Method and system for automatically detecting referee's decision in ball games

Abstract

In general, a method and system for automatically detecting a referee's decision during a ball game is provided. The method may include: receiving a plurality of images of a ball game field generated during a ball game; determining a first subset of images in the plurality of images based on a predetermined ball game rule, the first subset of images representing a first event suspected to be an event based on a specified rule; determining a second subset of images in the plurality of images representing a second event based on the predetermined ball game rule, wherein the second event is subsequent to the event based on the predetermined ball game rule; and analyzing the images of the second subset based on the predetermined ball game rule, and further determining a referee's decision regarding whether the first event is an event based on the specified rule based on the analysis of the images of the second subset.

Description

Field of the Invention

[0001] The present invention relates to the field of automatic analysis of sports events, and more particularly to a method and system for automatically detecting referee decisions during a sports event. Background of the Invention

[0003] In sports events today, the final ruling about rule-based events is made by referees. Some existing systems for sports event analysis can detect events suspected to be rule-based events. Taking football matches as an example, such systems can for example detect the event that the ball completely crosses the goal line between the goal posts and under the goal crossbar, and then judge the event as a scoring event. However, the detected event may occur in the case of violating predetermined ball game rules (such as caused by fouls and / or offside), and therefore may be cancelled by referees. Some existing systems for sports event analysis can further generate event-related data about detected events (such as suspected to be rule-based events), and further submit the event-related data to referees to help referees make rulings about the event.

[0004] One shortcoming of existing systems for sports event analysis is that they can at best only detect events that are suspected to be rule-based events. Today, these detected events may not be directly identified as rule-based events because they should be approved by the referee. SUMMARY OF THE INVENTION

[0006] One aspect of the present invention provides a system for automatically detecting a referee's decision during a ball game, the system comprising: a database comprising a plurality of images of a ball game field generated during a ball game; an event detection module coupled to the database, the event detection module: determining a first subset of images representing a first event suspected of being an event based on a specified rule from among the plurality of images based on predetermined ball game rules, and determining a second subset of images representing a second event from among the plurality of images based on the predetermined ball game rules, wherein according to the predetermined ball game rules, the second event is after the event based on the specified rule; an event analysis module coupled to the event detection module, the event analysis module analyzing the images of the second subset based on the predetermined ball game rules, and further determining the referee's decision as to whether the first event is an event based on the specified rule based on the analysis of the images of the second subset.

[0007] Another aspect of the present invention provides a method for automatically detecting a referee's decision during a ball game, the method comprising: receiving multiple images of a ball game field generated during the ball game; determining, by an event detection module based on predetermined ball game rules, a first subset of images in the multiple images representing a first event suspected to be an event based on specified rules; determining, by the event detection module, a second subset of images in the multiple images representing a second event based on the predetermined ball game rules, wherein, according to the predetermined ball game rules, the second event is after the event based on the specified rules; and, based on the predetermined ball game rules, analyzing the images of the second subset by an event analysis module, and further determining, based on the analysis of the images of the second subset, the referee's decision as to whether the first event is an event based on the specified rules.

[0008] Another aspect of the present invention provides a system for automatically detecting a referee's decision regarding a scoring event during a ball game, the system comprising: a database comprising a plurality of images of a ball game field generated during the ball game; an event detection module coupled to the database, the event detection module: determining a first subset of images representing a first event suspected of being a scoring event from among the plurality of images based on predetermined ball game rules, and determining a second subset of images representing a second event from among the plurality of images based on the predetermined ball game rules, wherein according to the predetermined ball game rules, the second event is after the scoring event; an event analysis module coupled to the event detection module, the event analysis module analyzing the images of the second subset based on the predetermined ball game rules, and further determining the referee's decision regarding whether the first event is an event based on specified rules based on the analysis of the images of the second subset.

[0009] Another aspect of the present invention provides a method for automatically detecting a referee's decision regarding a scoring event during a ball game, the method comprising: receiving multiple images of a ball game field generated during the ball game; determining, by an event detection module, a first subset of images representing a first event suspected to be a scoring event in the multiple images based on predetermined ball game rules, and determining, by the event detection module, a second subset of images representing a second event in the multiple images based on the predetermined ball game rules, wherein according to the predetermined ball game rules, the second event is after the scoring event; analyzing, by an event analysis module, the images of the second subset based on the predetermined ball game rules, and further determining, based on the analysis of the images of the second subset, the referee's decision regarding whether the first event is an event based on specified rules.

[0010] These, additional aspects, and / or other aspects and / or advantages of the invention are set forth in the detailed description which follows, may be inferred from the detailed description, and / or may be learned by practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] For a better understanding of embodiments of the invention and to show how the same may be put into practice, reference will now be made, by way of example only, to the accompanying drawings in which like numerals represent corresponding elements or parts throughout.

[0013] In the attached picture:

[0014] Figures 1A to 1C Various configurations of a system for automatically detecting a referee's decision during a ball game according to some embodiments of the present invention;

[0015] Figure 2A and Figure 2B is a schematic diagram of a reference pattern for a specific second event in a soccer game and in a field hockey game, respectively, according to some embodiments of the present invention;

[0016] Figure 3 is a flow chart of a method of a system for automatically detecting a referee's decision during a ball game according to some embodiments of the present invention.

[0017] It will be appreciated that for simplicity and clarity of illustration, the elements shown in the figures are not necessarily drawn to scale. For example, for clarity, the sizes of some elements may be enlarged relative to other elements. In addition, reference numerals may be repeated in multiple figures to indicate corresponding or similar elements where deemed appropriate.

[0018] Detailed Description of the Invention

[0019] In the following description, various aspects of the present invention are described. For the purpose of explanation, specific configurations and details are set forth in order to provide a thorough understanding of the present invention. However, it will also be apparent to those skilled in the art that the present invention can be practiced without the specific details set forth herein. In addition, well-known features may be omitted or simplified so as not to obscure the present invention. With specific reference to the accompanying drawings, it is emphasized that the details shown are presented by way of example and solely for the purpose of illustrative discussion of the present invention, and in order to provide what are considered to be the most useful and easily understood descriptions of the principles and concepts of the present invention. In this regard, no attempt is made to show the structural details of the present invention in a more detailed degree than the details necessary for a basic understanding of the present invention, and the description in conjunction with the accompanying drawings enables those skilled in the art to understand how several forms of the present invention can be embodied in practice.

[0020] Before describing in detail at least one embodiment of the present invention, it should be understood that the present invention is not limited in its application to the structural details and the arrangement of parts set forth in the following description and shown in the accompanying drawings. The present invention is applicable to other embodiments and combinations of the disclosed embodiments that can be practiced or implemented in various ways. Furthermore, it should be understood that the words and terms used herein are for descriptive purposes and should not be considered restrictive.

[0021] Unless otherwise specifically stated, as will be apparent from the following discussion, it should be recognized that throughout the specification discussion, terms such as "processing," "computing," "calculating," "determining," "enhancing," etc. are used to refer to the actions and / or processes of a computer or computing system or similar electronic computing device that manipulates data represented as physical quantities (e.g., electronic quantities) within registers and / or memories of the computing system and / or converts data represented as physical quantities (e.g., electronic quantities) within registers and / or memories of the computing system into other data similarly represented as physical quantities within the memory, registers, or other such information storage, transmission, or display devices of the computing system. Any disclosed module or unit may be implemented at least in part by a computer processor.

[0022] In general, a method and system for automatically detecting a referee's ruling during a ball game is provided. The method may include receiving a plurality of images of a ball game field generated during a ball game. In various embodiments, the plurality of images may be generated in real time, such as during an actual ball game, or the images may be pre-recorded images offline. The method may also include determining a first subset of images in the plurality of images based on predetermined ball game rules, for example, by an event detection module, the first subset of images representing a first event suspected to be an event based on a specified rule (e.g., a scoring event). The method may also include determining a second subset of images representing a second event in the plurality of images based on predetermined ball game rules, for example, by an event detection module, wherein the second event is after the event based on the specified rule according to the predetermined ball game rules. The method may also include analyzing the images of the second subset based on predetermined ball game rules, for example, by an event analysis module, and further determining the referee's ruling on whether the first event is an event based on the specified rule based on the analysis of the images of the second subset.

[0023] Now refer to Figures 1A to 1C , Figures 1A to 1C Various configurations of the system 100 for automatically detecting a referee's decision during a ball game according to some embodiments of the present invention; and Figure 2A and Figure 2B , Figure 2A and Figure 2B Schematic diagrams of reference patterns for specific second events in soccer and in field hockey, respectively, according to some embodiments of the present invention.

[0024] The system 100 may include a database 110. The database 110 may include a plurality of images 112 of the ball game field 90 generated during a ball game (e.g., Figure 1A ).

[0025] It should be noted that the term "ball game" as used herein in this application may refer to any ball game with predetermined ball game rules, such as but not limited to football, basketball, rugby, hockey, field hockey, etc.

[0026] In some embodiments, the image 112 may be generated in real time during an actual ball game. Alternatively or in addition, the image 112 may be pre-recorded (e.g., during a ball game) and stored in the database 110 for further analysis.

[0027] In some embodiments, the system 100 may include at least one set 120 of at least one camera. For example, the set 120 may include a first camera 120a, a second camera 120b, a third camera 120c, and / or a fourth camera 120d (e.g., Figure 1B ). One or more cameras in the set 120 may be positioned at a predetermined position of the ball game field 90. One or more cameras in the set 120 may generate multiple images 112 of the ball game field 90 in real time during an actual ball game, and deliver the multiple generated images 112 to the database 110. In various embodiments, one or more cameras in the set 120 may be one or more panoramic cameras, fixed cameras, or dynamic cameras.

[0028] In some embodiments, the system 100 may include multiple sets 120 of at least one camera. Each of the multiple sets 120 may be located at a different ball game field in the corresponding multiple ball game fields 90 to simultaneously generate, for example, a corresponding plurality of images of the corresponding multiple ball games being played on the corresponding multiple ball game fields 90.

[0029] For example, the system 100 may include a first set 122 of at least one camera located at a first ball playing surface 92 and a second set 124 of at least one camera located at a second ball playing surface 94 (e.g., Figure 1C). The first set 122 of one or more cameras may generate a first plurality of images 112a of the first ball game field 92 during the first ball game, and / or the second set 124 of one or more cameras may generate a second plurality of images 112b of the second ball game field 94 during the second ball game. Both the first set 122 of one or more cameras and the second set 124 of one or more cameras may further deliver the first plurality of images 112a and the second plurality of images 112b to the database 110, respectively.

[0030] In some implementations, the system 100 can simultaneously analyze multiple pluralities of images 112 (eg, the first plurality of images 112a and the second plurality of images 112b) corresponding to multiple ball games (eg, the first ball game and the second ball game).

[0031] The system 100 may include a calibration module 118. The calibration module 118 may be coupled to the database 110 (e.g., Figures 1A to 1C ). The calibration module 118 may calibrate each of the plurality of images 112 (or at least some of the plurality of images) so that each pixel in the image is associated with a specific geographic location on the ball game field 90.

[0032] The system 100 may include a classification module 130. The classification module 130 may be coupled to the database 110 and / or receive a plurality of images 112 (e.g., Figures 1A to 1C shown).

[0033] The classification module 130 may determine at least one background image based on the plurality of images 112 or at least some of the plurality of images 112. The classification module 130 may further generate a corresponding plurality of foreground images based on the plurality of images 112 (or at least some of the plurality of images 112) and based on the at least one background image. For example, the classification module 130 may subtract the at least one background image from each of the plurality of images 112 (or at least some of the plurality of images 112) to generate a corresponding foreground image.

[0034] In some embodiments, at least one background image includes stationary (or substantially stationary) objects related to the ball game. For example, at least one background image may include pixels representing the ball game field 90. In various embodiments, the classification module 130 uses a machine learning algorithm to determine and / or update at least one background image.

[0035] In some embodiments, the foreground image includes an object of interest related to the ball game. For example, the foreground image may include moving objects, such as players of the first team, players of the second team, and / or a ball.

[0036] The classification module 130 can generate classification images of at least some of the images in the plurality of images 112 and / or classification images of at least some of the images in the foreground images. In some embodiments, the classification images can be determined based on predetermined ball game rules 98. In some embodiments, each image in the classification images can include a plurality of pixel blocks, wherein each pixel block in the plurality of pixel blocks can be represented as an object of a specific category in a predetermined category of objects related to the ball game. For example, in a football game, the predetermined category of objects can include first team players, second team players, a ball, and / or a referee. In some embodiments, the classification module 130 filters at least some of the images in the classification images so as to enhance the plurality of pixel blocks in its images.

[0037] System 100 may include event detection module 140. Event detection module 140 may be coupled to classification module 130 and / or receive classification images (e.g., Figures 1A to 1C shown).

[0038] The event detection module 140 can determine and track the location of the object of interest (or at least some of the objects of interest) during the ball game based on the plurality of images 112 or based on the classified images. For example, the event detection module 140 can track the location of the first team players, the second team players, and / or the ball during the ball game. In some embodiments, the event detection module 140 saves and stores the determined tracked locations of the objects of interest.

[0039] The event detection module 140 may also determine a first event representing a suspected specified rule-based event (e.g., a foul and / or scoring event) during a ball game based on predetermined ball game rules 98 and a first subset of images in the plurality of images 112 or a first subset of classified images.

[0040] For example, the event based on the specified rule may be a scoring event in a soccer game. In this case, the first event may include the ball completely crossing the goal line between the goal posts and under the goal crossbar, or at least a goal is scored. However, the first event may not necessarily be identified as a scoring event, because the first event may occur in the case of violating at least one predetermined ball game rule 98. For example, a player of the scoring team may have committed a foul, or the player who shot the ball into the goal may be in an offside position.

[0041] Nowadays, in ball games (e.g., soccer games), the final decision on rule-based events is made by a referee. Therefore, it is necessary to detect the referee's decision on a first event to determine whether the first event is actually the event based on the specified rules. The system 100 can further track (e.g., via the event detection module 140) subsequent events that may occur during the ball game, and can also determine the referee's decision on the first event (e.g., whether the first event is an event based on the specified rules) based on at least one of the subsequent events and the predetermined ball game rules 98.

[0042] The event module 140 may also determine a second subset of images in the plurality of images 112 that represent a second event, or a second subset of classified images that represent a second event, wherein the second event follows the event based on the specified rule according to the predetermined ball game rules 98. For example, the event based on the specified rule may be a scoring event in a soccer game and the first event may be suspected to be its scoring event (e.g., as described above). In this case, the second event may include restarting the game from the center point of the ball game field 90 (e.g., kickoff).

[0043] In some embodiments, the database 110 includes audio signals acquired during a ball game. The event detection module 140 can receive the audio signal from the database 110 and determine a baseline power of the audio signal during the ball game. In some embodiments, the event detection module 110 can further determine a deviation of the audio signal power from the baseline power, and further determine a first event suspected of being an event based on a specified rule based on the deviation of the audio signal power. For example, if the audio signal power exceeds a predetermined threshold of the baseline power, in some embodiments, this may be an indication of an event based on a specified rule.

[0044] The system 100 may include an event analysis module 150. The event analysis module 150 may be coupled to the event detection module 140. The event analysis module 150 may analyze the second subset of images based on the predetermined ball game rules 98 and further determine a referee's ruling regarding the first event based on the analysis of the second subset of images (e.g., whether the first event is an event based on the specified rules).

[0045] The event analysis module 150 may determine at least one reference pattern of a second event based on the predetermined ball game rules 98, wherein the second event is after the event based on the specified rules according to the predetermined ball game rules 98. The event analysis module 150 may further analyze the second subset of images (e.g., representing the second event) based on the at least one reference pattern to verify whether the second event conforms to the at least one reference pattern. The event analysis module 150 may further determine the referee's ruling on the first event after its verification (e.g., whether the first event suspected of being an event based on the specified rules is actually an event based on the specified rules).

[0046] For example, Figure 2A At least one reference pattern of a second event is shown, including restarting the game from the center point 92 of the ball game field 90 after a scoring event occurs in a soccer game. In this case, the at least one reference pattern of the second event may include at least the following: (i) the ball 93 is at the center point 92 of the ball game field 90; (ii) there are only two players 94a of the same team 94 in the center circle 92a of the field 90 and the ball 93 is between the two players 94a; (iii) the first team players 94 and the second team players 95 are stationary (or substantially stationary); and / or (iv) the first team players 94 and the second team players 95 are both in opposite halves of the ball game field 90 (e.g., as shown in FIG. 1 ). Figure 2A ).

[0047] In another example, Figure 2B At least one reference pattern of a second event is shown, including a penalty corner kick after a foul by a defensive team 95 in a penalty circle 96 of a ball playing field 90 in a field hockey game. In this case, at least one reference pattern of the second event may include at least the following: (i) up to 5 defensive players are arranged behind the end line, either in the goal or on the end line at least 5 meters away from the ball; (ii) all other players of the defensive team 95 are behind the center line of the ball playing field 90; (iii) one offensive player 94a positions himself on the end line, and the ball 93 is in a circle at least 10 meters away from the goal posts closest to the sides of the goal; (iv) the remaining offensive players 94 position themselves on the field outside the shooting circle; and (v) all players on the end line except the offensive player 94a are prohibited from having any part of their body or stick touch the ground in the circle or on the center line 97 until the ball 93 is kicked (e.g., as shown in FIG. 1 ). Figure 2B ).

[0048] In various embodiments, the event analysis module 150 determines which team (e.g., the first team or the second team) of the two teams in the game has performed the first event (e.g., determined by the referee as an event based on the specified rules) based on the images of the second subset representing the second event, the predetermined ball game rules 98, and / or at least one reference pattern of the second event. Taking the scoring event in the football game as an example (e.g., as described above), the event analysis module 150 can determine the team that restarted the game from the center point of the court 90 based on the images of the second subset representing the second event (e.g., by determining the color of the clothes worn by the players in the center circle), and determine that the opponent team has scored a goal based on the predetermined ball game rules 98.

[0049] In some embodiments, the event analysis module 150 includes a plurality of at least one reference pattern corresponding to a plurality of specified rule-based events in a ball game (e.g., fouls, offsides, scores, etc.). In some embodiments, the reference pattern is determined based on predetermined ball game rules 98. In various embodiments, the event analysis module 150 uses a machine learning algorithm to determine and / or update its reference pattern.

[0050] System 100 may include an output module 160. Output module 160 may be coupled to event analysis module 150 and / or event detection module 140 (e.g., Figures 1A to 1C The output module 160 may generate at least one output related to the first event based on the determined referee's ruling regarding the first event.

[0051] In some embodiments, the at least one output includes a notification about the first event / the event based on the specified rule during the ball game. The notification can be delivered to at least one user of the system 100, for example.

[0052] In some embodiments, the at least one output includes at least one video clip representing the first event. For example, the output module 160 can receive images representing a first subset of the first event from the event detection module 140 or from the event analysis module 150, and further generate corresponding video clips representing the first event based on the images of the first subset. In some embodiments, the output module 160 can update the digitized table of events based on the specified rules based on the referee's ruling regarding the first event.

[0053] According to some embodiments, the system 100 is arranged to automatically detect a referee's decision regarding a scoring event during a ball game. The event detection module 140 may determine a first subset of images in the plurality of images representing a first event suspected of being a scoring event based on predetermined ball game rules 98 (e.g., as described above with respect to Figures 1A to 1CIn some embodiments, the first event occurs on a goal area (eg, one of two goal areas) of the ball game field 90.

[0054] The event detection module 140 may further determine a second subset of images in the plurality of images representing a second event based on a predetermined ball game rule, wherein the second event is after the scoring event (e.g., as described above with respect to Figures 1A to 1C In some embodiments, the second event occurs in a specific area of ​​the ball game field 90. In some embodiments, the specific area is not a goal area. For example, the specific area may be a predetermined area around the center point of the ball game field 90 (e.g., as described above with respect to Figure 2A described).

[0055] The event analysis module 150 can analyze the second subset of images based on the predetermined ball game rules and further determine the referee's ruling regarding the first event based on the analysis of the second subset of images (e.g., whether the first event is an event based on the specified rules; as described above with respect to Figures 1A to 1C described above).

[0056] Reference now Figure 3 , Figure 3 Flow chart of a method 200 for automatically detecting a referee's decision during a ball game according to some embodiments of the present invention.

[0057] The method 200 may be implemented by the system 100 which may be arranged to implement the method 200. It should be noted that the method 200 is not limited to Figure 3 For example, in various implementations, the method 200 need not move through each illustrated box or stage, or in exactly the same order as illustrated and described.

[0058] The method 200 may include receiving (stage 210) a plurality of images of a ball game field generated during a ball game. In various embodiments, the plurality of images are real-time images generated during an actual ball game, or the plurality of images are pre-recorded offline images during a ball game (e.g., as described above with respect to Figures 1A to 1C described).

[0059] In some embodiments, method 200 includes receiving (stage 212) a plurality of sets of images, wherein each set of images in the plurality of sets of images is generated during a different ball game in a corresponding plurality of ball games (e.g., as described above with respect to Figures 1A to 1C In some embodiments, method 200 includes simultaneously analyzing (stage 214) multiple sets of multiple images (e.g., as described above with respect to Figures 1A to 1Cdescribed).

[0060] The method 200 may include calibrating (stage 216) the plurality of images so that each pixel in the image is associated with a specific geographic location on the ball game field (e.g., via the calibration module 118, as described above with respect to Figures 1A to 1C described above).

[0061] The method 200 may include generating (stage 220) at least one background image based on the plurality of images or based on at least some of the plurality of images (e.g., by the classification module 130, as described above with respect to Figures 1A to 1C In some embodiments, the method 200 includes utilizing a machine learning algorithm to determine and / or update the at least one background image (stage 222).

[0062] In some embodiments, the method 200 includes generating (stage 224) a corresponding plurality of foreground images based on the plurality of images (or at least some of the plurality of images) and based on at least one background image (e.g., by the classification module 130, as described above with respect to Figures 1A to 1C described above).

[0063] The method 200 may include generating (stage 230) classified images of at least some of the plurality of images and / or classified images of at least some of the foreground images based on predetermined ball game rules (e.g., by the classification module 130, as described above with respect to Figures 1A to 1C In some embodiments, each image in the classification image includes a plurality of pixel blocks, wherein each pixel block in the plurality of pixel blocks represents a specific category of objects in a predetermined category of objects related to ball games (e.g., as described above with respect to Figures 1A to 1C described).

[0064] The method 200 may include determining and tracking (stage 232) the location of an object of interest (or at least some of the objects of interest) based on the plurality of images or based on the classified images during the ball game (e.g., by the event detection module 140, as described above with respect to Figures 1A to 1C described above).

[0065] The method 200 may include determining (stage 240) a first subset of images from the plurality of images or a first subset of classified images based on a predetermined ball game rule, the first subset representing a first event during the ball game that is suspected to be an event based on the specified rule (e.g., by the event detection module 140, as described above with respect to Figures 1A to 1C In some implementations, the event based on the specified rule is a scoring event.

[0066] The method 200 may include determining (stage 242) a second subset of images or classified images in the plurality of images that represent a second event, wherein the second event is subsequent to the specified rule-based event according to a predetermined ball game rule (e.g., by the event detection module 140, as described above with respect to Figures 1A to 1C described above).

[0067] In some embodiments, the method 200 includes receiving (stage 244) an audio signal acquired during a ball game, determining a baseline power of the audio signal, determining a deviation of the audio signal power from the baseline power, and further determining a first event suspected of being an event based on a specified rule based on the deviation of the audio signal power (e.g., by the event detection module 140, as described above with respect to Figures 1A to 1C described above).

[0068] The method 200 may include analyzing (stage 250) the second subset of images based on predetermined ball game rules and further determining a referee's ruling regarding the first event (e.g., whether the first event is an event based on specified rules) based on the analysis of the second subset of images (e.g., by the event analysis module 150, as described above with respect to Figures 1A to 1C described above).

[0069] In some embodiments, the method 200 includes determining (stage 252) at least one reference pattern of a second event based on a predetermined ball game rule, wherein according to the predetermined ball game rule, the second event is subsequent to the event based on the specified rule (e.g., by the event analysis module 150, as described above with respect to Figures 1A to 1C and FIG. 2A to FIG. 2B In some embodiments, method 200 includes utilizing a machine learning algorithm to determine and / or update a reference pattern of the second event (stage 253).

[0070] In some embodiments, method 200 includes analyzing (stage 254) a second subset of images (e.g., representing a second event) based on at least one reference pattern to verify whether the second event conforms to the at least one reference pattern (e.g., by event analysis module 150, as described above with respect to Figures 1A to 1C described above).

[0071] In some embodiments, the method 200 includes determining (stage 256) after the verification, a referee's ruling that the first event suspected of being an event based on the specified rule is in fact an event based on the specified rule (e.g., by the event analysis module 150, as described above with respect to Figures 1A to 1C described above).

[0072] In various embodiments, the method 200 includes determining (stage 258) based on the images representing the second subset of the second events, the predetermined ball game rules, and / or at least one reference pattern of the second event, one of the two teams (e.g., the first team or the second team) that performed the first event (e.g., determined by the referee to be an event based on the specified rule) (e.g., by the event analysis module 150, as described above with respect to Figures 1A to 1C described above).

[0073] The method 200 also includes generating (stage 260) at least one output related to the first event (e.g., via output module 160, as described above with respect to the first event) based on the determination by the referee that the first event is an event based on the specified rule. Figures 1A to 1C described above).

[0074] In some embodiments, the method 200 includes notifying the user associated with the first event (e.g., determined by the referee to be based on the specified rule) during the ball game (stage 262) (e.g., via the output module 160, as described above with respect to Figures 1A to 1C described above).

[0075] In some embodiments, method 200 includes generating (stage 264) at least one video segment representing the first event based on the first subset of images representing the first event (e.g., via output module 160, as described above with respect to Figures 1A to 1C described above).

[0076] In some embodiments, the event based on the specified rules is a scoring event and the method 200 includes updating (stage 266) a numerical score table (e.g., via the output module 160, as described above with respect to the first event) based on the referee's ruling regarding the first event. Figures 1A to 1C described above).

[0077] According to some embodiments, the method 200 can automatically detect a referee's decision regarding a scoring event during a ball game. The method 200 can include, for example, determining (stage 270) a first subset of images from a plurality of images based on predetermined ball game rules by the event detection module 140, the first subset of images representing a first event suspected of being a scoring event. (For example, as described above with respect to Figures 1A to 1C described).

[0078] The method 200 may include determining (stage 272) a second subset of images in the plurality of images representing a second event based on a predetermined ball game rule, such as by the event detection module 140, wherein the second event is after the scoring event according to the predetermined ball game rule (e.g., as described above with respect to Figures 1A to 1C described).

[0079] The method 200 may include analyzing (stage 274) the second subset of images based on predetermined ball game rules, for example, by the event analysis module 150, and further determining a referee's decision regarding the first event based on the analysis of the second subset of images (e.g., as described above with respect to Figures 1A to 1C described).

[0080] Advantageously, the disclosed system and method can realize automatic detection of referee's decisions regarding rule-based events during ball games, thereby overcoming the deficiencies of existing systems for analyzing sports events. The disclosed automatic system and method further realize simultaneous analysis of multiple ball games conducted at multiple ball game venues.

[0081] Various aspects of the present invention are described with reference to flowchart illustrations and / or partial diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It will be understood that each portion of the flowchart illustrations and / or partial diagrams, as well as combinations of portions in the flowchart illustrations and / or partial diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device for producing a machine, so that instructions executed by a processor of a computer or other programmable data processing device create functions / behaviors specified in the flowchart and / or partial diagrams or portions thereof.

[0082] These computer program instructions may also be stored in a computer-readable medium, which may direct a computer, other programmable data processing apparatus, or other device to function in a specific manner so that the instructions stored in the computer-readable medium produce an article of manufacture, which includes instructions for implementing the functions / behaviors specified in one or more portions of the flowchart and / or partial diagram. The computer program instructions may also be loaded into a computer, other programmable data processing apparatus, or other device so that a series of operational steps to be performed on the computer, other programmable apparatus, or other device produces a computer-implemented process, so that the instructions executed on the computer or other programmable device provide a process for implementing the functions / behaviors specified in one or more portions of the flowchart and / or partial diagram.

[0083] The foregoing flowcharts and illustrations illustrate the architecture, functionality and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each part in a flowchart or partial diagram may represent a part of a module, fragment or code, and the code includes one or more executable instructions for implementing a specific logical function. It should also be noted that, in some optional embodiments, the functions indicated in the part may occur in an order other than the order indicated in the accompanying drawings. For example, two parts shown in succession may actually be executed substantially simultaneously, or parts may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each part in a partial diagram and / or flowchart illustration, as well as a combination of parts in a partial diagram and / or flowchart illustration, may be implemented by a combination of a dedicated hardware-based system or dedicated hardware and computer instructions that performs a specified function or action.

[0084] In the above description, an embodiment is an example or implementation of the present invention. Each occurrence of "one embodiment", "an embodiment", "certain embodiments" or "some embodiments" does not necessarily all refer to the same embodiment. Although the various features of the present invention can be described in the context of a single embodiment, these features can also be provided individually or in any suitable combination. On the contrary, although the present invention can be described in the context of separate embodiments herein for the sake of clarity, the present invention can also be implemented in a single embodiment. Certain embodiments of the present invention can include features from different embodiments disclosed above, and certain embodiments can be combined with elements from other embodiments disclosed above. The disclosure of the present invention to elements in the context of a specific embodiment should not be understood as limiting its use to a separate specific embodiment. Moreover, it should be understood that the present invention can be accomplished or practiced in various ways, and the present invention can be implemented in certain embodiments different from the embodiments set forth in the above description.

[0085] The present invention is not limited to these figures or corresponding descriptions. For example, the process need not move through each shown box or state, or move in exactly the same order as shown and described. The meaning of the technical and scientific terms used herein is the meaning commonly understood by those of ordinary skill in the art to which the present invention belongs, unless otherwise specified. Although the present invention is described with respect to a limited number of embodiments, these should not be interpreted as limitations on the scope of the present invention, but should be examples of some preferred embodiments. Other possible variations, modifications and applications also fall within the scope of the present invention. Accordingly, the scope of the present invention should not be limited by what has been described so far, but by the attached claims and their legal equivalents.

Claims

1. A system for automatically detecting a referee's decision during a ball game, the system comprising: at least one camera operable to capture a plurality of images of a ball game field in real time during said ball game; An event detection module, wherein the event detection module is configured as follows: receiving, by the at least one camera, the plurality of images of the ball game field captured during the ball game; determining a first subset of images in the plurality of images based on predetermined ball game rules, the first subset of images representing a first event suspected to be an event based on the specified rule; as well as determining a second subset of images of the plurality of images representing a second event based on the predetermined ball game rules, wherein the second event is temporally subsequent to the event based on the specified rules according to the predetermined ball game rules; as well as An event analysis module, coupled to the event detection module, wherein the event analysis module is configured as follows: determining at least one reference pattern of the second event based on the predetermined ball game rules; analyzing the second subset of images based on the at least one reference pattern to verify whether the second event conforms to the at least one reference pattern; as well as Based on the verification, it is determined whether the first event suspected to be an event based on a specified rule is actually the event based on the specified rule.

2. The system according to claim 1, further comprising a classification module, which generates a classification image of the image based on at least some of the multiple images and the predetermined ball game rules.

3. The system according to claim 2, wherein: Each of the classified images includes a plurality of pixel blocks, and wherein each of the plurality of pixel blocks is represented as an object of a specific category among predetermined categories of objects related to the ball game.

4. The system according to claim 1 further includes an output module coupled to the event detection module and the event analysis module, wherein the output module generates at least one output related to the first event based on the referee's ruling regarding the first event.

5. The system according to claim 4, wherein: The at least one output includes at least one of: a notification regarding the first event during the ball game, at least one video clip representing the first event, an update to a digitized table of specified rule-based events.

6. The system of claim 1, further comprising at least one set of at least one camera located at the ball game venue, wherein the at least one camera generates a plurality of real-time images of the ball game venue during an actual ball game.

7. The system according to claim 1 further includes multiple sets of at least one camera, wherein each of the multiple sets of the at least one camera is located at a different ball game venue among the corresponding multiple ball game venues, and is configured to generate corresponding multiple groups of multiple images of the corresponding multiple ball games at the corresponding multiple ball game venues, and wherein the system simultaneously analyzes the multiple groups of multiple images.

8. A method for automatically detecting a referee's decision during a ball game, the method comprising: receiving, from at least one camera, a plurality of images of a ball game field captured in real time during the ball game; determining a first subset of images in the plurality of images based on predetermined ball game rules, the first subset of images representing a first event suspected to be an event based on the specified rule; determining a second subset of images of the plurality of images representing a second event based on the predetermined ball game rules, wherein the second event is temporally subsequent to the event based on the specified rules according to the predetermined ball game rules; determining at least one reference pattern of the second event based on the predetermined ball game rules; analyzing the second subset of images based on the at least one reference pattern to verify whether the second event conforms to the at least one reference pattern; as well as Based on the verification, it is determined whether the first event suspected to be a designated rule-based event is actually the designated rule-based event.

9. The method according to claim 8, further comprising generating, by a classification module, a classification image of the image based on at least some of the plurality of images and the predetermined ball game rules.

10. The method according to claim 9, wherein: Each of the classified images includes a plurality of pixel blocks, and wherein each of the plurality of pixel blocks is represented as an object of a specific category among predetermined categories of objects related to the ball game.

11. The method of claim 8, further comprising generating, by an output module, at least one output related to the first event based on the referee's ruling regarding the first event.

12. The method according to claim 11, wherein: The at least one output includes at least one of: a notification regarding the first event during the ball game, at least one video clip representing the first event, an update to a digitized table of specified rule-based events.

Images