Artificial intelligence sports equipment training system based on motion capture

Abstract

The invention relates to the technical field of exercise data analysis, and discloses an artificial intelligence sports equipment training system based on exercise capture, which comprises a training data acquisition module, a training data analysis module and a training data evaluation module, the training data acquisition module comprises a historical training data acquisition unit, a current training data acquisition unit and a training medical history acquisition unit. According to the system, historical training data of a current training person and body data of the current training person are collected and compared in a database, the training weight of the same level, training image data in the current training process and medical history uploaded by a user are output and analyzed, and the training weight of the same level, the training image data and the medical history uploaded by the user are obtained. According to the method, the historical training coefficient, the same-level training weight, the current training coefficient and the medical history influence coefficient are comprehensively calculated to obtain the training improvement coefficient, the training of the user is adjusted based on the training improvement coefficient, and the training weight of the user is reduced or increased, so that the exercise risk is reduced, and the exercise injury is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of sports data analysis, and more particularly to an artificial intelligence sports equipment training system based on motion capture. Background Art

[0002] Sports equipment refers to all kinds of instruments, equipment and supplies used in competitive sports competitions and fitness exercises. Sports equipment and sports are interdependent and mutually promoting. The popularization of sports and the diversification of sports events have developed the types, specifications, etc. of sports equipment. In order to get better physical fitness, people often use sports equipment such as dumbbells and barbells for exercise. However, at present, in most cases, the exercise personnel subjectively judge whether to increase the weight. For novice exercisers, the subjective judgment ability is often weak, and it is easy to blindly increase the weight during exercise, resulting in the risk of injury. Summary of the Invention

[0003] (I) Technical Problems to be Solved In view of the deficiencies of the prior art, the present invention provides an artificial intelligence sports equipment training system based on motion capture, which has the advantages of giving an increase or decrease in the exercise weight and avoiding sports injuries, and solves the above technical problems.

[0004] (II) Technical Solutions To achieve the above object, the present invention provides the following technical solution: An artificial intelligence sports equipment training system based on motion capture, comprising a training data acquisition module, a training data analysis module and a training data evaluation module; The training data acquisition module includes a historical training data acquisition unit, a body data acquisition unit, a current training data acquisition unit and a training medical history acquisition unit. The historical training data acquisition unit is used to acquire the historical training data of the current training personnel. The body data acquisition unit is used to acquire the body data of the current training personnel, compare it in the database, and output the training weight at the same level. The current training data acquisition unit is used to acquire the training image data during the current training process. The training medical history acquisition unit is for the user to upload the medical history; The training data analysis module includes a training data analysis unit, a training process analysis unit and a medical history analysis unit. The training data analysis unit analyzes and calculates the historical training coefficient based on the user's historical training data. The training process analysis unit analyzes the training process based on the training image data and outputs the current training coefficient. The medical history analysis unit comprehensively analyzes and outputs the medical history influence coefficient based on the user's uploaded medical history; The training data evaluation module comprehensively calculates the training improvement coefficient based on the historical training coefficient, the training weight at the same level, the current training coefficient and the medical history influence coefficient, and adjusts the user's training based on the training improvement coefficient.

[0005] As a preferred technical solution of the present invention, the specific expression for the historical training data acquisition unit to acquire the historical training data of the current trainee is as follows:

[0006] Among them, represents the historical training data set, respectively represent the first training data of the trainee, , the th training data, , the th training data, and the th training data has the following specific expression:

[0007] Among them, represents the th training weight of the trainee, represents the growth degree of the th training of the trainee, and the specific expression is as follows:

[0008] Among them, represents the th training weight of the trainee, represents the th training weight of the trainee.

[0009] As a preferred technical solution of the present invention, the specific expression for the current training data acquisition unit to acquire the training image data during the current training process is as follows:

[0010] Among them, represents the training image data set, respectively represent the first set of motion strokes intercepted during the training, , the th set of motion strokes, , the th set of motion strokes. The motion stroke includes the coordinates of the starting point and the ending point of the barbell end stroke, which are obtained by the Openpose model.

[0011] As a preferred technical solution of the present invention, the body data acquisition unit is used to acquire the body data of the current trainee, and the specific expression is as follows:

[0012] Among them, Represents the height of the current trainee, Represents the weight of the current trainee, and the body data acquisition unit records the height of the current trainee And the weight of the current trainee Compare in the database to obtain several training weights that are exactly the same as the height and weight of the current trainee, and calculate the same-level training weight. The specific expression is as follows:

[0013] Among them, Represents the same-level training weight, Represents the sum of a total of Training weights that are exactly the same as the height and weight of the current trainee, Represents the th training weight that is exactly the same as the height and weight of the current trainee.

[0014] As a preferred technical solution of the present invention, the specific expression for the training medical history collection unit to upload the medical history by the user is as follows:

[0015] Among them, Represents the medical history dataset, Represents the first medical history uploaded by the user, , the th medical history, , the th medical history.

[0016] As a preferred technical solution of the present invention, the specific expression for the training data analysis unit to analyze and calculate the historical training coefficient based on the user's historical training data is as follows:

[0017] Among them, Represents the growth degree of the th training of the trainee, Represents the historical training coefficient, Represents the sum of a total of Growth degrees of training.

[0018] As a preferred technical solution of the present invention, the specific expression for the training process analysis unit to analyze the training process based on the training image data and output the current training coefficient is as follows:

[0019] Among them, Represents the current training coefficient, Represents the sum of a total of Sum the travel deviations of the secondary motion travel. Denote the motion deviation of the

[0020] group of motion travels. Its specific expression is as follows: Denote the radian angle of 180°. Denote the arctangent value of Denote the slope between the coordinates of the starting point and the ending point of the

[0021] As a preferred technical solution of the present invention, the specific expression of the medical history influence coefficient output by the medical history analysis unit based on the comprehensively analyzed medical history uploaded by the user is as follows:

[0022] Wherein, Denote the number of medical histories affecting motion in the medical history dataset. Denote the medical history influence coefficient.

[0023] As a preferred technical solution of the present invention, the specific expression of the training improvement coefficient calculated by the training data evaluation module based on the historical training coefficient, the same-level training weight, the current training coefficient, and the medical history influence coefficient is as follows:

[0024] Wherein, Denote the balance value. Denote the training improvement coefficient. Denote the training weight of the trainer for the th time. Denote the same-level training weight. Denote the current training coefficient. Denote the historical training coefficient. Denote the medical history influence coefficient.

[0025] As a preferred technical solution of the present invention, the specific steps for the training data evaluation module to adjust the user's training based on the training improvement coefficient are as follows: When the training improvement coefficient the training threshold, increase the training weight of the current user by 2% - 3%; When 80% of the training threshold the training improvement coefficient the training threshold, keep the training weight of the current user unchanged; When the training improvement coefficient When the training threshold is *80%, the training weight of the current user is reduced by 1% - 2%.

[0026] Compared with the prior art, the present invention provides an artificial intelligence sports equipment training system based on motion capture, which has the following beneficial effects: The present invention collects the historical training data and physical data of the current trainer, compares them in the database, and outputs the training weight at the same level, the training image data during the current training process, and the medical history uploaded by the user, and analyzes them to obtain the historical training coefficient, the training weight at the same level, the current training coefficient, and the medical history influence coefficient. Based on this, the training improvement coefficient is comprehensively calculated, and the training of the user is adjusted based on the training improvement coefficient, either reducing or increasing the training weight of the user, thereby ensuring the reduction of sports risks and avoiding sports injuries. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0029] Please refer to Figure 1 , an artificial intelligence sports equipment training system based on motion capture, including a training data acquisition module, a training data analysis module, and a training data evaluation module; The training data acquisition module includes a historical training data acquisition unit, a physical data acquisition unit, a current training data acquisition unit, and a training medical history acquisition unit. The historical training data acquisition unit is used to collect the historical training data of the current trainer, the physical data acquisition unit is used to collect the physical data of the current trainer, and compare them in the database, and output the training weight at the same level. The current training data acquisition unit is used to collect the training image data during the current training process. The training medical history acquisition unit uploads the medical history by the user. The specific expression for the historical training data acquisition unit to collect the historical training data of the current trainer is as follows:

[0030] Wherein, represents the historical training data set, respectively represent the training data of the trainer for the first time, , the th training data, , the th training data, and the th training data has the following specific expression:

[0031] Among them, represents the th training weight of the trainer, represents the growth degree of the th training of the trainer. The specific expression is as follows:

[0032] Among them, represents the th training weight of the trainer, represents the th training weight of the trainer. The specific expression for the current training data acquisition unit to acquire the training image data during the current training process is as follows:

[0033] Among them, represents the training image data set, respectively represent the first set of movement strokes intercepted during the training process, , the th set of movement strokes, , the th set of movement strokes. The movement stroke includes the coordinates of the starting point and the ending point of the barbell end stroke, which are obtained by the Openpose model. The body data acquisition unit is used to acquire the body data of the current trainer. The specific expression is as follows:

[0034] Among them, represents the height of the current trainer, represents the weight of the current trainer. The body data acquisition unit compares the height of the previous trainer and the weight of the current trainer in the database to obtain several training weights that are exactly the same as the height and weight of the current trainer, and calculates the same-level training weight. The specific expression is as follows:

[0035] Among them, represents the same-level training weight, represents the sum of training weights that are exactly the same as the height and weight of the current trainer, represents the The training weight for a current trainee with exactly the same height and weight. The specific expression for the training medical history collection unit where the user uploads the medical history is as follows:

[0036] Among them, represents the medical history data set, represents the first medical history uploaded by the user, , the th medical history, , the th medical history; The training data analysis module includes a training data analysis unit, a training process analysis unit, and a medical history analysis unit. The training data analysis unit analyzes and calculates the historical training coefficient based on the user's historical training data. The training process analysis unit analyzes the training process based on the training image data and outputs the current training coefficient. The medical history analysis unit comprehensively analyzes the user-uploaded medical history and outputs the medical history influence coefficient. The specific expression for the training data analysis unit to analyze and calculate the historical training coefficient based on the user's historical training data is as follows:

[0037] Among them, represents the growth degree of the th training of the trainee, represents the historical training coefficient, represents the sum of the growth degrees of a total of trainings. The specific expression for the training process analysis unit to analyze the training process based on the training image data and output the current training coefficient is as follows:

[0038] Among them, represents the current training coefficient, represents the sum of the travel deviations of a total of movement trips, represents the movement deviation of the th group of movement trips. Its specific expression is as follows:

[0039] Among them, represents the radian angle of 180°, represents the arctangent value of, represents the The slope between the coordinates of the starting point and the ending point of the movement stroke of the group, where the stroke coordinates XOY plane obtained by photographing from the side of the barbell are perpendicular to the axis of the barbell. The medical history analysis unit comprehensively analyzes the user-uploaded medical history and outputs the specific expression of the medical history influence coefficient as follows:

[0040] Among them, represents the number of medical histories in the medical history dataset that have an impact on the movement, represents the medical history influence coefficient; The training data evaluation module comprehensively calculates the training improvement coefficient based on the historical training coefficient, the same-level training weight, the current training coefficient, and the medical history influence coefficient, and adjusts the user's training based on the training improvement coefficient. The specific expression is as follows:

[0041] Among them, represents the balance value, represents the training improvement coefficient, represents the training weight of the trainer for the th time, represents the same-level training weight, represents the current training coefficient, represents the historical training coefficient, represents the medical history influence coefficient. The specific steps for the training data evaluation module to adjust the user's training based on the training improvement coefficient are as follows: When the training improvement coefficient the training threshold, the training weight of the current user is increased by 2% - 3%; When the training threshold * 80% the training improvement coefficient the training threshold, the training weight of the current user remains unchanged; When the training improvement coefficient the training threshold * 80%, the training weight of the current user is decreased by 1% - 2%; Embodiment In this embodiment, the movement during training is barbell bench press. The height of the current trainer is 175 cm and the weight is 75 kg. , and there are 2 medical histories that affect the movement, with sports injuries in the wrist and elbow. , = 0.25. The historical training data of the current trainer is recorded in Table 1 below: Table 1

[0042] The slope between the coordinates of the starting point and the ending point of the currently trained motion stroke is shown in Table 2 below: Table 2

[0043] Currently trained coefficient , Training threshold * 80% = 40. At this time, the training weight of the current user is reduced by 1% - 2%.

[0044] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An artificial intelligence sports equipment training system based on motion capture, characterized in that: It includes a training data acquisition module, a training data analysis module and a training data evaluation module; The training data collection module includes a historical training data collection unit, a physical data collection unit, a current training data collection unit and a training medical history collection unit. The historical training data collection unit is used to collect the historical training data of the current trainee, the physical data collection unit is used to collect the physical data of the current trainee, and compare them in the database, and output the training weights at the same level, the current training data collection unit is used to collect the training image data in the current training process, and the training medical history collection unit is used to upload the medical history by the user; The training data analysis module includes a training data analysis unit, a training process analysis unit and a medical history analysis unit. The training data analysis unit analyzes and calculates the historical training coefficient based on the user's historical training data. The training process analysis unit analyzes the training process based on the training image data and outputs the current training coefficient. The medical history analysis unit comprehensively analyzes the medical history uploaded by the user and outputs the medical history influence coefficient. The training data evaluation module obtains a training improvement coefficient based on a comprehensive calculation of the historical training coefficient, the training weight at the same level, the current training coefficient and the medical history influence coefficient, and adjusts the user's training based on the training improvement coefficient.

2. The artificial intelligence sports equipment training system based on motion capture according to claim 1, characterized in that: The specific expression of the historical training data collection unit collecting the historical training data of the current trainee is as follows: in, represents the historical training dataset, Respectively represent the first training data of the trainer, , No. training data, , No. training data, and Training data The specific expression is as follows: in, Indicates that the training staff Training weight, Indicates that the training staff The specific expression of the training growth degree is as follows: in, Indicates that the training staff Training weight, Indicates that the training staff training weight.

3. The artificial intelligence sports equipment training system based on motion capture according to claim 2, characterized in that: The specific expression of the current training data acquisition unit acquiring the training image data in the current training process is as follows: in, represents the training image dataset, They represent the first group of movement strokes intercepted during the training process, , No. Group movement itinerary, , No. A movement stroke is formed, wherein the movement stroke includes the coordinates of the starting point of the barbell end stroke and the coordinates of the end point of the stroke, which are obtained by the Openpose model.

4. The artificial intelligence sports equipment training system based on motion capture according to claim 3, characterized in that: The body data collection unit is used to collect the body data of the current trainee, and the specific expression is as follows: in, Indicates the height of the current trainee. represents the weight of the current trainee, and the body data collection unit collects the height of the current trainee and the current trainee's weight Compare in the database to obtain several training weights that are completely consistent with the current trainee's height and weight, and calculate the training weight of the same level. The specific expression is as follows: in, Indicates the training weight at the same level, Expressing The training weights of the current trainees with the same height and weight are summed up. Indicates A training weight that is exactly the same height and weight as the current training personnel.

5. The artificial intelligence sports equipment training system based on motion capture according to claim 4, characterized in that: The specific expression of the medical history uploaded by the user in the training medical history collection unit is as follows: in, represents the medical history dataset, Indicates the first medical history uploaded by the user. , No. Medical history, , No. A medical history.

6. The artificial intelligence sports equipment training system based on motion capture according to claim 5, characterized in that: The specific expression of the historical training coefficient obtained by the training data analysis unit through analysis and calculation based on the user's historical training data is as follows: in, Indicates that the training staff The training increase represents the historical training coefficient, Expressing The growth rate of each training session is summed.

7. The artificial intelligence sports equipment training system based on motion capture according to claim 6, characterized in that: The training process analysis unit analyzes the training process based on the training image data and outputs the specific expression of the current training coefficient as follows: in, represents the current training coefficient, Expressing The travel deviation of each movement is summed up. Indicates The specific expression of the motion deviation of the group motion stroke is as follows: in, represents an angle of 180° in arc. express The inverse tangent of Indicates The slope between the coordinates of the start point and the end point of a group motion stroke.

8. The artificial intelligence sports equipment training system based on motion capture according to claim 7, characterized in that: The specific expression of the medical history analysis unit for comprehensively analyzing the medical history uploaded by the user and outputting the medical history influence coefficient is as follows: in, represents the number of medical histories that have an impact on exercise in the medical history dataset, Represents the medical history influence coefficient.

9. The artificial intelligence sports equipment training system based on motion capture according to claim 8, characterized in that: The training data evaluation module calculates the training improvement coefficient based on the historical training coefficient, the same-level training weight, the current training coefficient and the medical history influence coefficient as follows: in, represents the equilibrium value, represents the training improvement coefficient, Indicates that the training staff Training weight, Indicates the training weight at the same level, represents the current training coefficient, represents the historical training coefficient, Represents the medical history influence coefficient.

10. The artificial intelligence sports equipment training system based on motion capture according to claim 9, characterized in that: The specific steps of the training data evaluation module adjusting the user's training based on the training improvement coefficient are: When training improves the coefficient When training the threshold, the training weight for the current user is increased by 2% to 3%; When the training threshold * 80% Training improvement factor When training the threshold, the training weight for the current user remains unchanged; When training improves the coefficient When the training threshold is *80%, the training weight for the current user is reduced by 1% to 2%.

Images