged at an unobstructed position above the billiard table; a billiard process state image shooting unit, through the high-speed rotating motion target tracking camera group, shoots each billiard game process state image during the billiard game; the high-speed rotating motion target tracking unit uses a moving target to focus and track, and tracks and shoots the high-speed rotating billiard ball after hitting the ball; the high-speed rotating motion target tracking camera group comprises: a target tracking compound camera, a target tracking motion control unit, a high-speed rotating motion motor unit and a target tracking high-speed camera shadowless lamp; the target tracking motion control unit controls the high-speed rotating motion motor unit, and the high-speed rotating motion motor unit drives the target tracking compound camera to shoot each billiard game process state image during the billiard game; the high-speed camera shadowless lamp eliminates the shadow of the billiard ball and emits a set wavelength of light; the target tracking compound camera monitors the reflected light of the set wavelength of light emitted by the target tracking high-speed camera shadowless lamp, keeps focusing and tracking the center point of the billiard ball surface, and uses a moving target to focus and track, and tracks and shoots the high-speed rotating billiard ball after hitting the ball.
[0109] In one embodiment, the billiards standard suit rule information module includes:
[0110] The billiard suit and standard corresponding unit selects all billiard suit information of the corresponding standard according to the corresponding relationship between billiard suit and billiard standard;
[0111] The game process tracking rule unit automatically generates real-time tracking rules for each billiard game process according to all billiard ball suit information of the corresponding standard and the corresponding billiard game process; the automatic generation of real-time tracking rules for each billiard game process includes: according to all billiard ball suit information of the corresponding standard, one by one corresponding to all rules of the billiard game process from serving to the billiard ball falling into the pocket or the change of the game score until the end of the game, each billiard ball is automatically tracked according to each billiard ball suit of the cue ball, the scoring ball and the deciding ball, and the real-time tracking rules for each billiard game process are automatically generated.
[0112] The principle and effect of the above technical solution are: the billiards standard suit rule information module includes: a billiards suit standard corresponding unit, which selects all billiards suit information of the corresponding standard according to the corresponding relationship between billiards suit and billiard standard; a game process tracking rule unit, which automatically generates real-time tracking rules for each billiard game process according to all billiards suit information of the corresponding standard in the corresponding billiard standard game process; the automatic generation of real-time tracking rules for each billiard game process includes: according to all billiards suit information of the corresponding standard, one by one corresponds to all rules of the billiard game process from serving to billiards falling into the pocket or the change of game score until the end of the game, and automatically tracks each billiard monomer according to each billiard suit of the cue ball, scoring ball and deciding ball, and automatically generates real-time tracking rules for each billiard game process.
[0113] In one embodiment, the moving target tracking control module includes:
[0114] The high-speed rotation motion tracking information unit automatically generates high-speed rotation motion target tracking rule information according to the real-time tracking rules of each billiards game process;
[0115] The high-speed billiards camera control unit intelligently controls the high-speed rotating target tracking camera group according to the high-speed rotating target tracking rule information, automatically tracks and shoots the image of the high-speed rotating billiards after hitting the ball, and obtains multiple continuous images of the high-speed rotating billiards after hitting the ball.
[0116] The principle and effect of the above technical solution are as follows: the moving target tracking control module includes: a high-speed rotating motion tracking information unit, which automatically generates high-speed rotating motion target tracking rule information according to the real-time tracking rules of each billiards game process; a high-speed motion billiards camera control unit, which intelligently controls the high-speed rotating motion target tracking camera group according to the high-speed rotating motion target tracking rule information, automatically tracks and shoots the image of the high-speed rotating billiards after hitting the ball, and obtains multiple continuous images of the high-speed rotating billiards after hitting the ball; intelligently controls the high-speed rotating motion target tracking camera group, and automatically tracks and shoots the image of the high-speed rotating billiards after hitting the ball, including: according to the high-speed rotating motion target tracking rule information, the focus point of the high-speed rotating motion target tracking camera always follows the rapid movement of the high-speed rotating billiards after hitting the ball; when the tracking target billiards ball falls into the pocket, the adjacent billiards target is automatically tracked until the billiards target tracking ends at the end of the game; before the start of each round of billiards, all billiards images on the table are scanned and photographed to obtain the first image of all billiards on the table; when During the first shot, multiple billiard ball images of the first shot process are continuously taken until all the billiard balls on the table are stationary; the moving billiard ball target during the first shot is tracked, and the billiard ball target that falls into the pocket during the first shot or the billiard ball target that causes the match score to change is identified; before the second shot, when all the remaining billiard balls on the table are stationary, a second image of the remaining billiard balls on the table is taken; after the second shot, multiple billiard ball images of the second shot process are continuously taken until all the billiard balls on the table are stationary; the moving billiard ball target during the second shot is tracked, and the billiard ball target that falls into the pocket during the second shot or the billiard ball target that causes the match score to change is identified; after the billiard balls are stationary again during the second shot, a third image of the remaining billiard balls on the table is taken; when a billiard ball falls into the pocket or causes the match score to change in each shot, the billiard ball that falls into the pocket or the billiard ball that causes the match score to change is marked according to the last image of the billiard ball that falls into the pocket or the billiard ball that causes the match score to change; multiple billiard ball images of the shot process are reversely marked, and the billiard ball images that fall into the pocket or the billiard ball images that cause the match score to change are reversely selected in turn. In one embodiment, the billiard ball suit intelligent recognition module includes:
[0117] A billiard ball suit recognition unit for hitting a ball in a match, which recognizes the billiard ball suit in each billiard ball match process state image and the billiard ball suit in the multiple continuous images of the billiard ball in high-speed motion and rotation according to each billiard ball match process state image and the multiple continuous images of the billiard ball in high-speed motion and rotation;
[0118] The billiard suit intelligent judgment unit can intelligently judge the remaining billiard suits on the table, the suits of the pocketed billiard balls and the suits that cause changes in the game scores according to the billiard suits in the status image of each billiard game process and the billiard suits in multiple continuous images of high-speed rotating billiard balls and the real-time tracking rules of each billiard game process.
[0119] The principle and effect of the above technical solution are as follows: the billiard suit intelligent recognition module includes: a billiard suit recognition unit for hitting the ball in a match, which identifies the billiard suit in each billiard match process state image and the billiard suit in multiple continuous images of high-speed rotating billiards according to each billiard match process state image and multiple continuous images of high-speed rotating billiards; a billiard suit intelligent determination unit, which intelligently determines the remaining billiard suits on the table, the billiard suits in pockets, and the billiard suits that cause changes in the match scores according to the billiard suits in each billiard match process state image and the billiard suits in multiple continuous images of high-speed rotating billiards according to the real-time tracking rules of each billiard match process; intelligently determines the remaining billiard suits on the table, the billiard suits in pockets, and the billiard suits that cause changes in the match scores according to the billiard suits in each billiard match process state image and the billiard suits in multiple continuous images of high-speed rotating billiards. The billiard suit and the billiard suit that causes the change of the match score include: according to each billiard game process state image and multiple continuous images of high-speed rotating billiards, respectively identifying the billiard suit in each billiard game process state image and the billiard suit in multiple continuous images of high-speed rotating billiards, obtaining the billiard suit information of the remaining billiard balls on the table, and obtaining the billiard suit information of the pocketed billiard balls or the billiard suit information that causes the change of the match score; according to the real-time tracking rules of each billiard game process, determining the billiard suit information of the remaining billiard balls on the table, and obtaining the billiard suit information of the pocketed billiard balls and the billiard suit information that causes the change of the match score comply with the rules of the game; intelligently determining the billiard suits of the remaining billiard balls on the table, the billiard suits of the pocketed billiard balls, and the billiard suits that cause the change of the match score.
[0120] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and the implementation methods. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.
Claims
1. A billiard ball suit recognition method, characterized in that: include: S100, using a high-speed rotating moving target tracking camera group to capture each billiards game process state image during the billiards game, and using a moving target focus tracking to track and capture the high-speed rotating billiards ball after hitting the ball; S200, according to the correspondence between billiard ball suits and billiard ball formats, during the corresponding billiard ball format competition process, automatically generate real-time tracking rules for each billiard ball competition process; S300, according to the real-time tracking rules of each billiards game process, intelligently controls the high-speed rotating target tracking camera group, automatically tracks and shoots the image of the high-speed rotating billiard ball after hitting the ball, and obtains multiple continuous images of the high-speed rotating billiard ball after hitting the ball; S400, identifying the billiard suits in each billiard game process state image and the billiard suits in multiple continuous images of high-speed rotating billiard balls, and intelligently determining the suits of remaining billiard balls on the table, the suits of pocketed billiard balls, and the billiard suits that cause changes in game scores.
2. A billiard ball suit recognition method according to claim 1, characterized in that: S100 includes: S101, setting a high-speed rotating moving target tracking camera group at an unobstructed position above the billiard table; S102, capturing each billiards game process state image during the billiards game by a high-speed rotating moving target tracking camera group; S103, using moving target focus tracking, tracking and photographing the billiard ball that rotates at high speed after being hit.
3. A billiard ball suit recognition method according to claim 1, characterized in that: S200 includes: S201, according to the correspondence between billiard ball suits and billiard ball standards, select all billiard ball suit information of the corresponding standard; S202, according to the information of all billiard balls of the corresponding standard, during the billiards standard game process, automatically generate real-time tracking rules for each billiards game process; automatically generate real-time tracking rules for each billiards game process including: according to the information of all billiard balls of the corresponding standard, one by one correspond to all rules of the billiards game process from serving to the ball falling into the pocket or the change of the game score until the end of the game, automatically track each billiard ball according to each billiard ball suit of the cue ball, the scoring ball and the deciding ball, and automatically generate real-time tracking rules for each billiards game process.
4. A billiard ball suit recognition method according to claim 1, characterized in that: S300 includes: S301, generating high-speed rotating motion target tracking rule information according to the real-time tracking rule of each billiards game process; S302, intelligently controlling the high-speed rotating target tracking camera group according to the high-speed rotating target tracking rule information, automatically tracking and photographing the image of the high-speed rotating billiard ball after hitting the ball, and obtaining multiple continuous images of the high-speed rotating billiard ball after hitting the ball.
5. A billiard ball suit recognition method according to claim 1, characterized in that: S400 includes: S401, identifying the billiard ball suit in each billiard game process state image and the billiard ball suit in the multiple continuous images of the billiard ball moving and rotating at high speed according to each billiard game process state image and the multiple continuous images of the billiard ball moving and rotating at high speed; S402, based on the billiard suits in each billiard game process status image and the billiard suits in multiple continuous images of high-speed rotating billiard balls, and based on the real-time tracking rules of each billiard game process, intelligently determine the remaining billiard suits on the table, the billiard suits in the pockets, and the billiard suits that cause changes in the game scores.
6. A billiard ball suit recognition system, characterized in that: include: The high-speed moving target tracking module uses a high-speed rotating moving target tracking camera group to capture the state image of each billiards game process during the billiards game, and uses mobile target focus tracking to track and capture the high-speed moving rotating billiards after hitting the ball; The billiards standard color rule information module automatically generates real-time tracking rules for each billiards game process during the corresponding billiards standard game process according to the corresponding relationship between billiards color and billiards standard; The moving target tracking control module intelligently controls the high-speed rotating moving target tracking camera group according to the real-time tracking rules of each billiards game process, automatically tracks and shoots the image of the high-speed rotating billiard ball after hitting the ball, and obtains multiple continuous images of the high-speed rotating billiard ball after hitting the ball; The billiard suit intelligent recognition module identifies the billiard suit in each billiard game process status image and the billiard suit in multiple continuous images of high-speed rotating billiards, and intelligently determines the billiard suits remaining on the table, the billiard suits in the pocket, and the billiard suits that cause changes in the game score.
7. A billiard ball suit recognition system according to claim 6, characterized in that: High-speed moving target tracking module, including: A high-speed rotating motion tracking camera unit is provided at an unobstructed position above the billiard table. A high-speed rotating motion target tracking camera group is provided; The billiards process status image shooting unit, through the high-speed rotating moving target tracking camera group, shoots each billiards game process status image during the billiards game; The high-speed rotating target tracking unit utilizes the moving target focusing tracking to track and shoot the high-speed rotating billiard ball after hitting the ball.
8. A billiard ball suit recognition system according to claim 6, characterized in that: Billiards standard suit rules information module, including: The billiard suit and standard corresponding unit selects all billiard suit information of the corresponding standard according to the corresponding relationship between billiard suit and billiard standard; The game process tracking rule unit automatically generates real-time tracking rules for each billiard game process according to all billiard ball suit information of the corresponding standard and the corresponding billiard game process; the automatic generation of real-time tracking rules for each billiard game process includes: according to all billiard ball suit information of the corresponding standard, one by one corresponding to all rules of the billiard game process from serving to the billiard ball falling into the pocket or the change of the game score until the end of the game, each billiard ball is automatically tracked according to each billiard ball suit of the cue ball, the scoring ball and the deciding ball, and the real-time tracking rules for each billiard game process are automatically generated.
9. A billiard ball suit recognition system according to claim 6, characterized in that: The moving target tracking control module includes: The high-speed rotation motion tracking information unit automatically generates high-speed rotation motion target tracking rule information according to the real-time tracking rules of each billiards game process; The high-speed billiards camera control unit intelligently controls the high-speed rotating target tracking camera group according to the high-speed rotating target tracking rule information, automatically tracks and shoots the image of the high-speed rotating billiards after hitting the ball, and obtains multiple continuous images of the high-speed rotating billiards after hitting the ball.
10. A billiard ball suit recognition system according to claim 6, characterized in that: Billiards suit intelligent recognition module, including: A billiard ball suit recognition unit for hitting a ball in a match, which recognizes the billiard ball suit in each billiard ball match process state image and the billiard ball suit in the multiple continuous images of the billiard ball in high-speed motion and rotation according to each billiard ball match process state image and the multiple continuous images of the billiard ball in high-speed motion and rotation; The billiard suit intelligent judgment unit can intelligently judge the remaining billiard suits on the table, the suits of the pocketed billiard balls and the suits that cause changes in the game scores according to the billiard suits in the status image of each billiard game process and the billiard suits in multiple continuous images of high-speed rotating billiard balls and the real-time tracking rules of each billiard game process.
Citation Information
Patent Citations
Method for controlling camera to follow and shoot automatically based on gathering trend in match field
CN105245845A
Automatic parameter calibration method for billiard ball design and color recognition device and computer storage medium
CN108198225A
Multi-target sphere motion trajectory tracking method and system based on rigid body collision characteristics
CN112907618A
Automatic billiard scoring system and method based on image analysis technology
CN118079355A
Billiard scoring system based on image recognition processing
CN118230216A
Cited By
Billiard target rapid identification method and system
CN119863610A
Billiards target quick identification method and system
CN119863610B
