Description:ECHNOLOGICAL FIELD
[0001]
The present disclosure generally relates to sport accessories, and more particularly relates to a system and a method for analysing player’s arm 5 related sports data.
BACKGROUND
[0002]
Cricket is a popular sport that originated in England and is now played in many countries around the world. It is a bat-and-ball game played between two teams of 11 players each. The objective of the game is to score 10 more runs than the opposing team while also taking all 10 of the opposing team's wickets.
[0003]
At the beginning of the game, a coin toss is held to decide which team will bat first. The team that wins the toss can either choose to bat first or bowl first. The team that bats first sends two batsmen out to the field to face the first 15 ball, while the opposing team sends out their bowler to bowl the first over.
[0004]
The batting team scores runs by hitting the ball and running back and forth between two sets of wickets, which are located at opposite ends of the rectangular playing field, known as the pitch. If the ball is hit to the boundary of the field, the batting team scores four runs, and if it is hit over the boundary 20 without touching the ground, the team scores six runs.
[0005]
The fielding team tries to prevent the batting team from scoring runs by fielding the ball and trying to get the batsmen out. The most common way to get a batsman out is by hitting the wickets with the ball, known as a "bowled" dismissal. Other ways include catching the ball after the batsman hits it, "run 25 out" when a fielder throws the ball and hits the wickets while a batsman is attempting to run, and more.
3
[0006]
Once a team loses all 10 wickets or reaches the end of their allotted overs, the teams switch roles, with the fielding team becoming the batting team and vice versa. The team with the most runs at the end of the game wins. A typical game of cricket can last anywhere from a few hours to several days, depending on the format of the game. 5
[0007]
Cricket has seen the adoption of various technologies to improve the accuracy of decision-making by umpires and enhance the viewing experience of fans. Some of the technologies used in cricket include:
[0008]
Decision Review System (DRS): This system uses ball-tracking and edge-detection technology to help umpires make more accurate decisions 10 regarding dismissals, such as lbw (leg before wicket) and caught behind. Each team is allowed to challenge a limited number of decisions per innings.
[0009]
Hawk-Eye: This technology tracks the trajectory of the ball and predicts its path, which helps umpires make decisions on lbw appeals and gives fans a better view of close decisions. 15
[0010]
Snickometer: This technology uses audio sensors and cameras to detect faint edges off the bat, which helps umpires make decisions on caught-behind appeals.
[0011]
Hotspot: This technology uses infrared cameras to detect whether the ball has made contact with the bat or pad, which helps umpires make decisions 20 on lbw appeals and caught-behind appeals.
[0012]
LED stumps and bails: These are electronic versions of the traditional wooden stumps and bails. They light up when the ball hits them, which helps umpires make decisions on run-outs and stumpings.
[0013]
Overall, the use of technology in cricket has improved the accuracy 25 of decision-making and added to the excitement of the game.
4
[0014]
Bowling is one of the two main disciplines in the game of cricket, alongside batting. Bowling involves the act of delivering the ball to the batsman with the intention of dismissing them and restricting the scoring opportunities for the opposing batting team.
[0015]
In cricket, there is one bowler on the field at any given time, aiming 5 to deliver the ball to the batsman. The bowler's primary objective is to take wickets, which means getting the batsmen out, while also maintaining a tight line and length to limit the batsman's ability to score runs.
[0016]
Bowling requires a combination of skill, technique, and strategy. A good bowler must possess attributes such as accuracy, pace, swing, spin, and 10 variations in delivery. They need to bowl in the right areas on the pitch, exploit the weaknesses of the batsman, and create opportunities for their team.
[0017]
Bowlers use various types of deliveries to outfox the batsman, including fast bowling, swing bowling, seam bowling, spin bowling, and variations like leg spin, off-spin, googly, and doosra. They analyze the pitch 15 conditions, batsman's style, and field placements to decide the most effective delivery to bowl.
[0018]
Some common strategies employed by bowlers include bowling a yorker to target the batsman's toes, bowling a bouncer to surprise and intimidate the batsman, or using spin to deceive and create confusion. Bowlers also rely on 20 their ability to consistently hit the right line and length to build pressure on the batsman.
[0019]
There are several technologies that are used to help cricketers improve their bowling skills. Some of these technologies include:
[0020]
Video analysis: Bowlers may record their bowling sessions and 25 analyze the footage to identify areas for improvement. Coaches and players can study the video to assess bowling actions, identify faults, and work on correcting
5
them. Video analysis
may also help bowlers understand their strengths and weaknesses, adjust their strategies, and learn from successful bowlers.
[0021]
Performance tracking sensors: Wearable sensors may be utilized by bowlers to measure their run-up speed, release point, and bowling action. These sensors can provide valuable data on the bowler's technique and help identify 5 any inconsistencies or areas that need improvement. Bowlers can also track their progress over time using these performance tracking sensors.
[0022]
Virtual reality: Bowlers can utilize virtual reality technology to simulate match scenarios. They can practice bowling against different types of batsmen in various pitch conditions, allowing them to develop their decision-10 making skills, learn to adapt to different situations, and improve their accuracy and variations.
[0023]
Some sports equipment manufacturers have developed “smart” balls that include sensors to measure the revolutions, speed, and movement of the ball after it leaves the bowler's hand. This data can provide information related to the 15 quality of each delivery, helping bowlers to identify areas for improvement and refine their techniques.
[0024]
Overall, these technologies are used to assist cricketers in improving their bowling skills, enhancing their performance, and contributing to the success of their team on the field. 20
[0025]
Sports such as cricket is a professional game. Players participating in such game must remain fit for delivering performance to stay in the game. For improving their own performance, it is firstly important for the players to understand their own performance and also understand their strengths and weaknesses during each game. There are existing solutions in the market which 25 utilize motion sensor/s to sense actions such as batting or bowling deliveries by a player. However, the existing technologies, which mostly work in controlled laboratory conditions, use bulky equipment and are expensive. However, these
6
existing solutions are
not always able to accurately detect each action of the player and thereby, resulting in false positives and false negatives. Due to the false positives and false negatives, the players does not get accurate data about their performance. In absence of such accurate data about the performance, the players do not know or understand their performance well and thus, are also 5 unable to curate a better improvement plan for their performance.
[0026]
Therefore, there is a need for a system and a method for automatically analysing a player’s arm related sports data. There is also a need for a system and a method for determining performance and improvement data of the player based on the analysis of the player’s arm related sports data. 10
BRIEF SUMMARY
[0027]
Accordingly, there is a need for automatically analysing a player’s arm related sports data and determining performance as well as improvement data of the player based on the analysis of the player’s arm related sports data. In order to analyse the player’s arm related sports data, it is important to use one 15 or more sensors to sense data of a player while playing and analyse the sensed data for curating a performance plan for a player.
[0028]
Some example embodiments disclosed herein provide a system analysing a player’s arm related sports data. The system comprises one or more sensors to collect sports data and a filtering module to filter the collected sports 20 data. The system also comprises an identification module to identify the action from the collected sports data. Further, the action is at least one from the group comprising bowling, throwing, and pitching. The system further comprises a classification module to perform classification of the action into a type.
[0029]
According to some example embodiments, the action comprises 25 bowling of cricket sport.
7
[0030]
According to some example embodiments, the type comprises at least one from a group comprising pace, spin, inswing, outswing, length, short, bouncer, wide, leg-break, off-break, googly, doosra, arm-ball, carrom-ball, flipper, and top-spinner.
[0031]
According to some example embodiments, the system further 5 comprises counting run-up steps of the player. According to some example embodiments, the system further comprises tracking run-up speed using the sensors. According to some example embodiments, the system further comprises counting total steps of the player. According to some example embodiments, the system further comprises aggregating number of balls bowled by the player. 10 According to some example embodiments, the system further comprises indicating bowling load and injury stress using the total steps, the run-up steps, the run-up speed, gait analysis, landing stress, running symmetry, and the number of balls bowled.
[0032]
According to some example embodiments, the system further 15 comprises detecting posture, three dimensional co-ordinates, hand speed, ball grip and follow through distance of the player’s action.
[0033]
According to some example embodiments, one or more sensors comprise at least one of accelerometers, gyroscopes, magnetometers, piezoelectric sensors, visual sensors, electromagnetic trackers, and optical 20 sensors.
[0034]
According to some example embodiments, the system further comprises tracking a ball released by the player’s arm using the visual sensors and optical sensors.
[0035]
According to some example embodiments, the system further 25 comprises marking spot on a pitch where the ball has bounced. According to some example embodiments, the system further comprises aggregating a plurality of spots on the pitch where the ball had bounced to develop a pitch map.
8
[0036]
According to some example embodiments, the classification further comprising using at least a deep learning model comprising at least a recurrent neural network using at least a gated recurrent unit, attention-based long short-term memory, and convolutional neural network.
[0037]
According to some example embodiments, the system further 5 comprises a voice command feature for the bowler using an audio sensor attached to the wearable device. According to some example embodiments, the system further comprises audio feedback using a speaker/transducer to convey bowling actions and session level statistics to the bowler.
[0038]
Some example embodiments disclosed herein provide a method for 10 analysing a player’s arm related sports data. The method comprises the steps of collecting the sports data of at least one action of the player by using at least one sensor. The method also comprises the steps of filtering the collected data. The method further comprises the steps of identifying the action from the collected sports data, wherein the action is at least one from the group comprising bowling, 15 throwing, and pitching. The method further comprises the steps of performing classification of the action into a type.
[0039]
The foregoing summary is illustrative only and is not intended to be in any way limiting. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features will 20 become apparent by reference to the drawings and the following detailed description.
BRIEF DESCRIPTION OF THE DRAWINGS
[0040]
Having thus described example embodiments of the invention in general terms, reference will now be made to the accompanying drawings, which 25 are not necessarily drawn to scale, and wherein:
9
[0041]
FIG. 1A-1E illustrates a system architecture for analysing a player’s arm related sports data, in accordance with different embodiments;
[0042]
FIG. 2 illustrates an exemplary communication between a wearable device having an electronic circuitry attached or embedded thereto and a computing device, in accordance with an example embodiment; 5
[0043]
FIG. 3 illustrates a block diagram of an electronic circuitry for analysing a player’s arm related sports data, in accordance with an example embodiment;
[0044]
FIG. 4 illustrates a block diagram for building and training an artificial intelligence (AI) model for performance analysis of a player, in 10 accordance with an example embodiment;
[0045]
FIG. 5 illustrates a block diagram of deep learning algorithms utilized for building and training the AI model, in accordance with an example embodiment;
[0046]
FIG. 6 illustrates a flow diagram of a method for analysing a player’s 15 arm related sports data, in accordance with an example embodiment;
[0047]
FIG. 7 illustrates a flow diagram of a method for determining bowling load and injury stress of a player, in accordance with an example embodiment;
[0048]
FIG. 8 illustrates a flow diagram of a method for tracking a ball 20 released by the player’s arm, in accordance with an example embodiment;
[0049]
FIG. 9 illustrates a flow diagram of a method for developing a pitch map based on tracking a ball, in accordance with an example embodiment;
[0050]
FIG. 10 illustrates a flow diagram of a method for enabling voice command and voice feedback features on a wearable device, in accordance with 25 an example embodiment.
10
DETAILED DESCRIPTION
[0051]
In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. It will be apparent, however, to one skilled in the art that the present disclosure can be practiced without these specific details. In other 5 instances, systems, apparatuses, and methods are shown in block diagram form only in order to avoid obscuring the present disclosure.
[0052]
Reference in this specification to “one embodiment” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one 10 embodiment of the present disclosure. The appearance of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Further, the terms “a” and “an” herein do not denote a limitation of quantity, but rather denote the presence of at least 15 one of the referenced items. Moreover, various features are described which may be exhibited by some embodiments and not by others. Similarly, various requirements are described which may be requirements for some embodiments but not for other embodiments.
[0053]
Some embodiments of the present invention will now be described 20 more fully hereinafter with reference to the accompanying drawings, in which some, but not all, embodiments of the invention are shown. Indeed, various embodiments of the invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will satisfy applicable 25 legal requirements. Like reference numerals refer to like elements throughout. As used herein, the terms “data,” “content,” “information,” and similar terms may be used interchangeably to refer to data capable of being transmitted, received and/or stored in accordance with embodiments of the present invention.
11
Thus, use of any such terms should not be taken to limit the spirit and scope of
embodiments of the present invention.
[0054]
The embodiments are described herein for illustrative purposes and are subject to many variations. It is understood that various omissions and substitutions of equivalents are contemplated as circumstances may suggest or 5 render expedient but are intended to cover the application or implementation without departing from the spirit or the scope of the present disclosure. Further, it is to be understood that the phraseology and terminology employed herein are for the purpose of the description and should not be regarded as limiting. Any heading utilized within this description is for convenience only and has no legal 10 or limiting effect.
[0055]
Embodiments of the present disclosure may provide a system and a method for analysing a player’s arm related sports data. The system and the method analysing the player’s arm related sports data in such an improved manner are described with reference to FIG.1A to FIG. 10 as detailed below. 15
[0056]
FIG. 1A illustrates a system architecture 100A for analysing a player’s arm related sports data, in accordance with an example embodiment. A smart cricket ball 102 of a player 114 is shown. As can be seen, the smart cricket ball 102 may include an outer shell unit containing a core. Also, an electronic circuitry 104 may be removably attached or embedded/integrated within the core 20 of the smart cricket ball 102. It will be apparent to a person skilled in the art that the electronic circuitry 104 may be removably attached or embedded/integrated within the core of any other ball of sports such as baseball, American football, and the like.
[0057]
The electronic circuitry 104 may be capable of collecting and 25 processing sports data of the player 114. Such sports data may include events or activities such as ball movement, speed, spin, impact, and/or any such activity related to the ball. The electronic circuitry 104 may be capable of processing the collected sports data to filter out the noise or irrelevant data and then analyze the
12
sport
data for performance analysis. The collecting and processing of sports data for performance analysis of the player 114 by the electronic circuitry 104 is explained in greater details in FIG. 2 to FIG. 10 below.
[0058]
Moreover, the electronic circuitry 104 may also be in communication with a computing device 112 through a network 116. In this scenario, the 5 electronic circuitry 104 may communicate or transmit the performance analysis and/or filtered sensor data of the player 114 to the computing device 112 for display or real-time updates and/or further processing to get performance analytics and deeper player performance insights. Such computing device 112 may belong to the player 114 or may belong to a cricket board/committee or may 10 belong to a cricket team for which the player 114 is playing.
[0059]
Further, the electronic circuitry 104 may be in communication with a server 108 through a network 106. Herein, the electronic circuitry 104 may communicate with the server 108 for transmitting the collected sports data to the server 108. In this exemplary embodiment, instead of the electronic circuitry 15 104, the server 108 processes the collected sports data to determine performance analysis of the player 114. In a preferred embodiment, some of the data processing is carried out at a mobile device.
[0060]
Furthermore, the server 108 may be in communication with the computing device 112 through a network 110. The server 108 may process the 20 collected data to determine the performance analysis of the player 114. Then, the server 108 may also be capable of transmitting the performance analysis of the player 114 to the computing device 112.
[0061]
As used herein, the term “network” may refer to a long-term cellular network (such as GSM (Global System for Mobile Communication) network, 25 LTE (Long-Term Evolution) network or a CDMA (Code Division Multiple Access) network) or a short-term network (such as Bluetooth network, Wi-Fi network, NFC (near-field communication) network, LoRaWAN, ZIGBEE or Wired networks (like LAN, el all) etc.).
13
[0062]
As used herein, the term “computing device” may refer to a mobile phone, a personal digital assistance (PDA), a tablet, a laptop, a computer, VR Headset, Smart Glasses, or any such device capable of rendering performance analysis of the player 114.
[0063]
As used herein, the term ‘electronic circuitry’ (as explained in FIG.2 5 below) may refer to (a) hardware-only circuit implementations (for example, implementations in analog circuitry and/or digital circuitry); (b) combinations of circuits and computer program product(s) comprising software and/or firmware instructions stored on one or more computer readable memories that work together to cause an apparatus to perform one or more functions described 10 herein; and (c) circuits, such as, for example, a microprocessor(s) or a portion of a microprocessor(s), that require software or firmware for operation even if the software or firmware is not physically present. This definition of ‘circuitry’ applies to all uses of this term herein, including in any claims. As a further example, as used herein, the term ‘circuitry’ also includes an implementation 15 comprising one or more processors and/or portion(s) thereof and accompanying software and/or firmware. As another example, the term ‘circuitry’ as used herein also includes, for example, a baseband integrated circuit or applications processor integrated circuit for a mobile phone or a similar integrated circuit in a server, a cellular network device, other network device, and/or other computing 20 device.
[0064]
FIG. 1B illustrates a system architecture 100B for analysing a player’s arm related sports data, in accordance with an example embodiment. In this exemplary embodiment, the electronic circuitry 104 may be attached or embedded with clothing of the player 114. For an example, the electronic 25 circuitry 104 may be attached or embedded to a t-shirt (i.e., upper half clothing) worn by the player 114.
[0065]
FIG. 1C illustrates a system architecture 100C for analysing a player’s arm related sports data, in accordance with an example embodiment. In
14
this exemplary
embodiment, the electronic circuitry 104 may be attached or embedded with the clothing of the player 114 at a different position. For an example, the electronic circuitry 104 may be attached or embedded to a pair of trousers or pants (i.e. lower half clothing) worn by the player 114.
[0066]
FIG. 1D illustrates a system architecture 100D for analysing a 5 player’s arm related sports data, in accordance with an example embodiment. As an example, the electronic circuitry 104 may be directly attached to (i.e. worn by) the player 114 on his/her wrist or hand. In another example, the electronic circuitry 104 may be directly attached to (i.e. worn by) the player 114 on his/her leg or thigh. 10
[0067]
FIG. 1E illustrates a system architecture 100E for analysing a player’s arm related sports data, in accordance with an example embodiment. In this exemplary embodiment, one or more imaging units (such as cameras) 120A-120B may be installed in a cricket field for capturing videos and images of the player 114. A person skilled in the art would not limit the camera units to be only 15 on the field. The camera units may be a part of the mobile device 112 also. Such captured videos and images of the player 114 may be transmitted to the electronic circuitry 104 or to the server 108 for detecting posture detection of the player 114, motion of at least a sports equipment (such as ball 102), three dimensional co-ordinates of a body part of the player 114 from the posture, hand 20 speed of the player 114, ball grip and follow through distance of the player’s action.
[0068]
In some embodiments, the one or more imaging units (such as cameras) 120A-120B are installed or positioned in stumps of a bowler's delivery side. In some other embodiments, the one or more imaging units (such as 25 cameras) 120A-120B are installed or positioned at different positions of the cricket field. Although FIG. 1E shows only two imaging units; however, using more than two imaging units or less than two imaging units is also within the scope of the present disclosure.
15
[0069]
FIG. 2 illustrates an exemplary communication between a wearable device 118 having an electronic circuitry 104 attached or embedded thereto and a computing device 112, in accordance with an example embodiment. As shown in FIG. 2, the wearable device 118 may have the electronic circuitry 104. In an exemplary embodiment, the electronic circuitry 104 is removably attached with 5 the wearable device 118. In another exemplary embodiment, the electronic circuitry 104 is embedded or integrated into the wearable device 118. Further, the wearable device 118 may be communicably coupled with the computing device 112 via a short-range network such as Bluetooth or NFC (near-field communication). This wearable device 118 may be worn by the player 114 either 10 at his wrist or hand or thigh or leg while playing. In some embodiments, the wearable device 118 may also process the collected sports data for performance analysis of the player 114 and communicate the performance analysis of the player 114 directly to the computing device 112 via the short-range network.
[0070]
Moreover, the wearable device 118 may include a camera, a 15 microphone, LiDAR (Light Detection and Ranging) sensors and any such module or component well known in the art. As used herein, the wearable device 118 may refer to a smart watch, a smart band, a smart clothing or any such wearable device that is obvious to a person skilled in the art.
[0071]
FIG. 3 illustrates a block diagram of an electronic circuitry 104 for 20 analysing sports data, in accordance with an example embodiment. The electronic circuitry 104 may include, but is not limited to, an interface 302, a receiver 304, a transmitter 306, one or more sensors 308, an identification module 310, a classification module 312, a filtering module 314, a processor 316, a tracking module 318, a memory 320, the voice command module 322, 25 and the audio feedback module 324.
[0072]
The interface 302 may be configured to receive an input from the player 114 or any other user. The input may include various types of information related to the sports activity being performed. Additionally, the interface 302 is
16
also capable of outputting performance analysis of the player 112. This means
that it may provide feedback regarding the player's performance based on the analysis of the collected sports data.
[0073]
The receiver 304 may be configured to receive any kind of information and/or data from the computing device 112 and/or the server 108. 5 Such information and/or data may include notifications, performance analysis or any other relevant data that is being received by the electronic circuitry 104.
[0074]
The transmitter 306 may be configured to transmit any kind of information and/or data to the computing device 112 and/or the server 108. Such information and/or data may include performance analysis etc. 10
[0075]
The one or more sensors 308 may be configured to collect sports data of the player 114. As explained in FIG. 1 above, the sports data may include events or activities such as, ball movement, speed, spin, impact, and/or any such activity related to the ball. The one or more sensors 308 may be comprise at least one of accelerometers, gyroscopes, magnetometers, piezoelectric sensors, visual 15 sensors, electromagnetic trackers, optical sensors, and any such that is obvious to a person skilled in the art. For an example, the accelerometer may sense speed at which the player 114 is running while delivering the ball, step count or ball speed.
[0076]
In a more elaborative way, the collected sports data may include 20 various parameters and information related to the bowler's performance. The collected data for a bowler may be:
[0077]
Running: Data related to the bowler's running activity, including metrics such as the number of steps taken, speed, acceleration, and distance covered during the run-up to deliver the ball. 25
17
[0078]
Arm Speed: Data capturing the speed at which the bowler's arm moves during the delivery of the ball, which is an important factor in determining the pace and effectiveness of the delivery.
[0079]
Run-Up Steps: Information about the number of steps the bowler takes during the run-up to the bowling crease before releasing the ball. This may 5 provide information of the bowler's rhythm and approach.
[0080]
Run-Up Speed: Data indicating the speed at which the bowler approaches the bowling crease during the run-up, which may impact the delivery and generate additional momentum.
[0081]
Ball Release Parameters: Information about the specific parameters 10 of the ball release, including release angle, wrist rotation, and arm angle at release. These factors influence the trajectory, swing, and spin of the delivered ball.
[0082]
Ball Type: Data classifying the type of delivery, such as spin (off spin, leg spin, doosra), fast (medium, fast, swing, cutters), or variations specific 15 to bowling techniques.
[0083]
Accuracy: Metrics assessing the bowler's ability to consistently deliver the ball in the desired areas, such as hitting the right length and line according to the intended strategy.
[0084]
Bowling Technique: Information related to the bowler's technique, 20 including foot placement, body alignment, and follow-through distance. This data helps in analysing the efficiency and effectiveness of the bowling action.
[0085]
Performance Analysis: Derived information and analysis obtained from the collected data, such as average speed, line and length analysis, variations effectiveness, and overall bowling performance metrics. It helps in 25 evaluating the bowler's strengths, weaknesses, and areas for improvement.
18
[0086]
The one or more sensors 308 may be communicably coupled with the filtering module 314 to communicate the collected sports data of the player 114. The filtering module 314 may be configured to filter the collected sports data. Signal processing and time-series analysis is used to filter the collected sports data. That is, time-series techniques such as signal processing, time-series 5 analysis, and statistical modelling are used to extract relevant features from the collected sports data to describe the motion pattern and frequency components. Also, time-series analysis techniques are used to extract features from the collected sports data, such as trend analysis, seasonal decomposition, and autocorrelation from the collected sports data. 10
[0087]
There are various signal processing techniques that can be used to process motion sensor data. Some of the commonly used techniques include:
[0088]
Filtering: Filtering is a technique used to remove unwanted noise or artifacts from sensor data. A popular type of filter used in motion sensor data processing is the Kalman filter, which can be used to estimate the true value of 15 a signal by combining noisy measurements with a dynamic model.
[0089]
Feature extraction: Feature extraction involves identifying relevant features or patterns in the sensor data. In motion sensor data processing, this might involve extracting features related to the frequency, amplitude, or direction of movement. 20
[0090]
Time-frequency analysis: Time-frequency analysis techniques, such as the wavelet transform and Fourier transforms, can be used to analyze the spectral content of motion sensor data over time. This can be useful for identifying patterns or changes in movement over time.
[0091]
Fourier transform is a mathematical technique that is commonly used 25 to extract features from sensor data. The Fourier transform converts a time-domain signal into its frequency-domain representation, which can be useful for
19
identifying patterns or features in the signal that may not be easily visible in the
time-domain.
[0092]
To extract features from sensor data using Fourier transform, the following steps are typically followed:
[0093]
Pre-processing: The sensor data is typically pre-processed to remove 5 noise, artifacts, or other unwanted signals that may interfere with the analysis.
[0094]
Windowing: The sensor data is segmented into smaller sections, or windows, which are then analyzed separately. This helps to avoid issues related to spectral leakage or edge effects.
[0095]
Fourier transform: The Fourier transform is applied to each window 10 of sensor data, which results in a frequency-domain representation of the signal. This representation can be visualized using a power spectrum, which shows the distribution of power across different frequencies.
[0096]
Feature extraction: Features can be extracted from the power spectrum to identify patterns or characteristics in the signal. For example, the 15 frequency of the peak in the power spectrum may be used as a feature to identify a specific vibration frequency in the signal.
[0097]
Classification: The extracted features can be used to classify the sensor data into different categories or classes.
[0098]
Overall, Fourier transform is a powerful technique for analyzing 20 sensor data and extracting features that can be used for a wide range of applications in fields such as engineering, healthcare, and environmental monitoring. Wavelet transform is a mathematical technique that is commonly used to extract features from sensor data. Unlike Fourier transform, which only provides frequency-domain information, wavelet transform provides both time-25 domain and frequency-domain information. This makes it a useful technique for analyzing non-stationary signals that vary in frequency and amplitude over time.
20
[0099]
To extract features from sensor data using wavelet transform, the following steps are typically followed:
[00100]
Pre-processing: The sensor data is typically pre-processed to remove noise, artifacts, or other unwanted signals that may interfere with the analysis.
[00101]
Wavelet transform: The wavelet transform is applied to the sensor 5 data, which results in a time-frequency representation of the signal. This representation can be visualized using a spectrogram, which shows the distribution of power across different frequencies and time.
[00102]
Feature extraction: Features can be extracted from the spectrogram to identify patterns or characteristics in the signal. For example, the location and 10 amplitude of specific wavelet coefficients may be used as features to identify a specific pattern in the signal.
[00103]
Classification: The extracted features can be used to classify the sensor data into different categories or classes. For example, the presence of a specific pattern in the signal may be used to classify the health of a machine 15 component.
[00104]
In summary, wavelet transform is a powerful technique for analyzing sensor data and extracting features that can be used for a wide range of applications in fields such as engineering, healthcare, and environmental monitoring. It is particularly useful for analyzing non-stationary signals that may 20 not be easily analyzed using Fourier transform.
[00105]
Overall, signal processing techniques are used to extract meaningful information from motion sensor data, which can be used for a wide range of applications in fields such as sports, healthcare, and robotics.
[00106]
Signal processing techniques are used to extract features from the 25 collected sports data, such as filtering, Fourier transforms, or wavelet transforms.
21
These techniques are used to extract features such as frequency, amplitude,
phase, and energy from the collected sports data.
[00107]
Such filtering may include removal of the noise from the collected sports data while retaining the relevant sports data. Noise that may be data redundant or not relevant for further processing is discarded at this stage. An 5 example of such irrelevant sports data may be data between a time interval after expiry of an existing over and before start of a new over. Such filtering of the collected sports data at an initial stage results in efficient processing of the collected sports data and would move focus only on the relevant data.
[00108]
In a preferred embodiment a peak recognition method is deployed 10 after analyzing bowling data. Peaks are identified on live data and when defined threshold limit is reached for a specific mode, then the data captured, and pattern recognition algorithm is employed to identify it as a bowling action.
[00109]
In some exemplary embodiments, the filtering of the collected sports data also include cleaning, formatting and normalizing the relevant sports data 15 for inputting to the next stage of the identification module 310.
[00110]
The one or more sensors 308 and the filtering module 314 may be communicably coupled with the identification module 310 for communicating filtered sports data and collected sports data. The identification module 310 may be configured to detect actions from the collected and/or filtered sports data. The 20 action comprises at least one from the group comprising bowling, throwing, and pitching.
[00111]
For identifying the action, the identification module 310 analyzes the collected sports data using predefined parameters and patterns associated with different actions of the player 114. For instance, in case of bowling the 25 identification module 310 analyzes the collected sport data related to arm speed, arm angle, wrist rotation, run-up steps, run-up speed, and step length to determine that the action corresponds to a bowling action. Further, in case of
22
throwing the identification module 310 examines the collected sport data related
to arm speed, arm angle, release point, and throwing trajectory to determine that the action corresponds to a throwing action. Additionally, in case of pitching the identification module 310 analyzes the collected sport data related to arm speed, arm angle, release point, and pitching trajectory to determine that the action of 5 the player 114 corresponds to a pitching action.
[00112]
By analyzing these parameters and patterns in the collected sports data, the identification module 310 may accurately identify the specific action performed by the player, whether it is a bowling action, throwing action, or pitching action. 10
[00113]
In some exemplary embodiments, based on identifying the action of the player 114, the identification module 310 may further be configured to detect a posture of the player 114, three dimensional co-ordinates of a body part of the player 114 from the posture, hand speed of the player 114 while delivering a ball, a ball grip and follow through distance of the player’s action. For an 15 instance, based on a feed from a camera, posture of the player 114 may be detected and also co-ordinates of the body of the player 114 may be detected. Additionally, in some embodiments, using the camera feed, an arm angle, arm speed of the player 114 may also be detected during the delivery of the ball.
[00114]
In addition to the identification module 310, the one or more sensors 20 308 may also be communicably coupled with the tracking module 318 for tracking various parameters of the player 114. In particular, the tracking module 318 may be configured to track the player's run-up steps and speed. By accurately counting the number of run-up steps taken by the player 114 and tracking the speed at which it approaches the bowling crease, the tracking 25 module 318 gathers necessary information of the player's approach and run-up dynamics. This information contributes to a deeper understanding of the player's technique and performance.
23
[00115]
Furthermore, the tracking module 318 may count total steps of the player 114 taken during the bowling session. This data may provide valuable information of the player's overall physical exertion and workload.
[00116]
Furthermore, by aggregating a number of balls bowled by the player 114, the tracking module 318 may contribute towards understanding of the 5 player's performance over time. These data (for example, the total steps, the run-up steps, the run-up speed, and the number of balls bowled) may be utilized to indicate bowling load and potential injury stress, offering important information for training and injury prevention strategies.
[00117]
In conjunction with its primary tracking functionality, the tracking 10 module 318 extends its capabilities to include the tracking of the ball after it is released by the player's arm. By utilizing the visual sensors and optical sensors, the tracking module 318 precisely follows the ball's path in real-time. As a result, the tracking module 318 may accurately mark the spot on the pitch where the ball has bounced, capturing important data on the ball's behavior and bounce 15 characteristics.
[00118]
To further elaborate the analysis, the tracking module 318 employs sophisticated algorithms to aggregate multiple bounce spots on the pitch. By collecting and combining data from numerous instances where the ball has made contact with the ground, the tracking module 318 generates a pitch map. This 20 pitch map provides a detailed representation of the ball's trajectory, bounce patterns, and variations across the pitch, offering meaningful information for strategic decision-making and understanding pitch conditions.
[00119]
The identification module 310 may be communicably coupled with the classification module 312 for communicating the identified action. The 25 classification module 312 may be configured to perform classification of the identified action (e.g., a bowling delivery) into a type (e.g., bowling type). Further, the classified type comprises at least one from a group comprising pace,
24
spin, inswing, outswing, length, short, bouncer, wide, leg
-break, off-break, googly, doosra, arm-ball, carrom-ball, flipper, and top-spinner.
[00120]
Herein, the classification further comprising using at least a deep learning model comprising at least a recurrent neural network using at least a gated recurrent unit, attention-based long short-term memory, and convolutional 5 neural network.
[00121]
Further, the voice command module 322 may be configured to enable a voice command feature for the bowler using an audio sensor attached to the wearable device 200.
[00122]
Further, the audio feedback module 324 may be configured to 10 provide an audio feedback using a speaker/transducer to convey bowling actions and session level statistics to the bowler. This is further explained in greater detail in conjunction with FIG. 10.
[00123]
The memory 320 may be configured to store or save data and/or information of the player 114 such as performance analysis. The memory 320 15 may be a non-volatile memory or a volatile memory. Examples of non-volatile memory may include, but are not limited to a flash memory, a Read Only Memory (ROM), a Programmable ROM (PROM), Erasable PROM (EPROM), and Electrically EPROM (EEPROM) memory. Examples of volatile memory may include but are not limited to Dynamic Random Access Memory (DRAM), 20 and Static Random-Access Memory (SRAM).
[00124]
The processor 316 may be communicably coupled with the interface 302, the receiver 304, the transmitter 306, the one or more sensors 308, the identification module 310, the classification module 312, the filtering module 314, the tracking module 318, the memory 320, the voice command module 322, 25 and the audio feedback module 324. Further, the interface 302, the receiver 304, the transmitter 306, the one or more sensors 308, the identification module 310, the classification module 312, the filtering module 314, the processor 316, the
25
tracking
module 318, the memory 320, the voice command module 322, and the audio feedback module 324 may also be communicably coupled with each other.
[00125]
Moreover, each of the identification module 310, the classification module 312, the filtering module 314, the tracking module 318, the voice command module 322, and/or the audio feedback module 324 may be 5 implemented as a hardware only module or a software only module or a combination of both hardware as well as software. Each of these modules 310, 312, 314, 318, 322, and 324 may also be communicable coupled with a processor and a memory for performing operations and functions as described herein.
[00126]
Also, the server 108 may also comprise the same modules or 10 components (i.e. a receiver, a transmitter, an identification module, a classification module, a filtering module, a processor, a tracking module the voice command module, the audio feedback module, and a memory) as of the electronic circuitry 104 for performing the same functions/operations as described herein. 15
[00127]
FIG. 4 illustrates a block diagram for building and training an artificial intelligence (AI) model for performance analysis of the player 114, in accordance with an example embodiment. As described above in FIG. 3, the one or more sensors 308 may collect sports data and provide the collected sports data to a feature extraction 402. This feature extraction 402 may perform the same 20 functions and/or operation/s (such filtering the collected sports data received from the sensors 308) as performed by the filtering module 314. The feature extraction 402 may filter the collected sports data to remove any noise and process the collected sports data as needed. The filtered sports data may then be transmitted to a feature extraction 402. 25
[00128]
Signal processing and time-series analysis is used to filter the collected sports data. That is, time-series techniques such as signal processing, time-series analysis, and statistical modelling are used to extract relevant features from the collected sports data to describe the motion pattern and
26
frequency components. Also, time
-series analysis techniques are used to extract features from the collected sports data, such as trend analysis, seasonal decomposition, and autocorrelation from the collected sports data.
[00129]
In addition, signal processing techniques are used to extract features from the collected sports data, such as filtering, Fourier transforms, or wavelet 5 transforms. These techniques are used to extract features such as frequency, amplitude, phase, and energy from the collected sports data.
[00130]
Based on the filtered sports data, a performance analysis 406 is performed to determine performance of the player 114. The performance of the player 114 may include, but is not limited to, a bowling session summary from 10 the collected/filtered sports data from the sensors 308. Such bowling session summary may include, but is not limited to, total balls bowled, maximum arm speed, minimum arm speed, average arm speed, intensity (balls per hour), total time, heart rate during the session, heart rate zones – warmup, far burn, cardio, threshold, and/or peak zones. 15
[00131]
The performance of the player 114 may include bowling data points. The bowling data points may include, but is not limited to, arm speed, arm release angle, wrist rotation/angel at release, run up steps, run up speed (acceleration, steps rhythm/consistency), run up distance, stride length, landing foot and balance. 20
[00132]
The performance of the player 114 may also include bowling parameters from vision. The bowling parameters from vision may include, but is not limited to, action detection, run up stride, run up rhythm/consistency, action type (front-on, midway, sideway-on), bowling arm detection (right arm, left arm) and arm angles (angle of arm from the shoulder at release, angle 25 between upper and lower arm, arm angle at release). Further parameters include ball type (spin: off spin, leg spin, doosra, googly, flipper, spin type: finger spin, wrist spin, fast: medium, fast, outswing, inswing, leg cutter), ball classification (full toss, shot pitch, good length, yorker or outside off, inline, outside leg), ball
27
parameters (ball release parameters
- on seam, wobble seam, cross seam, ball speed – at release, after pitching, swing/spin angle – after pitching, landing foot and balance).
[00133]
There are several different bowling techniques that a bowler can employ to deliver the ball and restrict the batsman's scoring opportunities. Each 5 bowling technique has its own specific approach and objective, and skilled bowlers can choose the right technique for each delivery based on factors such as the batsman's strengths, weaknesses, and the pitch conditions. In cricket, the bowler and the team captain strategically position fielders on the field to create favorable field placements. Field placement involves positioning the fielders in 10 specific areas to maximize the chances of taking wickets or restricting the batsman from scoring runs. The bowler and captain work together to devise a fielding strategy that complements the chosen bowling technique and aims to create pressure on the batsman.
[00134]
Each fielding position on the cricket field has a specific name, such 15 as slips, gully, point, cover, mid-off, mid-on, square leg, fine leg, and third man. The captain and bowler determine the positioning of the fielders based on the pitch conditions, the bowler's strengths and weaknesses, the opposition's batting style, and the score.
[00135]
Here are some of the most common bowling techniques used in 20 cricket:
[00136]
Pace: Refers to a fast-paced delivery with significant speed and minimal spin.
[00137]
Spin: Involves imparting spin to the ball to make it deviate from its original path as it travels towards the batsman. 25
[00138]
Inswing: Refers to a delivery that curves towards the batsman (for a right-handed batsman, it moves from the off-side towards the leg-side).
28
[00139]
Outswing: Involves a delivery that curves away from the batsman (for a right-handed batsman, it moves from the leg-side towards the off-side).
[00140]
Length: Describes the area where the ball bounces on the pitch, usually categorized as full length, good length, or short length.
[00141]
Short: Represents a delivery with a shorter length, typically aimed at 5 unsettling the batsman or inducing a mistimed shot.
[00142]
Bouncer: Refers to a delivery that is bowled short and rises towards the batsman's upper body or head, intended to surprise, or intimidate the batsman.
[00143]
Wide: Represents a delivery that is bowled too wide outside the 10 batsman's reach, resulting in an extra run being awarded to the batting team.
[00144]
Leg-break: Describes a type of delivery for spin bowlers that spins from the leg side to the off- side for a right-handed batsman (opposite for a left-handed batsman).
[00145]
Off-break: Refers to a type of delivery for spin bowlers that spins 15 from the off-side to the leg side for a right-handed batsman (opposite for a left-handed batsman).
[00146]
Googly: Represents a deceptive delivery bowled by a leg-spinner, which spins in the opposite direction to that expected by the batsman.
[00147]
Doosra: Describes another deceptive delivery by an off-spinner, 20 which spins in the opposite direction compared to a traditional off-spin delivery.
[00148]
Arm-ball: Refers to a delivery bowled by an off-spinner or left-arm spinner that goes straight without significant spin.
29
[00149]
Carrom-ball: Represents a delivery bowled by a spinner, which is released using the fingers and flicked with a snap of the wrist, resulting in a skiddy delivery with late spin.
[00150]
Flipper: Describes a delivery bowled by a leg-spinner, which is released with an action similar to a fast bowler, resulting in a quicker and flatter 5 trajectory.
[00151]
Top-spinner: Represents a delivery bowled by a spinner that has a forward-spinning motion, causing the ball to bounce higher than expected.
[00152]
These are just a few of the many different bowling techniques or deliveries used in cricket. Skillful bowlers are able to vary their deliveries with 10 pace, swing, spin, and precise line and length to trouble the batsmen and take wickets. By intelligently selecting the right delivery, adjusting their run-up, and exploiting the conditions, bowlers aim to restrict the batsmen's scoring opportunities and maintain control over the game.
[00153]
The performance of the player 114 may also include feedback. The 15 feedback may include, but is not limited to, training loads, alert zones, progress based on goals, ranking against peer group /leaderboard, auto training plan. The Auto training plan may include, but is not limited to, preparing for t20/one day/test match, bowling position plan (opener/first change/middle overs/ end of innings), bowling against the batting style, auto feedback / alerts, ball length / 20 recommendation depending on the field, comparison with elites.
[00154]
The present invention also encompasses wicket keeping (cricket) parameters. The wicket keeping (cricket) parameters may include, but is not limited to, reaction time, hand speed, total catches, throughs, total steps, heat map using GPS and/or other sensor data. 25
[00155]
The present invention also encompasses fielding (cricket) parameters. The wicket fielding (cricket) parameters may include, but is not
30
limited to, running up speed, total distance covered, acceleration, heat
-map. total catches, throughs, reaction time for catch, throughout.
[00156]
The present invention also encompasses health parameters which may include, but is not limited to, heart rate chart, heart rate zones (warmup, far burn, cardio, threshold, peak), insights and risk heart zone, calories and steps. 5
[00157]
The present invention also encompasses calories parameters of the player 114. This accounts for all those parameters to measure more accurate calories burnt for the wearable device 118.
[00158]
The present invention also encompasses audio sensor/s provided on the wearable device 118 or the electronic circuitry 104 to enable different sports 10 and fitness modes (e.g. – I am bowling, stop bowling, I am running now). This will enable touchless event trigger and allow the player 114 to focus on activity that he/she are doing. Such audio feedback method may involve using speaker or other sound device to convey bowling actions and session level statistics, offer feedback and guidance to the player 114. 15
[00159]
The present invention also encompasses an AI model for ball impact. Any ball impact is allowed to be identified using only optical sensor data. When used with other sensors data, improve impact detection automatically accuracy. The present invention also encompasses auto detection of the player 114 being a right/left hand bowler. Using one or more sensor data, and identified range of 20 activity over the sample data, auto identify players playing style is done and starting giving them performance data without need to manually giving these details.
[00160]
Based on the filtered sports data, an analytics 404 is performed for the player 114. The analytics 404 develops specialized fine-tuned feature 25 extraction for sporting activity using proprietary sports activity algorithms.
31
[00161]
The analytics 404 for the player 114 is provided as an input to an artificial intelligence (AI) model building and training 408. The AI model building and training 408 uses deep learning models. Further, deep learning models such as recurrent neural networks (RNNs), like Gated Recurrent Unit (GRU), Attention-based LSTM (ATT-LSTM), and convolutional neural 5 networks (CNNs) like Inception-v3, MobileNet are used to analyse the filtered sports data 314 and feature extraction 402. Further, RNNs are used for modelling long-term dependencies in the data, while CNNs are valid for detecting patterns in spatiotemporal data. Deep learning methods are also used to pull out spatial and temporal features from the collected sports data and to model sequential 10 data. By identifying and analysing movement patterns of the player 114, activities of the player 114 are accurately identified and classified.
[00162]
A recurrent neural network (RNN) is a type of neural network that is designed to handle sequential data. It processes inputs in a sequence, one at a time, while maintaining a "hidden state" that encodes information from previous 15 inputs. The hidden state is updated at each time step based on the current input and the previously hidden state.
[00163]
A gated recurrent unit (GRU) is a type of RNN that uses gating mechanisms to control the flow of information in the network. It is similar to the more well-known Long Short-Term Memory (LSTM) network in that it is 20 designed to address the vanishing gradient problem that can occur in standard RNNs. The GRU has fewer parameters than the LSTM, making it faster to train and less prone to overfitting.
[00164]
The GRU has two gating mechanisms: an update gate and a reset gate. The update gate determines how much of the previous hidden state should 25 be retained and how much of the new input should be incorporated into the updated hidden state. The reset gate determines how much of the previous hidden state should be forgotten and how much of the new input should be used to create a new candidate hidden state.
32
[00165]
At each time step, the GRU takes in an input vector x(t) and the previous hidden state h(t-1). It then computes the update gate z(t) and the reset gate r(t) based on the input and the previous hidden state. It then uses these gates to update the candidate hidden state h~(t) and the new hidden state h(t) as follows: 5
z(t) = sigmoid(Wz * [h(t-1), x(t)])
r(t) = sigmoid(Wr * [h(t-1), x(t)])
h~(t) = tanh(W * [r(t) * h(t-1), x(t)])
h(t) = (1 - z(t)) * h(t-1) + z(t) * h~(t)
[00166]
Where sigmoid and tanh are activation functions, Wz, Wr, and W are 10 weight matrices that are learned during training, and [h(t-1), x(t)] denotes the concatenation of the previous hidden state and the current input.
[00167]
In summary, the GRU uses gating mechanisms to control the flow of information in the network and to selectively update the hidden state at each time step based on the input and the previous hidden state. This allows the network to 15 capture long-term dependencies in sequential data and to avoid the vanishing gradient problem that can occur in standard RNNs.
[00168]
An attention-based LSTM is a type of recurrent neural network (RNN) that uses an attention mechanism to selectively focus on specific parts of the input sequence. It is similar to a regular LSTM, but it adds an additional 20 component that allows the network to selectively attend to different parts of the input sequence, based on their relevance to the task at hand.
[00169]
The basic idea behind attention is to give the network the ability to focus on the most relevant parts of the input sequence for a particular task. This is done by introducing a set of attention weights, which are used to compute a 25 weighted sum of the input sequence, where the weights indicate the relative
33
importance of each input element. The attention weights are learned during
training and can be thought of as a form of soft attention, in which the network can attend to multiple parts of the input sequence at once, and the attention weights can vary from one time step to the next.
[00170]
In an attention-based LSTM, the attention mechanism is added to the 5 LSTM architecture in the following way:
[00171]
The input sequence is fed into the LSTM layer, which produces a sequence of hidden states. The last hidden state of the LSTM layer is used as a context vector, which summarizes the information in the input sequence. The attention mechanism computes a set of attention weights, which are used to 10 compute a weighted sum of the input sequence.
[00172]
The context vector and the weighted sum of the input sequence are concatenated, and fed into a final output layer, which produces the output for the task.
[00173]
The attention mechanism can be implemented in various ways, but 15 one common approach is to use a feedforward neural network to compute the attention weights. The input to the feedforward network is a concatenation of the current hidden state of the LSTM and the input at the current time step, and the output is a scalar value that represents the relevance of the input at that time step. The attention weights are then computed using a SoftMax function, which 20 ensures that the weights sum to one.
[00174]
In summary, an attention-based LSTM is a type of RNN that uses an attention mechanism to selectively focus on different parts of the input sequence, based on their relevance to the task at hand. This allows the network to capture long-term dependencies in the input sequence, and to make more accurate 25 predictions by attending to the most relevant information.
34
[00175]
A convolutional neural network (CNN) is a type of artificial neural network that is commonly used for analyzing visual imagery. CNNs are designed to recognize patterns and features in images by using layers of filters, known as convolutional layers, to extract increasingly complex features from the input image. 5
[00176]
The basic architecture of a CNN consists of a series of convolutional layers, followed by one or more fully connected layers, and an output layer. The convolutional layers apply a set of learnable filters to the input image, producing a set of feature maps that highlight the presence of certain patterns or features in the image. The fully connected layers combine these features to make a 10 prediction about the class or category of the image.
[00177]
There are several types of CNNs that are commonly used in computer vision applications:
[00178]
LeNet: LeNet was one of the earliest CNNs and was introduced in the 1990s by Yann LeCun. It was designed to recognize handwritten digits and 15 was used for automated check reading.
[00179]
ResNet: ResNet is known for its ability to train very deep CNNs (up to hundreds of layers) without suffering from the vanishing gradient problem. It uses residual connections to allow the network to learn residual functions, which can be thought of as the difference between the input and output of a block of 20 layers.
[00180]
Inception: The Inception network is known for its use of "inception modules", which consist of multiple parallel convolutional layers with different filter sizes. This allows the network to capture features at different scales and resolutions. 25
[00181]
MobileNet is a convolutional neural network (CNN) that was specifically designed for mobile and embedded devices with limited
35
computational resources. It was introduced as a way to make deep learning
models more efficient and practical for deployment on mobile devices.
[00182]
The main innovation of MobileNet is the use of depthwise separable convolutions, which are a form of convolutional operation that factorizes the standard convolution into a depthwise convolution and a pointwise convolution. 5 This reduces the number of parameters and the computational cost of the model, while still allowing it to achieve good accuracy on image classification tasks.
[00183]
These are just a few examples of the many different types of CNNs that have been developed over the years. Each type of CNN has its own strengths and weaknesses, and the choice of which one to use depends on the specific 10 application and the available resources.
[00184]
For training the AI model, the deep learning model are trained using the extracted features or filtered sports data as an input and the desired motion pattern as an output. Taking as an example here, the sports data is monitored continuously from one or more sensors 308. When a particular pre-defined 15 threshold limit is reached with regard to data from the one or more sensors 308, collected sports data and filtered sports data is sampled for pattern recognition for building an AI model by the AI model building and training 408.
[00185]
In an example embodiment, for instance, bowling threshold from accelerometer may be set at 10g and bowling threshold from gyroscope may be 20 set at 1400 degrees per second (dps). One such sports data is received at desired threshold, the sports data is processed for pattern recognition.
[00186]
In the case of bowling by the player 114, the analytics 404 checks the sports data above the threshold limit around the threshold index (e.g., index 50). The pattern identifies the acceleration on each axis and deceleration after a peak 25 value from the accelerometer sports data. Similarly, an increase and a decrease in dps of gyroscope data are also checked. Both the data are matched to confirm that the peaks on both the sensors 308 were detected at the same time using
36
timestamp. The start of acceleration and dip in acceleration defines the start of
the ball delivery and end of the ball delivery. Once the pattern is recognized as a delivery, the data from the start of the delivery index to the end of the delivery index is passed to a speed calculation algorithm. Linear velocity is calculated using accelerometer data, and angular velocity is calculated using gyroscope 5 data.
[00187]
Incase, only one sensor is used, then the pattern is matched only for that sensor, for example acceleration and de acceleration in case of accelerometer data. Linear velocity is calculated using the accelerometer data and converted to predict the angular velocity. In case if only gyroscope is used 10 then analytics 404 checks the increase in rotation speed on gyroscope data around the peak to find the pattern that ball is delivered and calculate angular velocity. Linear velocity is predicted.
[00188]
Furthermore, when an action is detected, the collected or filtered sports data is processed to check if the ball has been released by the bowler 15 (release of the ball). As the hand/arm motion during the bowling action has consistent acceleration and deceleration, sudden dip and spike happens when the ball is released from the hand. Similarly, in the case of gyroscope data, a sudden dip may be observed in the rotational data. The dip and spike threshold limit may be checked to match with the recognized dip values when the ball is released. 20 This dip has to be within the start of the bowling action and the end of the release ball data.
[00189]
The determined performance by the performance analysis 406 as well as an output of the AI model building and training 408 are evaluated by an evaluation 410. Such evaluation involves evaluation of the trained AI model's 25 performance using a set of test data and making adjustments as necessary. Moreover, outcomes of the AI model are evaluated and checked if calibration is needed for pattern recognition.
37
[00190]
Based on the evaluation 410, the AI model is deployed by model deployment 412. For this, use the trained and evaluated model in a real-time system to identify motion patterns in new data from one or more sensors 308. That is, based on the new data from the sensors 308, the AI model is used for performance analysis of the player 114. 5
[00191]
FIG. 5 illustrates a block diagram of deep learning algorithms utilized for building and training the AI model, in accordance with an example embodiment. As shown in Fig, 5, deep learning models such as recurrent neural networks (RNNs) 502 are used. One example of RNN is Gated Recurrent Unit (GRU). Attention-based LSTM (ATT-LSTM) 504 is also utilized as a deep 10 learning algorithm for training the AI model. And convolutional neural networks (CNNs) 506 is also utilized as deep learning algorithm for training the AI model. Examples of CNN may include Inception-v3, MobileNet which are used to analyse the filtered sports data 314 and feature extraction 402. Further, RNNs are used for modelling long-term dependencies in the data, while CNNs are valid 15 for detecting patterns in spatiotemporal data. Deep learning methods are also used to pull out spatial and temporal features from the collected sports data and to model sequential data. By identifying and analysing movement patterns of the player 114, activities of the player 114 are accurately identified and classified.
[00192]
FIG. 6 illustrates a flow diagram of a method 600 for analysing a 20 player’s arm related sports data, in accordance with an example embodiment. The method flow diagram 600 starts at step 602.
[00193]
At step 604, sports data of at least one action of a player 114 is collected by using one or more sensors 308. As explained in FIG. 1 above, the sports data may include any activity or event related to the player 114, such as 25 ball movement, speed, spin, impact, and/or any such activity related to the ball. The one or more sensors 308 may be comprise at least one of accelerometers, gyroscopes, magnetometers, piezoelectric sensors, visual sensors, electromagnetic trackers, and optical sensors and any such that is obvious to a
38
person skilled in the art
. For an example, the accelerometer may sense speed at which the player 114 is delivering a ball.
[00194]
At step 606, the collected sports data is filtered. Signal processing and time-series analysis is used to filter the collected sports data. That is, time-series techniques such as signal processing, time-series analysis, and statistical 5 modelling are used to extract relevant features from the collected sports data to describe the motion pattern and frequency components. Also, time-series analysis techniques are used to extract features from the collected sports data, such as trend analysis, seasonal decomposition, and autocorrelation from the collected sports data. 10
[00195]
In addition, signal processing techniques are used to extract features from the collected sports data, such as filtering, Fourier transforms, or wavelet transforms. These techniques are used to extract features such as frequency, amplitude, phase, and energy from the collected sports data.
[00196]
At step 608, the action is identified from the collected sport data. In 15 some exemplary embodiments, the action is at least one from the group comprising bowling, throwing, and pitching. The action may be identified by the identification module 310. Using advanced algorithms and pattern recognition techniques, the identification module 310 examine various parameters and characteristics present in the sport data to accurately identify the specific action 20 performed by the player.
[00197]
For example, in the case of bowling of the cricket sport, the identification module 310 analyzes the collected sports data related to parameters such as arm speed, arm angle, wrist rotation, run-up steps, run-up speed, and step length. By comparing these data points to predefined patterns 25 and thresholds, the identification module 310 may confidently determine that the action being performed is a bowling action.
39
[00198]
Similarly, for throwing or pitching actions, the identification module 310 utilizes specific parameters and criteria relevant to those actions to make accurate identifications. The identification. This enables the system to provide targeted performance analysis to the identified action, thereby enhancing the player's training, skill development, and overall performance in the cricket sport. 5
[00199]
At step 610, classification of the action into a type is performed. In some exemplary embodiments, the type comprises at least one from a group comprising pace, spin, inswing, outswing, length, short, bouncer, wide, leg-break, off-break, googly, doosra, arm-ball, carrom-ball, flipper, and top-spinner. For performing the classification of the identified action into the type, the 10 classification module 312 may be employed.
[00200]
For example, in case of cricket bowling, the classification module 312 may consider parameters such as the speed of the ball, the movement of the ball in the air or off the pitch, and the length of the delivery. Based on these parameters and predefined thresholds or patterns, the classification module 312 15 may classify the action as either pace or spin. Further classification may be performed to identify specific variations such as inswing, outswing, leg-break, off-break, googly, doosra, etc.
[00201]
Herein, the classification further comprises using at least a deep learning model. In an exemplary embodiment, the deep learning model 20 comprises at least of a recurrent neural network using at least a gated recurrent unit, attention-based long short-term memory, and convolutional neural network.
[00202]
The method flow diagram 600 ends at step 612.
[00203]
For better understanding of the above steps 602 – 614 of method 600, 25 consider an example of a live cricket match where a fast bowler is delivering a ball. The system, equipped with visual sensors and optical sensors, tracks the bowler's arm movements, run-up speed, and other relevant data. Based on this
40
real
-time data, the identification module 310 recognizes that the action being performed is bowling.
[00204]
Then, the classification module 312 examines the collected data to classify the type of bowling. If the ball is consistently delivered at high speeds, with minimal swing or spin, the classification module 312 categorizes it as pace 5 bowling. On the other hand, if the ball exhibits significant swing or spin, the classification module may identify it as swing or spin bowling, respectively.
[00205]
These real-time identification and classification processes enable the system to provide valuable information and analysis to the bowler, helping them to understand their performance, make adjustments, and enhance their bowling 10 skills during the game.
[00206]
FIG. 7 illustrates a flow diagram of a method 700 for determining bowling load and injury stress of a player, in accordance with an example embodiment. The method flow diagram 700 starts at step 702.
[00207]
At step 704, run-up steps of the player may be counted. This refers to 15 the number of steps taken by the player during the run-up phase before delivering the ball. By accurately counting the run-up steps, the system may gather valuable information about the player's approach and positioning before the actual bowling action. The count of run-up steps provides an essential parameter for analysing the player's bowling technique and performance. It may help to assess 20 the consistency and rhythm of the player's run-up, which are crucial factors in achieving optimal performance and minimizing the risk of injury.
[00208]
At step 706, a run-up speed may be tracked using the sensors. By monitoring the player's speed during the run-up, the system may capture valuable data related to the player's approach dynamics. The sensors detect and measure 25 the player's movement, allowing for accurate tracking of the speed at which the player covers the distance during the run-up. This information provides the player's explosiveness, acceleration, and overall speed, which are important
41
factors in determining the effectiveness of the bowling action. Tracking the run
-up speed enables the system to evaluate the player's ability to generate power and momentum leading up to the delivery, contributing to a detailed analysis of their bowling performance.
[00209]
At step 708, total steps of the player may be counted. This refers to a 5 measurement of the overall activity and movement of the player during the bowling session. By counting the total steps, the system may assess the player's level of physical exertion and activity throughout the session. This information helps to determine the player's workload, which is an essential factor in understanding their fitness level and managing the risk of injury. Tracking the 10 total steps enables the system to analyse the player's mobility, endurance, and overall movement patterns, providing valuable perceptions of their physical condition and performance capacity.
[00210]
At step 710, number of balls bowled by the player may be aggregated. This step involves keeping a cumulative count of the balls delivered by the player 15 during the bowling session. By aggregating this data, the system may assess the workload and volume of bowling undertaken by the player. The number of balls bowled provides a key metric for understanding the intensity and duration of the bowling session. It helps in evaluating the player's endurance, stamina, and ability to sustain consistent performance over a prolonged period. Additionally, 20 tracking the number of balls bowled enables the system to monitor the player's progress, set performance levels, and compare their bowling load against predefined targets or standards.
[00211]
At step 712, bowling load and injury stress may be indicated using the total steps, the run-up steps, the run-up speed, gait analysis, landing stress, 25 running symmetry, and the number of balls bowled. By considering these parameters collectively, the system may provide information about the player's workload, exertion, and potential injury risks. The combination of total steps, run-up steps, gait analysis, landing stress, running symmetry, and run-up speed
42
helps
to estimate the physical demands placed on the player's body during the bowling session. The number of balls bowled further contributes to evaluating the player's cumulative workload and the impact on their physical condition. By indicating the bowling load and injury stress, the system provides information to guide training programs, manage player workload, and minimize the risk of 5 injuries associated with excessive or inappropriate bowling activities.
[00212]
The method flow diagram 700 ends at step 714.
[00213]
FIG. 8 illustrates a flow diagram of a method 800 for tracking a ball released by the player’s arm, in accordance with an example embodiment. The method flow diagram 800 starts at step 802. 10
[00214]
At step 804, a posture of the player 114, three dimensional co-ordinates, hand speed, ball grip and follow through distance of the player’s action may be detected. At this step, various aspects of the player's action are detected. Firstly, the posture of the player 114 is identified, which refers to the positioning and alignment of the player's body during the bowling action. This 15 information helps in understanding the biomechanics and technique of the player.
[00215]
Additionally, three-dimensional co-ordinates are captured, which provide precise spatial information about the player's body movements. Hand speed is measured, which indicates the velocity at which the player releases the 20 ball. Ball grip is also detected, which refers to the specific manner in which the player holds and controls the ball. Lastly, the follow-through distance is determined, which measures the distance covered by the player's arm after the ball release. These detected parameters collectively contribute to a detailed analysis of the player's bowling action and provide valuable information for their 25 technique and performance.
[00216]At step 806, a ball released by the player’s arm may be tracked using the visual sensors and optical sensors. In this step, the visual sensors capture
43
visual information, such as video footage or images, of the player's action. These
sensors enable the system to closely monitor the trajectory and movement of the ball from the moment it leaves the player's hand.
[00217]
Optical sensors, on the other hand, employ specialized optical technology to track the ball's position and motion. They may utilize techniques 5 like motion tracking or object recognition to precisely follow the ball's path. By combining the data from both visual and optical sensors, the system may accurately track the ball's movement throughout its trajectory. This tracking information is essential for various purposes, such as analysing the ball's speed, spin, trajectory, and accuracy, which are important factors in evaluating the 10 player's bowling performance. The method flow diagram 800 ends at step 808.
[00218]
By way of an example, consider a scenario of a cricket match where a fast bowler is about to deliver a ball. At this point, the system detects various aspects of the player's action. It analyses the bowler's posture, ensuring they are in the correct position with proper alignment of their body. The system also 15 captures three-dimensional coordinates of the bowler's body, providing precise spatial information about their movements. It measures the hand speed of the bowler as they release the ball, giving an indication of the velocity at which, the ball will be delivered. Additionally, the system detects the bowler's ball grip, which refers to how they hold and control the ball, influencing its movement. It 20 also measures the follow-through distance, which tracks the distance covered by the bowler's arm after releasing the ball. These parameters are continuously monitored and analysed in real-time, providing valuable insights into the bowler's technique and performance.
[00219]
As the bowler releases the ball, visual sensors and optical sensors 25 come into play. The visual sensors, such as high-speed cameras, capture video footage of the entire delivery. These cameras track the ball's trajectory from the bowler's hand to its intended destination, recording its movement in real-time. The optical sensors, on the other hand, utilize specialized technology to precisely
44
track the ball's
position and motion. They may employ techniques like motion tracking algorithms or object recognition to monitor the ball's path accurately. By combining the data from the visual and optical sensors, the system may provide real-time updates on the ball's speed, spin, trajectory, and accuracy. This tracking information is vital for assessing the bowler's performance, identifying 5 any patterns or variations in their deliveries, and making informed decisions for strategic purposes.
[00220]
FIG. 9 illustrates a flow diagram of a method 900 for developing a pitch map based on tracking a ball, in accordance with an example embodiment. The method flow diagram 900 starts at step 902. 10
[00221]
At step 904, spot on a pitch may be marked where the ball has bounced. In this step, as the ball bounces on the pitch during a cricket match, the system marks the spot where the ball has landed. This is done using the tracking information obtained from the previous steps of method 800. The system precisely determines the coordinates on the pitch where the ball contacted the 15 ground.
[00222]
By marking these spots, the system creates a visual representation of the ball's trajectory and bounce locations. This information is essential for analysing the behaviour of the pitch and understanding how it affects the ball's movement. It helps in evaluating factors such as pitch conditions, bounce 20 variations, and any potential patterns that may emerge.
[00223]
At step 906, a plurality of spots on the pitch may be aggregated where the ball had bounced to develop a pitch map. In other words, once multiple spots on the pitch have been marked where the ball has bounced, the system proceeds to aggregate this data to develop a pitch map. The pitch map is a graphical 25 representation that combines all the marked spots to provide a holistic view of the ball's behaviour across different deliveries. By aggregating the bounce spots, the system may identify trends and patterns in how the ball interacts with the pitch surface.
45
[00224]
The pitch map may reveal information related to pitch characteristics such as variations in bounce height, swing, spin, or any other ball movement. This information is highly valuable for bowlers, batsmen, coaches, and analysts as it facilitates in strategizing gameplay, making informed decisions, and understanding the dynamics of the pitch. The method flow diagram 900 ends at 5 step 908.
[00225]
By way of an example, consider a scenario of a cricket match where a fast bowler delivers a ball. As the ball contacts the pitch, the tracking module 318 deployed on the field detects the exact spot where the ball bounces. Let's say the ball bounces near the off-stump, and the tracking module 318 marks this 10 spot on the pitch using its tracking capabilities. This process is repeated for each delivery throughout the match, capturing the bounce spots at various locations on the pitch.
[00226]
Now, with multiple bounce spots marked on the pitch, the tracking module 318 aggregates this data to develop a pitch map. The pitch map is created 15 by combining all the recorded bounce spots. For example, after analyzing several deliveries, the tracking module 318 identifies that the majority of bounce spots are concentrated around the off-stump region and slightly favor the bowler's end. By representing this information graphically on a pitch map, a pattern emerges, indicating that the pitch is offering consistent bounce and swing to the bowlers 20 in that area.
[00227]
As the match progresses, more deliveries are bowled, and the tracking module 318 continues to update the pitch map in real-time. The pitch map evolves, revealing additional information. For instance, the tracking module 318 may detect that certain areas of the pitch produce erratic bounce, making it 25 challenging for batsmen to predict the ball's behavior accurately. This information helps both the bowling and batting teams strategize their gameplay.
[00228]
FIG. 10 illustrates a flow diagram of a method 1000 for enabling voice command and voice feedback features on a wearable device, in accordance
46
with an example embodiment
. The method flow diagram 1000 starts at step 1002.
[00229]
At step 1004, a voice command feature may be enabled for the bowler using an audio sensor attached to the wearable device 200. The wearable device 200, such as a smartwatch or a specialized bowling device, incorporates an audio 5 sensor capable of capturing the bowler's voice commands. For instance, the bowler may use voice commands to initiate specific actions or trigger functions related to their bowling technique.
[00230]
These voice commands may include I am bowling, stop bowling, I am running now, etc. The audio sensor captures the bowler's voice, processes it, 10 and translates it into actionable commands that the wearable device 200 may recognize and respond to. By enabling voice commands, the bowler gains a convenient and hands-free method of interacting with the wearable device 200, allowing for seamless control over various aspects of their bowling performance.
[00231]
At step 1006, an audio feedback may be provided using a 15 speaker/transducer to convey bowling actions and session level statistics to the bowler. After the bowler performs a bowling action or completes a session, the wearable device 200 processes the collected data and generates relevant feedback. This feedback may include bowling actions analysis, session-level statistics, or performance analysis. Instead of relying solely on visual or textual 20 feedback, the device utilizes its speaker or transducer to convey this information audibly to the bowler. For example, the wearable device 200 may provide real-time verbal feedback on the bowler's speed, accuracy, or recommended improvements in their technique.
[00232]
The method flow diagram 1000 ends at step 1008. 25
[00233]
As will be appreciated by those skilled in the art, the techniques described in the various embodiments discussed above are not routine, or conventional, or well understood in the art. The described techniques are
47
designed to capture sensor data from various sports players and provide sports
performance information specifically related to the bowling technique in sports such as cricket, tennis, badminton, paddle, table tennis, baseball, and golf, among others. By using sensors, these techniques aim to minimize false positives (incorrectly identifying an action as a specific bowling type) and false negatives 5 (failing to identify an actual bowling action).
[00234]
To achieve accurate identification of player’s bowling actions, the sensor data is utilized to train AI/ML models. These models learn to recognize patterns in the sensor data that correspond to specific motions or actions, thereby reducing false positives and false negatives. Various deep learning models are 10 employed in this process, including recurrent neural networks (RNNs), Gated Recurrent Unit (GRU), attention-based LSTM (Att-LSTM), and convolutional neural networks (CNNs) such as Inception-v3.
[00235]
RNNs are particularly effective in capturing long-term dependencies in sequential data, making them suitable for modeling the temporal aspects of 15 bowling actions. CNNs, on the other hand, excel at detecting spatial and temporal patterns in the sensor data. By leveraging these advanced models, the system can analyze the custom feature-extracted sensor data using techniques like MobileNet. The training of the AI/ML models includes utilizing the sensor data, player performance data, and feedback data. This approach enables the 20 models to evaluate and rate different bowling actions in a given sport accurately. The real-time sensor data, along with offline feedback data, is combined to enhance the training process and provide timely and accurate performance analysis.
[00236]
Accordingly, blocks of the flow diagram support combinations of 25 means for performing the specified functions and combinations of operations for performing the specified functions for performing the specified functions. It will also be understood that one or more blocks of the flow diagram, and combinations of blocks in the flow diagram, may be implemented by special
48
purpose hardware
-based computer systems which perform the specified functions, or combinations of special purpose hardware and computer instructions.
[00237]
Many modifications and other embodiments of the inventions set forth herein will come to mind to one skilled in the art to which these inventions 5 pertain having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is to be understood that the inventions are not to be limited to the specific embodiments disclosed and that modifications and other embodiments are intended to be included within the scope of the appended claims. Moreover, although the foregoing descriptions 10 and the associated drawings describe example embodiments in the context of certain example combinations of elements and/or functions, it should be appreciated that different combinations of elements and/or functions may be provided by alternative embodiments without departing from the scope of the appended claims. In this regard, for example, different combinations of elements 15 and/or functions than those explicitly described above are also contemplated as may be set forth in some of the appended claims. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.
, Claims:1.
A system for analysing a player’s arm related sports data, the system comprising:
one or more sensors to collect sports data of at least one action of the player;
a filtering module to filter the collected sports data; 5
an identification module to identify the action from the collected sports data, wherein the action is at least one from the group comprising bowling, throwing, and pitching; and
a classification module to perform classification of the action into a type.
10
2.
The system of claim 1, wherein the action comprises bowling of cricket sport.
3.
The system of claim 2, wherein the type comprises at least one from a group comprising pace, spin, inswing, outswing, length, short, bouncer, wide, 15 leg-break, off-break, googly, doosra, arm-ball, carrom-ball, flipper, and top-spinner.
4.
The system of claim 1, further comprising:
counting run-up steps of the player; 20
tracking run-up speed using the sensors;
counting total steps of the player;
aggregating number of balls bowled by the player; and
50
indicating bowling load and injury stress using the total steps, the run-up steps, the run-up speed, and the number of balls bowled.
5.
The system of claim 1, further comprising detecting posture, three dimensional co-ordinates, hand speed, ball grip and follow through distance of 5 the player’s action.
6.
The system of claim 1, wherein one or more sensors comprise at least one of accelerometers, gyroscopes, magnetometers, piezoelectric sensors, visual sensors, electromagnetic trackers, and optical sensors. 10
7.
The system of claim 1, further comprising a tracking module configured for tracking a ball released by the player’s arm using the visual sensors and optical sensors.
15
8.
The system of claim 7, further comprising:
marking spot on a pitch where the ball has bounced; and
aggregating a plurality of spots on the pitch where the ball had bounced to develop a pitch map.
20
9.
The system of claim 1, wherein the classification further comprising using at least a deep learning model comprising at least a recurrent neural network using at least a gated recurrent unit, attention-based long short-term memory, and convolutional neural network.
25
10.
The system of claim 1, further comprising:
51
a voice command module configured for enabling voice command feature for the bowler using an audio sensor attached to the wearable device; and
an audio feedback module configured for providing audio feedback using a speaker/transducer to convey bowling actions and session level statistics to the bowler. 5
11.
A method for analysing a player’s arm related sports data, the method comprising:
collecting the sports data of at least one action of the player by using at least one sensor; 10
filtering the collected data;
identifying the action from the collected sports data, wherein the action is at least one from the group comprising bowling, throwing, and pitching; and
performing classification of the action into a type.
15
12.
The method of claim 11, wherein the action comprises bowling of cricket sport.
13.
The method of claim 12, wherein the type comprises at least one from a group comprising pace, spin, inswing, outswing, length, short, bouncer, wide, 20 leg-break, off-break, googly, doosra, arm-ball, carrom-ball, flipper, and top-spinner.
14.
The method of claim 11, further comprising:
counting run-up steps of the player; 25
52
tracking run-up speed using the sensors;
counting total steps of the player;
aggregating number of balls bowled by the player; and
indicating bowling load and injury stress using the total steps, the run-up steps, the run-up speed, and the number of balls bowled. 5
15.
The method of claim 11, further comprising detecting posture, three dimensional co-ordinates, hand speed, ball grip and follow through distance of the player’s action.
10
16.
The method of claim 11, wherein one or more sensors comprise at least one of accelerometers, gyroscopes, magnetometers, piezoelectric sensors, visual sensors, electromagnetic trackers, and optical sensors.
17.
The method of claim 11, further comprising tracking a ball released by 15 the player’s arm using the visual sensors and optical sensors.
18.
The method of claim 17, further comprising:
marking spot on a pitch where the ball has bounced; and
aggregating a plurality of spots on the pitch where the ball had bounced to 20 develop a pitch map.
19.
The method of claim 11, wherein the classification further comprising using at least a deep learning model comprising at least a recurrent neural network using at least a gated recurrent unit, attention-based long short-term 25 memory, and convolutional neural network.
53
20.
The method of claim 11, further comprising:
a voice command feature for the bowler using an audio sensor attached to the wearable device; and
audio feedback using a speaker/transducer to convey bowling actions and 5 session level statistics to the bowler.