Intelligent Analysis and Automatic Grouping of Activity Sensors

Abstract

The method may include acquiring sensor data from one or more activity sensors, each of the activity sensors being coupled to a respective region of the sports user. The method may include acquiring image data of the sports user and each of the activity sensors coupled to the sports user. The method may include identifying, by a machine learning model, each muscle associated with each region to which the activity sensor is coupled based on the sensor data and the image data. The method may include identifying a movement of the sports user based on the sensor data from the activity sensors and the identified body parts. The method may include analyzing the identified movement of the sports user, including evaluating a body posture of the sports user, identifying one or more movement patterns of the sports user, and / or performing an injury assessment of the sports user.

Description

[Technical field]

[0001] FIELD OF THE DISCLOSURE This disclosure relates generally to intelligent analysis and automatic grouping of activity sensors. [Background technology]

[0002] A player's performance in a sport may depend on how the player performs sport-specific movements and / or moves his / her body. The muscle movements and status of the player may be related to the way the player performs sport-specific movements and / or moves his / her body. Evaluating muscle movements and status may improve the understanding and analysis of sports competitions.

[0003] The subject matter claimed in this disclosure is not limited to embodiments that solve any disadvantages, nor does it operate exclusively in the environments described above. Rather, this background is only provided to illustrate one example technology area where some embodiments described in this disclosure may be practiced. Summary of the Invention [Means for solving the problem]

[0004] According to an aspect of an embodiment, the method may include acquiring sensor data from one or more activity sensors, each coupled to a respective region of the sports user. The method may include acquiring image data of the sports user and each of the activity sensors coupled to the sports user. The method may include identifying, by a machine learning model and based on the sensor data and the image data, respective muscles associated with each region to which the activity sensors are coupled. The method may include identifying movements of the sports user based on the sensor data from the activity sensors and the identified body parts. The method may include analyzing the identified movements of the sports user, including assessing a body posture of the sports user, identifying one or more movement patterns of the sports user, and / or performing an injury assessment of the sports user.

[0005] The object and advantages of the embodiments will be realized and attained at least by the elements, features, and combinations particularly pointed out in the appended claims. It is to be understood that both the foregoing general description and the following detailed description are explanatory only and are not restrictive of the invention as claimed. [Brief description of the drawings]

[0006] Example embodiments are described with additional specificity and detail through the accompanying drawings.

[0007] [Figure 1] FIG. 1 illustrates an exemplary system for sensor location detection according to this disclosure. [Diagram 2] FIG. 2 is a diagram of a sensor according to the present disclosure. [Diagram 3] FIG. 3 is a flow chart of an example method for providing feedback regarding muscle performance of one or more sports users according to the present disclosure. [Figure 4] FIG. 4 is a flow chart of an example method for analyzing an athletic user's movements based on activity sensor data according to the present disclosure. [Diagram 5] FIG. 5 is an example of a computing system. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] One or more activity sensors may be coupled to each area of ​​a sports user to facilitate tracking and analysis of muscle movement and status. The areas of the sports user to which the activity sensors are coupled may correspond to one or more muscles of the sports user such that the activity sensors may assess muscle activity based on metrics such as sweat activity levels and / or electromyogram (EMG) measurements. However, existing activity sensors and methods of collecting sensor data, such as intramuscular EMG and / or surface EMG, may be inefficient and include various drawbacks. In particular, both intramuscular EMG and surface EMG include tedious and time-consuming preparations to ensure accurate collection of sensor data. As an example, intramuscular EMG includes preparations that require precise selection of muscles and needle positioning to facilitate collection of sensor data. As another example, the use of surface EMG may be relatively easier to prepare than the use of intramuscular EMG, but performing surface EMG also requires preparations such as rubbing the surface of the skin where the activity sensor will be attached with alcohol. Furthermore, since pairing multiple activity sensors with sports users must be done manually for each activity sensor, as the number of activity sensors and / or the number of sports users increases, identifying and pairing each activity sensor coupled to one or more sports users becomes increasingly complicated. Thus, identifying and pairing each of the activity sensors with a particular sports user can become increasingly cumbersome without automatic configuration of the activity sensors.

[0009] The embodiments described in this disclosure may facilitate automatic identification and / or pairing of activity sensors to one or more sports users through intelligent analysis and grouping of the activity sensors, improving the availability of sensor data associated with each of the sports users. In some embodiments, one or more activity sensors may be coupled to an area of ​​multiple sports users without performing any preparation. Single camera and / or multi-camera systems may collect image data including position information about the sports users and any movements associated with each sports user to estimate a three-dimensional body posture corresponding to each sports user. The activity sensors and camera systems may continuously transmit sensor data and image data, respectively, to a receiver, and the sensor data and image data may be combined to identify a user-sensor pairing. The sensor data and image data may be analyzed to provide feedback regarding the sports users' muscle movements.

[0010] Embodiments of the present disclosure will now be described with reference to the accompanying drawings.

[0011] FIG. 1 illustrates an exemplary system 100 for sensor placement detection according to the present disclosure. The system 100 may include one or more activity sensors 110a-110f ("activity sensors 110") and one or more cameras 130a and 130b ("cameras 130"). Each of the activity sensors 110a-110f may be coupled to a respective area 120a-120f ("area 120") of the sports user. For example, a first activity sensor 110a may be coupled to a first area 120a of the sports user, and a second activity sensor 110b may be coupled to a second area 120b of the sports user. Image data captured by the cameras and sensor data captured by the activity sensors 110 may be acquired by a machine learning model 140.

[0012] The activity sensor 110 may obtain sensor data detailing muscle activity of the sports user to which the activity sensor 110 is coupled. In some embodiments, the sensor data detailing muscle activity may include EMG measurements indicative of electrical activity generated by the muscles of the sports user. Such electrical activity may be measured based on electrical potentials of muscles corresponding to the regions 120 of the sports user on which the activity sensor 110 is positioned. The EMG measurements may provide information about the corresponding muscles, such as activation levels and / or recruitment order of the corresponding muscles. Additionally or alternatively, the activity sensor 110 may include a pressure sensor configured to measure contraction and / or lengthening of muscles corresponding to the regions 120 of the sports user. Additionally or alternatively, the activity sensor 110 may include a sweat sensor configured to measure the presence of a biomarker in the sweat of the sports user and / or a chemical potential sensor configured to measure lactate accumulation in the muscles of the sports user. Additionally or alternatively, the activity sensor 110 may be configured to measure other types of data, including temperature, electrical potential, and / or other types of biometric data. Additionally or alternatively, the activity sensor 110 may be configured to measure three-dimensional data of the sports user based on orientation, rotation angle and / or direction, speed, and / or other physical parameters related to the sports user's movement.

[0013] The activity sensors 110 may be positioned in corresponding regions 120 of the sports user's skin surface. In this manner, the activity sensors 110 may include external sensors that do not need to be in direct contact with one or more muscles of the sports user to obtain sensor data related to the muscles of the sports user. Because the regions 120 of the sports user correspond to the muscles of the sports user, positioning the activity sensors 110 in the regions 120 of the sports user in proximity to one or more muscles of interest in analyzing the sensor data may facilitate more accurate collection of sensor data. In these and other embodiments, the surface area of ​​each of the activity sensors 110 may be smaller than the surface area of ​​each of the regions 120 corresponding to the muscles of the sports user such that positioning of the activity sensors 110 does not need to be precise. For example, a particular activity sensor may be positioned on the sports user's upper arm to obtain sensor data related to the sports user's biceps. In this example, a particular activity sensor on the sports user's upper arm may not require precise positioning as it may be positioned in one or more regions on the sports user's upper arm to obtain sensor data related to the biceps. Additionally or alternatively, the placement of the activity sensor 110 may be more sensitive when the targeted muscles are small and / or multiple muscles correspond to the same area 120 of the athletic user.

[0014] The cameras 130 may be configured to capture one or more images indicative of the positioning and / or movement of the activity sensors 110 coupled to the area 120 of the sports user. In some embodiments, the cameras 130 may be positioned to capture images including the entire sports user and / or all of the activity sensors 110 coupled to the area 120 of the sports user. Additionally or alternatively, each of the cameras 130 may be positioned to capture images including a portion of the sports user and / or a portion of the activity sensors 110 coupled to the area 120 of the sports user. For example, a first camera may be positioned near the ground to capture images related to the lower body of one or more sports users, including images of the activity sensors 110 positioned in the area 120 located in the lower body of the sports user, and a second camera may be positioned to capture images related to the upper body of the sports user.

[0015] In these and other embodiments, two or more cameras 130 may capture images from different angles relating to the same sports user and / or the same activity sensor 110. Capturing images including multiple angles of the same sports user and / or the same activity sensor 110 may reduce the number of blind spots caused by movement of the sports user and / or the activity sensor 110 (e.g., moving out of the camera's field of view and / or blocking of the line of sight of a first sports user by a second sports user). Additionally or alternatively, capturing images from different angles relating to the same sports user and / or the same activity sensor 110 may facilitate more accurate identification and / or pairing of the activity sensor 110, as described in more detail below.

[0016] The camera 130 may include any device, system, component, or collection of components configured to capture an image. Although two cameras 130 are illustrated in relation to the system 100 with reference to FIG. 1, any number of cameras may be envisioned. The camera 130 may include optical elements, such as lenses, filters, holograms, splitters, etc., and an image sensor in which an image may be recorded. Such an image sensor may include any device that converts an image represented by incident light into an electronic signal. The image sensor may include a number of pixel elements that may be arranged in a pixel array (e.g., a grid of pixel elements). For example, the image sensor may be comprised of a charge-coupled device (CCD) or a complementary metal-oxide semiconductor (CMOS) image sensor. The pixel array may include a two-dimensional array having an aspect ratio of, for example, 1:1, 4:3, 5:4, 3:2, 16:9, 10:7, 6:5, 9:4, 17:6, or other ratios. The image sensor may be optically aligned with various optical elements, such as lenses, that focus light onto the pixel array. For example, any number of pixels may be included, such as 8 million pixels, 15 million pixels, 20 million pixels, 50 million pixels, 1 million pixels, 2 million pixels, 6 million pixels, 10 million pixels, etc.

[0017] The camera 130 may operate at a particular frame rate or may be capable of capturing a particular number of images within a given time. The camera 130 may operate at a frame rate of about 30 frames per second or greater. In a specific example, the camera 130 may operate at a frame rate of about 100 to about 300 frames per second. In some embodiments, a smaller subset of available pixels in the pixel array may be used to allow the camera 130 to operate at a higher frame rate. For example, if a moving object (e.g., an athlete, a region 120 corresponding to the athlete, and / or an activity sensor 110 coupled to the region 120) is known or estimated to be located in a particular quadrant, region, or space of the pixel array, only that quadrant, region, or space may be used in capturing an image to allow a faster refresh rate for capturing another image. This may allow for the use of a less expensive camera while enjoying a higher effective frame rate with less usage than the entire pixel array.

[0018] Camera 130 may also include various other components. Such components may include one or more illumination features, such as a flash or other light source, a light diffuser, or other components for illuminating an object. In some embodiments, the illumination features may be configured to illuminate a moving object, for example, when the moving object is in close proximity to the image sensor, such as when the moving object is within 3 meters of the image sensor.

[0019] Any number of different triggers can be used to cause the camera 130 to capture one or more images of a moving object. As non-limiting examples, the camera 130 may be triggered when a moving object is known or estimated to be in the field of view of the camera 130, when a moving object first starts or changes its motion (e.g., when an athlete starts running, when an athlete jumps, when an athlete's limbs extend into the field of view of the camera, etc.), when a moving object is detected at the top row of pixels in the pixel array, etc. Another example of a trigger can be a sustained peak in the spectrum of the reflected microwaves. For example, if there is a consistent peak over a predetermined time at a given frequency known to be the frequency of an expected moving object, this can serve as a trigger event.

[0020] In some embodiments, the camera 130 may have a field of view within which an image can be captured. The field of view may correspond to a pixel array. In some embodiments, the field of view may be limited such that a moving object spends only a limited amount of time in the field of view. In such embodiments, the camera 130 may be triggered to capture an image while the moving object is in the field of view. The time during which the moving object is in the field of view of the camera 130 may be referred to as an optimal capture time frame. In some embodiments, the optimal capture time frame may include when only the entire moving object is in the field of view, or when only a portion of the moving object is in the field of view. Other factors may also contribute to the optimal capture time frame, such as the distance between the image sensor and the moving object, the amount of illumination that can be provided by the illumination function, etc. For example, the optimal capture time frame may occur when the moving object is moving at a distance of 3 meters to 1 meter from the camera 130, which is the range within which the flash of the camera 130 may provide illumination of the moving object.

[0021] The sensor data recorded by the activity sensor 110 and the image data recorded by the camera 130 may be acquired by a computing system, such as the computing system 500 of FIG. 5. The sensor data recorded by the activity sensor 110 and the image data recorded by the camera 130 may include a timestamp indicating when the respective data was recorded. In some embodiments, the activity sensor 110, the camera 130, and / or the computing system to which the sensor and image data are transmitted may include an internal clock indicating the time the data was recorded and / or acquired. Additionally or alternatively, the data may be provided with a relative timestamp indicating when the data was captured relative to the time the first image data and / or first sensor data was acquired (e.g., the first image data and / or first sensor data is provided with a timestamp of "t0").

[0022] The sensor data and image data can be used to train the machine learning model 140 to recognize which muscles a given activity sensor is associated with. In some embodiments, the machine learning model 140 may be trained to identify muscles and / or assess muscle activity of the sports user based on the acquired image data and sensor data. In some embodiments, the machine learning model 140 may be trained to identify muscles associated with the region 120 to which the activity sensor 110 is associated. Because the initial placement of the activity sensor 110 on the sports user may be arbitrary and such activity sensor 110 may not be calibrated, the machine learning model 140 may be trained and configured to pair the acquired sensor data with the corresponding activity sensor identified in the image data. Additionally or alternatively, the machine learning model 140 may be trained to recognize patterns of muscle activity of the sports user based on the image data acquired by the camera 130 and the sensor data acquired by the activity sensor 110. For example, the machine learning model 140 may be trained to recognize patterns such as increased muscle activity in response to identifying flexed muscles in the image data, increased lactate levels in response to identifying a sports user resting after running a marathon, etc.

[0023] In some embodiments, the machine learning model 140 may employ a machine learning algorithm, and the training of the machine learning model 140 may be supervised, unsupervised, or some combination thereof. For example, the machine learning model 140 may be trained using a decision tree, a naive Bayes classifier, a K-nearest neighbor, a support vector machine, a linear regression, a logistic regression, dimensionality reduction, and / or an artificial neural network. The machine learning model 140 may be employed in any of the methods disclosed herein to perform operations that increase in effectiveness and accuracy over time as the machine learning model learns and is periodically retrained to make more accurate predictions or decisions. For example, any of the operations in the methods disclosed herein, or any other operations, may be performed by the machine learning model 140 to perform these operations with increasing effectiveness and accuracy over time as the machine learning model 140 learns.

[0024] Training of the machine learning model 140 to identify and / or pair each of the activity sensors to each region of the sports user can be facilitated by using training sensor data and training image data. In some embodiments, training of the machine learning model 140 can be based on correlation of the training sensor data and the training image data. In some embodiments, the activity sensors configured to acquire the training sensor data may already be identified and / or paired to each region of the training sports user such that the training sensor data and the training image data include accurate sensor region pairings.

[0025] The trained machine learning model 140 may be configured to output information about the environment and / or the sports user based on the newly captured sensor data and image data. In some embodiments, the trained machine learning model 140 may identify the number of activity sensors included in a given environment. For example, the trained machine learning model 140 may count the number of activity sensors based on newly captured image data of the given environment and / or determine the number of activity data sources based on the newly captured sensor data. In some embodiments, the environment may include multiple sports users, and the trained machine learning model 140 may be configured to identify which sports user each of the activity sensors is coupled to by matching the newly captured image data with the timing of data collection and / or the newly captured sensor data. In this way, the trained machine learning model 140 may additionally or alternatively identify the muscles from which the activity sensors record sensor data. Additionally or alternatively, the machine learning model 140 may be configured to generate a three-dimensional body position model of the sports user and / or a three-dimensional motion simulation model of the sports user's movement.

[0026] In one example, the given environment may include a tennis court with a first tennis player positioned on a first side of the tennis court and a second tennis player positioned on a second side of the tennis court opposite the first side. Each of the tennis players may have a first activity sensor coupled to the tennis player's upper arm and a second activity sensor coupled to the tennis player's forearm such that the given environment includes four activity sensors. Image data of the given environment, the tennis players, and / or the activity sensors may be captured by one or more cameras positioned around the given environment, such as from an elevated position behind either tennis player, from an umpire's chair, from both ends of the tennis net, etc. Based on the image data, the machine learning model 140 may determine that the number of sport users included in the given environment is two, and two activity sensors are coupled to each of the sport users. When the first tennis player swings a tennis racket, the activity sensor coupled to the first tennis player may record sensor data indicative of muscle activity at a given time point. At the same given time point, the activity sensor coupled to the second tennis player may record sensor data indicative of muscle tension rather than muscle activity.

[0027] Based on differences in sensor data between the activity sensor coupled to the first tennis player and the activity sensor coupled to the second tennis player, the machine learning model 140 may pair the recorded sensor data to each of the activity sensors. After pairing the activity sensors, the machine learning model 140 may be configured to analyze the tennis player's movements based on the sensor data. For example, the machine learning model 140 may determine that the first tennis player is performing a smash as opposed to a slice based on the differences in sensor data obtained from the first activity sensor and the second activity sensor coupled to the first tennis player.

[0028] FIG. 2 illustrates an activity sensor 200 according to the present disclosure. In some embodiments, the activity sensor 200 may include an EMG sensor 210, an acceleration sensor 240, and / or any other sensor configured to obtain information regarding muscle activity. In some embodiments, the activity sensor 200 may include one or more output leads 260 to facilitate attachment of the activity sensor 200 to an athletic user. In some embodiments, the activity sensor 200 may include a processor 220, a wireless module 230, and / or any other components to facilitate on-board processing of recorded sensor data. The data processing elements of the activity sensor 200, including the processor 220 and the wireless module 230, may be similar to or the same as the computing system 500 described in more detail below in connection with FIG. 5. In some embodiments, the activity sensor 200 may include a sensor (e.g., the EMG sensor 210 and / or the acceleration sensor 240) and / or the processor 220, and / or a power system 250 configured to provide energy to the wireless module 230.

[0029] In some embodiments, the activity sensor 200 may include one or more sensors configured to collect information regarding a physical state and / or placement of the activity sensor 200 when coupled to an athletic user. For example, the activity sensor 200 may include an acceleration sensor 240 configured to determine the motion of the athletic user, a gyroscope configured to determine the orientation of the activity sensor 200 and / or the athletic user, and / or any other such sensors. In some embodiments, physical information obtained by such sensors may be obtained by machine learning models as another source of information to facilitate muscle identification, activity sensor pairing, etc.

[0030] The power system 250 may be configured to provide energy to the activity sensor 200 during operation. In some embodiments, the power system 250 may include a rechargeable battery, such as a lithium ion battery. Additionally or alternatively, the power system 250 may include a single-use battery, such as a button cell battery, and the activity sensor 200 may be disposed of after use. In embodiments including a rechargeable power system 250, the activity sensor 200 may be charged during use. For example, one or more activity sensors 200 may be collected before and / or after use and placed in a wireless charging area (e.g., a docking container). Additionally or alternatively, two or more activity sensors 200 may be coupled together, such as by a wired connection, to form a system of activity sensors. The power system 250 associated with the system of activity sensors may be charged using a wireless or wired charging port. For example, multiple activity sensors 200 may be attached to a garment, such as a sports jersey, and each of the activity sensors 200 attached to the garment may be connectively coupled. The garment may include a charging port configured to provide energy to the activity sensor 200 when the charging port is connected to an energy source.

[0031] Modifications, additions, or omissions may be made to activity sensor 200 as depicted in FIG. 2 without departing from the scope of the present disclosure, for example, activity sensor 200 may include more or fewer elements than those shown and described in this disclosure.

[0032] 3 is a flow chart of an example method 300 of providing feedback regarding muscle activity of one or more sports users according to the present disclosure. Method 300 may be performed by any suitable system, apparatus, or device. For example, activity sensor 200 may perform one or more operations associated with method 300. Although illustrated in discrete blocks, the activities and operations associated with one or more blocks of method 300 may be divided into additional blocks, combined into fewer blocks, or eliminated, depending on the particular implementation.

[0033] The method may begin at block 310, where sensor data from one or more first activity sensors coupled to a first sports user may be acquired. The first activity sensor coupled to the first sports user may acquire sensor data including the same or similar type of information as activity sensors coupled to other sports users. For example, all activity sensors may acquire EMG data, sweat data, and / or other biometric data regarding the sports user to which the activity sensor is coupled. Thus, in an environment including multiple sports users, it may be infeasible to identify the first activity sensor coupled to the first sports user based on the sensor data alone. In some embodiments, the sensor data may include a specific signature such that all of the sensor data originating from a particular activity sensor includes the same specific signature. Additionally or alternatively, the activity sensor may acquire orientation and / or motion information corresponding to the orientation and / or motion of the sports user based on sensor data obtained from an acceleration sensor, gyroscope, etc. included in the activity sensor, as described above in connection with FIG. 2.

[0034] In block 320, image data of a first sports user and a first sensor coupled to the first sports user may be acquired. The image data may be captured by one or more cameras positioned around a sports environment including one or more sports users and activity sensors coupled to each of the sports users. In some embodiments, the image data may facilitate determining the number of sports users included in a given sports environment and / or the number of activity sensors coupled to each of the sports users. In these and other embodiments, the image data may facilitate identifying a first sports user of a plurality of sports users included in a given sports environment.

[0035] One or more criteria for identifying a particular sports user, such as the first sports user, may be determined based on characteristics of a given sports environment. For example, the particular sports environment may include a soccer field, and the first sports user may be identified based on the color of a jersey worn by the first sports user and / or the number included on the back of the jersey. As another example, the particular sports environment may include a tennis court, and the first sports user may be identified based on a section of the tennis court where the first sports user stays during a tennis match.

[0036] At block 330, each region of the first sports user to which each of the first activity sensors is coupled may be identified. In some embodiments, each of the first activity sensors identified as coupled to the first sports user is analyzed sequentially to identify sensor data to which the activity sensor corresponds. In some embodiments, pairing a given activity sensor to its corresponding sensor data may include analyzing motion and / or sensor data associated with a given activity sensor over time.

[0037] For example, the image data may indicate that a given activity sensor should move in a particular direction, turn, detect a particular sensor data value, etc. in a particular time frame. One or more of the sensor data measurements obtained by the activity sensor in the particular time frame may not correspond to an assumption made based on the image data, and such sensor data measurements may be ignored. Additional assumptions may be made based on the image data of the given activity sensor in a second particular time frame, and one or more additional sensor data readings may be ignored. This process may be repeated until only one of the sensor data readings corresponds to the image data of the given activity sensor.

[0038] At block 340, a movement of the first sport user may be analyzed based on the sensor data from the first activity sensor and the first image data. In some embodiments, the movement of the first sport user may be analyzed to assess body posture, identify one or more movement patterns, and / or perform an injury assessment of the first sport user, as described above in connection with FIG.

[0039] Sensor data from one or more second activity sensors coupled to a second sport user may be acquired at block 350. The sensor data associated with the second activity sensor may be acquired by the same or a similar process as acquiring the sensor data associated with the first activity sensor, as described above in connection with block 310.

[0040] In block 360, image data of a second sports user and a second activity sensor coupled to the second sports user may be acquired. The image data associated with the second sports user and / or the second activity sensor may be acquired by the same or a similar process as acquiring image data associated with the first sports user and / or the first activity sensor, as described above in connection with block 320.

[0041] At block 370, each region of the second sports user to which each of the second activity sensors is coupled may be identified. In some embodiments, each of the second activity sensors identified as coupled to the second sports user may be analyzed sequentially to identify the sensor data to which the activity sensor corresponds. In these and other embodiments, sensor data identified as associated with the first activity sensor may be excluded from the second activity sensor and sensor data pairing analysis.

[0042] At block 380, the second sport user's movements may be analyzed based on the sensor data from the second activity sensor and the second image data. In some embodiments, the second sport user's movements may be analyzed to assess body posture, identify one or more movement patterns, and / or perform an injury assessment of the second sport user, as described above in connection with FIG.

[0043] Modifications, additions, or omissions may be made to method 300 without departing from the scope of the present disclosure. For example, the designations of different elements in the described embodiments are intended to help explain the concepts described herein and are not intended to be limiting. Additionally, method 300 may include any number of other elements or be implemented in systems or contexts other than those described.

[0044] 4 is a flow chart of an example method for analyzing an athletic user's movements based on activity sensor data according to the present disclosure. Method 400 may be performed by any suitable system, apparatus, or device. For example, activity sensor 200 may perform one or more of the operations associated with method 400. Although illustrated in discrete blocks, the operations and operations associated with one or more blocks of method 400 may be divided into additional blocks, combined into fewer blocks, or eliminated, depending on the particular implementation.

[0045] Method 400 may begin at block 410 with training a machine learning model based on image data and sensor data associated with a skilled athlete. Training the machine learning model to recognize and critique the body posture of a sports user may include providing image data and sensor data associated with one or more movements of a skilled athlete (e.g., a professional athlete). In some embodiments, the skilled athlete may repeat a movement to provide additional image data and sensor data such that the machine learning model can determine patterns of learning data associated with each movement and reduce the effects of statistical noise on the data.

[0046] At block 420, a movement of the athlete may be determined based on the image data and the sensor data associated with the athlete. In some embodiments, pairing the image data and the sensor data with the athlete and / or determining the movement of the athlete may be performed in accordance with that described above in connection with method 300.

[0047] At block 430, one or more body postures of the sports athlete may be analyzed. In some embodiments, the body posture analysis may be based on a comparison of the sports athlete's movements to those of a skilled sports athlete. In some embodiments, the trained machine learning model may evaluate the sports athlete's body posture based on image data and sensor data associated with the sports athlete. For example, the trained machine learning model may compare image data and / or sensor data associated with the sports athlete to training data associated with a skilled sports athlete and determine a degree of similarity between the sports athlete's body posture and the skilled sports athlete's body posture. Additionally or alternatively, the trained machine learning model may highlight certain differences between the sports athlete's and the skilled sports athlete's body posture. For example, the sports athlete may be more hunched, have less hunched shoulders, a wider stance, tense arms, poorer weight distribution, etc. than the skilled sports athlete.

[0048] At block 432, one or more patterns of motion of the sports athlete may be identified. In some embodiments, image data and sensor data related to the sports athlete may be collected over a period of time as training data, and patterns in the motion of the sports athlete may be identified based on the training data and / or newly acquired image data and sensor data. For example, the machine learning model may identify a habitual motion of the sports athlete, such as a thrust of the arm immediately before taking a basketball shot. As another example, the machine learning model may identify a difference in the motion of the sports athlete based on the image data and / or sensor data that differs from the habitual motion of the sports athlete, and such a difference in motion may be paired as an "incorrect" motion.

[0049] At block 434, an injury assessment may be performed on the athlete. In some embodiments, image data and sensor data captured immediately prior to and / or after the occurrence of an injury may be analyzed by machine learning models and / or provided to a review user for inspection. In some circumstances, the image data may provide information regarding possible contact points between one or more athletes, apparent imbalances in the posture of an injured athlete, and / or the like. In these and other circumstances, the sensor data may provide a more detailed perspective regarding the presence and / or severity of muscle damage caused by the injury, the timing of the injury (e.g., muscle tear before the athlete falls), and / or the cause of the injury.

[0050] Modifications, additions, or omissions may be made to method 300 without departing from the scope of the present disclosure. For example, the designations of different elements in the described embodiments are intended to help explain the concepts described herein and are not intended to be limiting. Moreover, method 300 may include any number of other elements and may be implemented in systems or contexts other than those described.

[0051] 5 illustrates an example computing system 500 according to at least one embodiment described in this disclosure. The computing system 500 may include a processor 510, a memory 520, a data storage device 530, and / or a communication unit 540, all of which may be communicatively coupled. Any or all of the activity sensor 200 of FIG. 2 may be implemented as a computing system consistent with the computing system 500, including the EMG sensor 210, the processor 220, the wireless module 230, the acceleration sensor 240, and / or the power system 250.

[0052] In general, the processor 510 may include any suitable special purpose or general purpose computer, computing entity, or processing device, including various computer hardware or software modules, and may be configured to execute instructions stored on any applicable computer-readable recording medium. For example, the processor 510 may include a microprocessor, a microcontroller, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or any other digital or analog circuitry configured to interpret and / or execute program instructions and / or process data.

[0053] 5 as a single processor, it is understood that processor 510 may include any number of processors distributed across any number of networks or physical locations configured to individually or collectively perform any number of operations described in this disclosure. In some embodiments, processor 510 may interpret and / or execute program instructions and / or process data stored in memory 520, data storage device 530, or memory 520 and data storage device 530. In some embodiments, processor 510 may fetch program instructions from data storage device 530 and load program instructions into memory 520.

[0054] After the program instructions are loaded into memory 520, the processor 510 may execute the program instructions, such as instructions to perform methods 300 and / or 400 of Figures 3 and / or 4, respectively. For example, the processor 510 may acquire sensor data associated with the sports user, acquire image data associated with the sports user, identify a body part associated with each activity sensor, and / or identify a movement of the sports user based on the sensor data and the associated body part.

[0055] Memory 520 and data storage 530 may include a computer-readable storage medium or one or more computer-readable storage media for carrying or having computationally executable instructions or data structures stored thereon. Such computer-readable storage media may be any available media that can be accessed by a general-purpose or special-purpose computer, such as processor 510. For example, memory 520 and / or data storage 530 may store acquired sensor data. In some embodiments, computing system 500 may or may not include either memory 520 or data storage 530.

[0056] By way of non-limiting example, such computer-readable recording media may include random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), compact disk read-only memory (CD-ROM) or other optical disk storage, magnetic disk storage or other magnetic storage, flash memory devices (e.g., solid-state memory devices), or any other recording media that can be used to transmit or store desired program code in the form of computer-executable instructions or data structures and that can be accessed by a general-purpose or special-purpose computer. Combinations of the above may also be included within the scope of computer-readable recording media. Computer-executable instructions may include, for example, instructions and data configured to cause processor 510 to perform a certain operation or group of operations.

[0057] The communication unit 540 may include any component, device, system, or combination thereof configured to transmit and receive information over a network. In some embodiments, the communication unit 540 may communicate with other devices at other locations, at the same location, or other components within the same system. For example, the communication unit 540 may include a modem, a network card (wireless or wired), an optical communication device, an infrared communication device, a wireless communication device (such as an antenna), and / or a chipset (such as a Bluetooth device, an 802.6 device (e.g., a metropolitan area network (MAN)), a WiFi device, a WiMax device, a cellular communication facility, etc.), and / or the like. The communication unit 540 may enable data to be exchanged with a network and / or any other device or system described in this disclosure. For example, the communication unit 540 may enable the system 500 to communicate with other systems, such as computing devices and / or other networks.

[0058] Those skilled in the art, after reviewing this disclosure, may recognize that modifications, additions, or omissions may be made to system 500 without departing from the scope of the disclosure. For example, system 500 may include more or fewer components than explicitly illustrated and described.

[0059] Additionally, the embodiments described in this disclosure may involve the use of special purpose or general purpose computers containing various computer hardware or software modules. Further, the embodiments described in this disclosure may be implemented using computer-readable media for carrying or having computer-executable instructions or data structures stored thereon.

[0060] Terms used in this disclosure, and particularly in the appended claims (e.g., the body of the appended claims), are generally intended to be "open-ended" (e.g., the term "including" should be interpreted as "including, but not limited to").

[0061] Furthermore, where a particular number of introduced claim recitations is intended, such intent will be expressly stated in the claim, and in the absence of such a statement, no such intent exists. For example, as an aid to understanding, the following appended claims may include the use of the introductory phrases "at least one" and "one or more" to introduce a claim recitation. However, even when the same claim includes the introductory phrases "one or more" or "at least one" and an indefinite article such as "a" or "an" (e.g., "a" and / or "an" means "at least one" or "one or more"), the use of such phrases should not be construed as meaning that the introduction of a claim recitation with the indefinite article "a" or "an" limits a particular claim that includes such an introduced claim recitation to embodiments that include only one such recitation, nor should the use of a definite article be construed as meaning that the introduction of a claim recitation with the indefinite article "a" or "an" limits a particular claim that includes such an introduced claim recitation to embodiments that include only one such recitation.

[0062] Moreover, even when a particular number is explicitly recited in an introduced claim recitation, one of ordinary skill in the art will recognize that such recitation should be interpreted to mean at least the recited number (e.g., the mere recitation of "two statements" without other modifiers means at least two statements, or more than two statements). Furthermore, when idiomatic phrases similar to "at least one of A, B, and C, etc." or "one or more of A, B, and C, etc." are used, such constructions are generally intended to include A alone, B alone, C alone, A and B together, A and C together, B and C together, or A, B, and C together, etc.

[0063] Furthermore, whether in the specification, claims, or drawings, any conjunction or phrase preceding two or more alternative terms should be understood to contemplate the possibility of including either one of the terms, either one of the terms, or both of the terms. For example, the phrase "A or B" should be understood to include the possibilities of "A" or "B," or "A and B."

[0064] All examples and conditional statements incorporated in this disclosure are intended for educational purposes to help the reader understand the concepts contributed by the inventor to further develop the disclosure and the technology, and are not to be construed as being limited to such specifically incorporated examples and conditions. Although the embodiments of the present disclosure have been described in detail, various changes, substitutions, and alterations can be made without departing from the spirit and scope of the present disclosure.

Claims

1. Identifying one or more first activity sensors coupled to a first sports user and one or more second activity sensors coupled to a second sports user from a defined environment including a plurality of sports users, the plurality of sports users having one or more first or second activity sensors coupled to each area of ​​each sports user, the one or more first or second activity sensors being configured to generate sensor data indicative of muscle activity; acquiring first sensor data from the one or more first activity sensors coupled to each area of ​​a first sport user; acquiring first image data of the first sports user and the one or more first activity sensors coupled to the first sports user; Identifying the first sports user as a first specific sports user based on the first sensor data and the first image data by a machine learning model; identifying, by the machine learning model, muscle activities of the first particular sports user associated with each region to which the one or more first activity sensors are respectively coupled based on the first sensor data and the first image data; and identifying, by the machine learning model, a specific movement performed by the first specific sport user based on the respective muscle activities of the identified first specific sport user.

2. 2. The method of claim 1, wherein the first sensor data from the one or more first activity sensors comprises at least one of electromyographic data, biometric data, movement data, orientation data, rotational data, electrical data, sweat data, temperature data, or pressure data.

3. determining a first body posture of the first specific sport user associated with the specific movement performed by the first specific sport user; obtaining training sensor data and training image data based on the movements of a skilled athlete using the machine learning model; identifying specific movements performed by the skilled sportsman based on the training sensor data and the training image data; determining a second body posture associated with the particular movement performed by the skilled athlete; comparing the first body posture to the second body posture; and identifying one or more differences in the execution of the particular movement between the first body posture of the first particular sport user and the second body posture of the skilled sports athlete.

4. The method further includes analyzing the specific movement of the first specific sport user by determining one or more movement patterns of the first specific sport user, wherein determining the movement patterns of the first specific sport user includes: obtaining training sensor data and training image data based on a movement of the first sports user in performing the specific movement by the machine learning model; Identifying one or more habitual movements of the first specific sport user in performing the specific movements; - comparing the movement of the first specific sport user in performing the specific movement with a habitual movement of the first specific sport user in performing the specific movement; The method of claim 1 , further comprising identifying one or more differences between the movement of the first specific sport user in performing the specific movement and the habitual movement of the first specific sport user in performing the specific movement.

5. The method further includes performing an injury assessment of the first specific sport user, the injury assessment comprising: determining a time when the injury occurred involving the first particular sport user; analyzing the sensor data and the image data relating to the movement of the first specific sports user at a time when an injury involving the first specific sports user occurred; and determining at least one of the presence of muscle damage caused by the injury, the severity of muscle damage caused by the injury, or the cause of the injury.

6. 2. The method of claim 1, wherein identifying the respective muscle activities associated with the respective regions to which each of the one or more first activity sensors is coupled comprises generating a three-dimensional body posture model or a three-dimensional movement simulation model of the first particular sport user.

7. and identifying a second specific sports user's movement, the identification comprising: acquiring second sensor data from the one or more second activity sensors, each of the one or more second activity sensors coupled to a respective region of the second sport user; acquiring second image data of the second sport user and the one or more second activity sensors coupled to the second sport user; Identifying the second sport user as the second specific sport user based on the second sensor data and the second image data by the machine learning model; determining, by the machine learning model, muscle activities of the second specific sports user associated with each region of the second specific sports user to which the one or more second activity sensors are coupled based on the second sensor data and the second image data; The method of claim 1 , further comprising: identifying a specific movement performed by the second specific sport user based on the respective muscle activities of the second specific sport user identified by the machine learning model.

8. 1. A system for grouping and analyzing sensor activity, comprising: one or more activity sensors coupled to respective regions of each of a plurality of sports users within a defined environment, the activity sensors configured to generate sensor data indicative of muscle activity; one or more cameras configured to collect image data of the user within the defined environment; a computing system communicatively coupled to the one or more activity sensors and the one or more cameras, the computing system comprising: identifying one or more first activity sensors coupled to a first sports user and one or more second activity sensors coupled to a second sports user from the defined environment including the plurality of sports users; acquiring first sensor data from the one or more first activity sensors each positioned on the first sports user and coupled to a respective area of ​​the first sports user; acquiring first image data of the first sports user and the one or more first activity sensors coupled to the first sports user; Identifying the first sports user as a first specific sports user by a machine learning model and based on the first sensor data and the first image data; identifying, by the machine learning model, muscle activities of the first particular sports user associated with each region to which the one or more first activity sensors are respectively coupled based on the first sensor data and the first image data; A system for grouping and analyzing sensor activities, configured to identify a specific movement performed by the first specific sports user based on the identified muscle activity of the first specific sports user by the machine learning model.

9. the first sensor data from the one or more first activity sensors; 9. The system of claim 8, comprising at least one of electromyographic data, biometric data, movement data, orientation data, rotational data, electrical data, sweat data, temperature data, or pressure data.

10. The computing system includes: determining a first body posture of the first specific sport user associated with the specific exercise performed by the first specific sport user; obtaining training sensor data and training image data based on the movements of a skilled athlete using the machine learning model; identifying specific movements performed by the skilled sportsman based on the training sensor data and the training image data; determining a second body posture associated with the particular movement performed by the skilled athlete; comparing the first body posture with the second body posture; 10. The system of claim 8, further configured to: identify one or more differences in the execution of the specific movement between the first body posture of the first specific sport user and the second body posture of the skilled sports athlete.

11. The computing system is further configured to analyze the specific movement of the first specific sports user by determining one or more movement patterns of the first specific sports user, and determining the movement patterns of the first specific sports user includes: acquiring training sensor data and training image data based on the movement of the first specific athlete user when performing the specific movement by the machine learning model; Identifying one or more habitual movements of the first specific sport user when performing the specific movement; comparing the movement of the first specific sport user when performing the specific movement with a habitual movement of the first specific sport user when performing the specific movement; and identifying one or more differences between the movement of the first specific sport user when performing the specific movement and the habitual movement of the first specific sport user when performing the specific movement.

12. The computing system is further configured to perform an injury assessment of the first specific sport user, the injury assessment comprising: determining a time when an injury occurred involving the first particular sports user; analyzing the sensor data and the image data relating to the movement of the first specific sports user at a time when an injury involving the first specific sports user occurred; and determining at least one of the presence of muscle damage caused by the injury, the severity of muscle damage caused by the injury, or the cause of the injury.

13. 9. The system of claim 8, wherein identifying the respective muscle activities associated with the respective regions to which each of the one or more first activity sensors is coupled comprises generating a three-dimensional body posture model or a three-dimensional movement simulation model of the first particular sport user.

14. The computing system is further configured to identify a second specific sports user, the identification comprising: acquiring second sensor data from the one or more second activity sensors, each coupled to a respective region of the one or more second sport users; acquiring second image data of the second sports user and the one or more second activity sensors coupled to the second sports user; Identifying the second sport user as the second specific sport user based on the second sensor data and the second image data by the machine learning model; determining, using the machine learning model, muscle activities of the second specific sports user associated with each region of the second specific sports user to which the one or more second activity sensors are coupled based on the second sensor data and the second image data; The system of claim 8 , wherein the machine learning model identifies specific movements performed by the second specific sports user based on the respective muscle activities of the identified second specific sports user.

15. One or more non-transitory computer-readable storage media configured to store instructions that, in response to being executed, cause the system to perform operations, the operations including: identifying one or more first activity sensors coupled to a first sports user and one or more second activity sensors coupled to a second sports user from a defined environment including a plurality of sports users, the plurality of sports users having one or more first or second activity sensors coupled to each region of each sports user, the one or more first or second activity sensors being configured to generate sensor data indicative of muscle activity; acquiring first sensor data from the one or more first activity sensors coupled to respective areas of the first sports user positioned on the first sports user; acquiring first image data of the first sports user and the one or more first activity sensors coupled to the first sports user; Identifying the first sports user as a first specific sports user based on the first sensor data and the first image data by a machine learning model; identifying, by the machine learning model, muscle activities of the first particular sports user associated with each region to which the one or more first activity sensors are respectively coupled based on the first sensor data and the first image data; and identifying, by the machine learning model, a specific movement performed by the first specific sports user based on the identified muscle activity of the first specific sports user.

16. The operation includes: determining a first body posture of the first specific sport user associated with the specific movement performed by the first specific sport user; obtaining training sensor data and training image data based on the movements of a skilled athlete using the machine learning model; identifying specific movements performed by the skilled athlete based on the training sensor data and the training image data; determining a second body posture associated with the particular movement performed by the skilled athlete; and comparing the first body posture to the second body posture; and identifying one or more differences between the first body posture of the first particular sport user and the second body posture of the skilled athlete when performing the particular movement.

17. The operation includes: The method further includes analyzing the specific movement of the first specific sport user by determining one or more movement patterns of the first specific sport user, wherein determining the movement patterns of the first specific sport user includes: acquiring training sensor data and training image data based on the movement of the first specific athlete user when performing the specific movement by the machine learning model; Identifying one or more habitual movements of the first specific sport user when performing the specific movement; comparing the movement of the first specific sport user when performing the specific movement with a habitual movement of the first specific sport user when performing the specific movement; and identifying one or more differences between the movement of the first specific sport user when performing the specific movement and the habitual movement of the first specific sport user when performing the specific movement.

18. The operation includes: The method further includes performing an injury assessment of the first specific sport user, the injury assessment comprising: determining a time when an injury occurred involving the first particular sports user; analyzing the sensor data and the image data related to the movement of the first particular sports user at a time when an injury involving the first particular sports user occurred; and determining at least one of the presence of muscle damage caused by the injury, the severity of muscle damage caused by the injury, or the cause of the injury.

19. 16. The one or more non-transitory computer-readable storage media of claim 15, wherein identifying the respective muscle activities associated with the respective regions to which each of the one or more first activity sensors is coupled comprises generating a three-dimensional body posture model or a three-dimensional movement simulation model of the first particular sport user.

20. The operation includes: and identifying a second specific sports user, the identification comprising: acquiring second sensor data from the one or more second activity sensors, each coupled to a respective area of ​​the second sport user; acquiring second image data of the second sports user and the one or more second activity sensors coupled to the second sports user; Identifying the second sport user as the second specific sport user based on the second sensor data and the second image data by the machine learning model; determining, by the machine learning model, muscle activities of the second specific sports user associated with each region of the second specific sports user to which the one or more second activity sensors are coupled based on the second sensor data and the second image data; 16. The one or more non-transitory computer-readable storage media of claim 15, wherein the machine learning model identifies a specific movement performed by the second specific sports user based on the respective muscle activities of the identified second specific sports user.

Images