CN108452504B - Swimming posture analysis device and method based on sensor - Google Patents

Swimming posture analysis device and method based on sensor Download PDF

Info

Publication number
CN108452504B
CN108452504B CN201810714914.2A CN201810714914A CN108452504B CN 108452504 B CN108452504 B CN 108452504B CN 201810714914 A CN201810714914 A CN 201810714914A CN 108452504 B CN108452504 B CN 108452504B
Authority
CN
China
Prior art keywords
time
sensor
axis
swimming
stroke
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810714914.2A
Other languages
Chinese (zh)
Other versions
CN108452504A (en
Inventor
黄俊颕
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Kunshan Happy Island Sports Electronic Technology Co ltd
Original Assignee
Kunshan Happy Island Sports Electronic Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Kunshan Happy Island Sports Electronic Technology Co ltd filed Critical Kunshan Happy Island Sports Electronic Technology Co ltd
Priority to CN201810714914.2A priority Critical patent/CN108452504B/en
Publication of CN108452504A publication Critical patent/CN108452504A/en
Application granted granted Critical
Publication of CN108452504B publication Critical patent/CN108452504B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B69/00—Training appliances or apparatus for special sports
    • A63B69/12—Arrangements in swimming pools for teaching swimming or for training
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B24/00—Electric or electronic controls for exercising apparatus of preceding groups; Controlling or monitoring of exercises, sportive games, training or athletic performances
    • A63B24/0062—Monitoring athletic performances, e.g. for determining the work of a user on an exercise apparatus, the completed jogging or cycling distance
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B71/00—Games or sports accessories not covered in groups A63B1/00 - A63B69/00
    • A63B71/06—Indicating or scoring devices for games or players, or for other sports activities
    • A63B71/0619—Displays, user interfaces and indicating devices, specially adapted for sport equipment, e.g. display mounted on treadmills
    • A63B71/0622—Visual, audio or audio-visual systems for entertaining, instructing or motivating the user
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B71/00—Games or sports accessories not covered in groups A63B1/00 - A63B69/00
    • A63B71/06—Indicating or scoring devices for games or players, or for other sports activities
    • A63B71/0619—Displays, user interfaces and indicating devices, specially adapted for sport equipment, e.g. display mounted on treadmills
    • A63B71/0622—Visual, audio or audio-visual systems for entertaining, instructing or motivating the user
    • A63B2071/0625—Emitting sound, noise or music
    • A63B2071/063—Spoken or verbal instructions
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B71/00—Games or sports accessories not covered in groups A63B1/00 - A63B69/00
    • A63B71/06—Indicating or scoring devices for games or players, or for other sports activities
    • A63B71/0619—Displays, user interfaces and indicating devices, specially adapted for sport equipment, e.g. display mounted on treadmills
    • A63B2071/0658—Position or arrangement of display
    • A63B2071/0661—Position or arrangement of display arranged on the user
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B71/00—Games or sports accessories not covered in groups A63B1/00 - A63B69/00
    • A63B71/06—Indicating or scoring devices for games or players, or for other sports activities
    • A63B71/0619—Displays, user interfaces and indicating devices, specially adapted for sport equipment, e.g. display mounted on treadmills
    • A63B2071/0658—Position or arrangement of display
    • A63B2071/0661—Position or arrangement of display arranged on the user
    • A63B2071/0666—Position or arrangement of display arranged on the user worn on the head or face, e.g. combined with goggles or glasses
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B2220/00—Measuring of physical parameters relating to sporting activity
    • A63B2220/17—Counting, e.g. counting periodical movements, revolutions or cycles, or including further data processing to determine distances or speed
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B2220/00—Measuring of physical parameters relating to sporting activity
    • A63B2220/18—Inclination, slope or curvature
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B2220/00—Measuring of physical parameters relating to sporting activity
    • A63B2220/20—Distances or displacements
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B2220/00—Measuring of physical parameters relating to sporting activity
    • A63B2220/30—Speed
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B2220/00—Measuring of physical parameters relating to sporting activity
    • A63B2220/40—Acceleration
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B2220/00—Measuring of physical parameters relating to sporting activity
    • A63B2220/64—Frequency, e.g. of vibration oscillation
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B2225/00—Miscellaneous features of sport apparatus, devices or equipment
    • A63B2225/50—Wireless data transmission, e.g. by radio transmitters or telemetry
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B2230/00—Measuring physiological parameters of the user
    • A63B2230/62—Measuring physiological parameters of the user posture

Landscapes

  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Physical Education & Sports Medicine (AREA)
  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

The invention relates to a swimming posture analysis device and an analysis method based on a sensor, which comprises the following steps: 1) the palm data acquisition device including sensor module is worn in the palm to the training person, and sensor module includes: the three-axis accelerometer, the gyroscope sensor and the wireless communication module are arranged on the base; 2) the trainer wears the bone conduction sensor device on the head; 3) and (3) analysis: 4) and carrying out timely voice feedback of analysis data by utilizing the bone conduction sensor device. The invention uses a three-axis accelerometer and a gyroscope sensor to obtain the swimming action characteristic data, converts the obtained three-axis acceleration and a gyroscope sensing value into graphical data after processing, analyzes basic parameters of swimming training including stroke frequency, stroke width, stroke speed, efficiency and the like, and uses the gyroscope sensor to measure the inclination angle and the rotation angle of the body of a swimmer.

Description

Swimming posture analysis device and method based on sensor
Technical Field
The invention relates to the technical field of swimming teaching methods, in particular to a swimming posture analysis device and method based on a sensor.
Background
In the highly competitive swimming competition, the difference of the trainer is smaller and smaller, and the victory or defeat is often only within milliseconds; therefore, introducing scientific methods and techniques to capture, analyze, and fine-tune swimming movements during training has become a common requirement for elite swimmers; however, the related technologies are still rarely applied to swimming training, and mainly because the research training tool needs to have a waterproof function, the technology of how to transmit data in water is a problem which is difficult to overcome. Swimming training is based on the theory of action combination and action structure, and swimmers overcome resistance in water to move forwards by operating body limbs, so that the swimming training can effectively improve exercise performance by knowing and applying various parameters of strength, direction, track and the like of the body limbs.
The current research method for swimming technical analysis is based on kinetic and image analysis. In the aspect of dynamics, parameters such as speed, stroke frequency, stroke width and efficiency of a trainer are mainly analyzed, in the aspect of image analysis, the situation of three-dimensional motion of the trainer in water is mainly analyzed, and the relation between the rotation, balance and streamline position of a body and the swimming speed is discussed, however, some problems still need to be overcome by using a dynamics or image analysis method in the aspect of swimming training and research, for example, at least two underwater cameras are needed when image analysis is used, and a long time is needed to process a digitization process; motion in water may also be due to light refraction or human error of skin marking, which may not be able to use optical or motion capture systems to assist in motion capture, even due to visual occlusion caused by viewing angle
In addition, both of these two research methods have the problem of delayed feedback, and the trainers and coaches cannot immediately know the lack of motion; however, real-time feedback messages are important for swimming teaching and swimming training, and excellent coaches and trainees can provide high-quality feedback messages in a shortest time, which is beneficial for trainees or learners to immediately correct and improve the loss of motion.
Therefore, it is necessary to provide a sensor-based swimming stroke analysis device and method to solve the above problems.
Disclosure of Invention
The invention aims to provide a swimming posture analysis device and an analysis method based on a sensor.
The technical scheme is as follows:
a sensor-based swimming stroke analysis device, comprising:
palm data acquisition device: the device comprises a three-axis accelerometer, a gyroscope sensor and a wireless communication module;
a bone conduction sensor device worn on the head of a trainer.
Further, palm data acquisition device sets up in elasticity bracelet in-band.
A sensor-based swimming stroke analysis method, comprising the steps of:
1) the palm data acquisition device including sensor module is worn in the palm to the training person, and sensor module includes: the three-axis accelerometer, the gyroscope sensor and the wireless communication module are arranged on the base;
2) the trainer wears the bone conduction sensor device on the head;
3) and (3) analysis:
3-1) analysis of the influence of whether the sensor is worn or not on the swimming time and frequency of 50 meters: in order to know whether the trainer affects the completion time or the number of times of scratching the hand due to wearing the sensor, the trainer is respectively measured whether the sensor is worn or not, and the measurement result is analyzed by a paired sample tcalibration so as to inspect the influence of the sensor on the completion time and the number of times of scratching the hand;
3-2) analyzing the scratching frequency, scratching width, scratching speed and scratching efficiency: the following information can be obtained from the rower cycle analyzed by the patterned sensing data:
average dividing frequency: the number of times of scratching hands completed in unit time is indicated, and the average scratching frequency can be obtained according to the relation between the cycle number and the time axis in the oscillogram, wherein the unit is times/minute;
average drawing: the distance of advancing in a single stroke cycle is expressed in the unit of meter/time; the average stroke information can be obtained by calculating the relationship between the number of times of the stroke cycle and the length of the swimming pool.
Average dividing speed: the distance traveled in unit time is obtained by multiplying the stroke frequency and the stroke width, and the unit is meter/second;
the efficiency of paddling: the efficiency of arm stroke is defined as the product of two parameters of stroke speed and stroke width, and the unit is meter2Per second/time;
3-3) analysis of torso rotations: the gyro sensor fixed on the head is used to obtain the rotation angle of the trunk during swimming, and the change of the gyro in the X axis represents the fluctuation of the body on the water surface, the Y axis represents the rolling of the body, and the Z axis represents the left-right swinging of the body;
3-4) analyzing swimming stroke actions: referring to a shortswimming gesture decomposition mode proposed by Ohgi (2002), the shortswimming action is simplified into four action stages of water grabbing, water pushing, arm lifting, returning and the like, and the main change conditions corresponding to the triaxial acceleration sensor in each action stage are analyzed as follows:
stage I (water grab): the palm grips water outwards, and the X axis presents a negative value;
stage II (push water): after the water grabbing is finished, pushing water to the thigh side, and inverting the X axis to be a positive value; the Y axis exhibits a negative value;
stage III (lifting): after water is pushed, the lifting arm discharges water, and the Y axis and the Z axis are quickly thrown backwards to present a larger negative value;
stage IV (recovery): the X axis has no great change, the Y axis has a negative value and the Z axis has a positive value;
analyzing the waveform state of the sensing data of the triaxial accelerometer according to the obtained characteristic relationship, judging the detailed action composition of the swimming trainer, and obtaining the detailed process of the rower in water;
3-5) analysis of motion stability:
the sensing value of a single rower cycle is represented by a random variable, different rower cycles are represented by different random variables, and the correlation coefficient of the random variables of adjacent rower cycles is calculated; assuming that two adjacent rower actions are represented by random variables Ut and Vt respectively, and the time lengths of Ut and Vt are TU and TV, the amount of the captured sensing data is different because the time lengths of the two rower actions may be different; before the correlation coefficient calculation is performed, the normalization procedure is performed, and the details thereof are as follows:
time normalization procedure: lengthening the lengths of TU and TV to a preset time T, taking Ut as an example, and firstly calculating the lengthening multiplying power p for the Ut as T TU; when the time axis is elongated, the sensing data located on the original time axis also needs to be correspondingly changed, let t be a sensing value sampling point in the original Ut, and the sensing data at the time point t is u, the elongated position of the sensing data at the time point t is set to be t × p, and the sensing data at the original time point t is changed to be u/p 2;
sensing value frequency normalization procedure: since the amount of sensing data represented by Ut and Vt is not the same, in this procedure, we use the interpolation concept to insert the sensing points, so that the amount of sensing data in the random variables Ut and Vt can be the same; assuming that the time T is divided into n equal parts, at any point T of the n equal parts, the interpolated sensing value of the fixed equal part T is obtained by using the normalized sensing data am-1 and am adjacent to the times tm-1 and tm on the left and right of the time axis at the time point T, and the formula is as follows:
Figure BDA0001717139600000031
calculation of correlation coefficient program: the random variables Ut and Vt after the normalization are represented by U and V, respectively, the time lengths of the two random variables are both T, and the number of the sensing data is n, the expected values and the standard deviations are U and V, and SU and SV, respectively, and then the covariance SUV of U and V can be calculated as:
Figure BDA0001717139600000032
by using the definition of the correlation coefficient, the correlation coefficient r can be obtained as:
Figure BDA0001717139600000041
wherein the final r value will be between-1 and 1, the closer to 1 represents the trend of two random variables U and V being similar, i.e. the stability of the rower's motion in two cycles is higher;
4) and carrying out timely voice feedback of analysis data by utilizing the bone conduction sensor device.
Further, in the step 3-1), there is enough rest between the two measurements to avoid the trainer from influencing the measurement result due to fatigue.
Further, in step 3-3), the angular velocity value per unit time is integrated to obtain the progressive angle, and since the sampling value of the gyro sensor is not a continuous function, we use the trapezoidal area formed by the time and the sensing value in the sampling time interval as the moving angle in a certain time, and the details are as follows: assuming that the angular velocity values of a plurality of sampling times (t1, t2, t 3.). are (w (t1), w (t2), w (t3),. so, the moving angle between t1 and t2 can be obtained as a trapezoidal area in which the w (t1) value is an upper base, the w (t2) value is a lower base, and the interval between t2 and t1 is high, and the area is (upper base + lower base) × high/2 ═ w (t1) + w (t2)) × (t2-t 1)/2; and by analogy, the angle values in the time interval of t2 and t3 are obtained to obtain the change process of the trunk rotation angle in the whole motion process.
Further, in step 3-5), the trainee with higher sport technical level can make the operation performance reach a certain degree of accuracy under the repeated training for a long time, and the action shows that the rhythm is more stable and consistent (Southard & Miracle, 1993); in the step 3-5), the trainees are averagely divided into two groups according to the stroke ages to be trained: the group A is the one which is trained for less than 5 years, the group B is the one which is trained for more than 5 years, and the relationship between the training stroke age and the action stability is analyzed; the correlation coefficient concept is used to measure the similarity of the sensed values of the trainer in the adjacent rower periods, and determine whether the rower movement has consistency.
Compared with the prior art, the invention uses the three-axis accelerometer and the gyroscope sensor to obtain the characteristic data of the swimming action, converts the obtained three-axis acceleration and the gyroscope sensing value into graphical data after processing, analyzes the basic parameters of the swimming training including the stroke frequency, the stroke width, the stroke speed, the efficiency and the like, and uses the gyroscope sensor to measure the inclination angle and the rotation angle of the body of the swimmer, under the condition, three advanced action analysis and application can be provided:
1) analyzing the action characteristics of the swimmer in detail, judging the underwater detailed action composition of the swimmer, and correcting the action loss of the swimmer according to the detailed action composition analysis;
2) calculating a trend change of the captured sensed values using a concept of correlation coefficients, thereby observing the stability of the swimmer's motion behavior and tracing a difference in the swimmer's motion skill performance against the course of practicing swimming;
3) the underwater bone conduction technology is used for providing voice feedback in time, so that a trainer can immediately correct swimming actions, and the overall training efficiency is improved.
Drawings
FIG. 1 is one of the schematic diagrams of a palm-worn palm data collection device of a trainer in an embodiment of the invention.
FIG. 2 is a second schematic diagram of the palm-worn palm data collection device of the trainer in accordance with the embodiment of the present invention.
Fig. 3 is one of schematic illustrations of a trainer's head-worn conduction sensor apparatus in an embodiment of the invention.
Fig. 4 is a second schematic diagram of a conduction sensor device worn on the head of a trainer in an embodiment of the invention.
Fig. 5 is a continuous waveform diagram generated by the Y-axis value of the acceleration sensor.
Fig. 6 shows the integrated results of the angular velocity data measured by the palm gyroscope of the trainer a.
FIG. 7 is a graph of the change in the three-axis pattern of acceleration before and after correction of the action of the short swim by trainer A.
Detailed Description
Example (b):
please refer to fig. 1, fig. 2, fig. 3, fig. 4; this embodiment shows a sensor-based swimming stroke analysis method, comprising the steps of:
1) the palm data acquisition device 100 including sensor module is worn in the palm to the training person, and palm data acquisition device 100 sets up in elasticity bracelet area 101, and sensor module includes: the three-axis accelerometer, the gyroscope sensor and the wireless communication module are arranged on the base;
2) the trainer wears the conduction sensor apparatus 200 on the head;
3) and (3) analysis:
3-1) analysis of the influence of whether the sensor is worn or not on the swimming time and frequency of 50 meters: in order to know whether the trainer affects the completion time or the number of times of scratching the hand due to wearing the sensor, the trainer is respectively measured whether the sensor is worn or not, and the measurement result is analyzed by a paired sample tcalibration so as to inspect the influence of the sensor on the completion time and the number of times of scratching the hand;
3-2) analyzing the scratching frequency, scratching width, scratching speed and scratching efficiency: the following information can be obtained from the rower cycle analyzed by the patterned sensing data:
average dividing frequency: the number of times of scratching hands completed in unit time is indicated, and the average scratching frequency can be obtained according to the relation between the cycle number and the time axis in the oscillogram, wherein the unit is times/minute;
average drawing: the distance of advancing in a single stroke cycle is expressed in the unit of meter/time; the average stroke information can be obtained by calculating the relationship between the number of times of the stroke cycle and the length of the swimming pool.
Average dividing speed: the distance traveled in unit time is obtained by multiplying the stroke frequency and the stroke width, and the unit is meter/second;
the efficiency of paddling: the efficiency of arm stroke is defined as the product of two parameters of stroke speed and stroke width, and the unit is meter2Per second/time;
3-3) analysis of torso rotations: the gyro sensor fixed on the head is used to obtain the rotation angle of the trunk during swimming, and the change of the gyro in the X axis represents the fluctuation of the body on the water surface, the Y axis represents the rolling of the body, and the Z axis represents the left-right swinging of the body;
3-4) analyzing swimming stroke actions: referring to a shortswimming gesture decomposition mode proposed by Ohgi (2002), the shortswimming action is simplified into four action stages of water grabbing, water pushing, arm lifting, returning and the like, and the main change conditions corresponding to the triaxial acceleration sensor in each action stage are analyzed as follows:
stage I (water grab): the palm grips water outwards, and the X axis presents a negative value;
stage II (push water): after the water grabbing is finished, pushing water to the thigh side, and inverting the X axis to be a positive value; the Y axis exhibits a negative value;
stage III (lifting): after water is pushed, the lifting arm discharges water, and the Y axis and the Z axis are quickly thrown backwards to present a larger negative value;
stage IV (recovery): the X axis has no great change, the Y axis has a negative value and the Z axis has a positive value;
analyzing the waveform state of the sensing data of the triaxial accelerometer according to the obtained characteristic relationship, judging the detailed action composition of the swimming trainer, and obtaining the detailed process of the rower in water;
3-5) analysis of motion stability:
the sensing value of a single rower cycle is represented by a random variable, different rower cycles are represented by different random variables, and the correlation coefficient of the random variables of adjacent rower cycles is calculated; assuming that two adjacent rower actions are represented by random variables Ut and Vt respectively, and the time lengths of Ut and Vt are TU and TV, the amount of the captured sensing data is different because the time lengths of the two rower actions may be different; before the correlation coefficient calculation is performed, the normalization procedure is performed, and the details thereof are as follows:
time normalization procedure: lengthening the lengths of TU and TV to a preset time T, taking Ut as an example, and firstly calculating the lengthening multiplying power p for the Ut as T TU; when the time axis is elongated, the sensing data located on the original time axis also needs to be correspondingly changed, let t be a sensing value sampling point in the original Ut, and the sensing data at the time point t is u, the elongated position of the sensing data at the time point t is set to be t × p, and the sensing data at the original time point t is changed to be u/p 2;
sensing value frequency normalization procedure: since the amount of sensing data represented by Ut and Vt is not the same, in this procedure, we use the interpolation concept to insert the sensing points, so that the amount of sensing data in the random variables Ut and Vt can be the same; assuming that the time T is divided into n equal parts, at any point T of the n equal parts, the interpolated sensing value of the fixed equal part T is obtained by using the normalized sensing data am-1 and am adjacent to the times tm-1 and tm on the left and right of the time axis at the time point T, and the formula is as follows:
Figure BDA0001717139600000071
calculation of correlation coefficient program: the random variables Ut and Vt after the normalization are represented by U and V, respectively, the time lengths of the two random variables are both T, and the number of the sensing data is n, the expected values and the standard deviations are U and V, and SU and SV, respectively, and then the covariance SUV of U and V can be calculated as:
Figure BDA0001717139600000072
by using the definition of the correlation coefficient, the correlation coefficient r can be obtained as:
Figure BDA0001717139600000073
wherein the final r value will be between-1 and 1, the closer to 1 represents the trend of two random variables U and V being similar, i.e. the stability of the rower's motion in two cycles is higher;
4) and carrying out timely voice feedback of analysis data by utilizing the bone conduction sensor device.
In the step 3-1), enough rest exists between the two measurements so as to avoid the influence of the trainer on the measurement result due to fatigue.
In step 3-3), the angular velocity value of each unit time is integrated to obtain a progressive angle, and since the sampling value of the gyro sensor is not a continuous function, a trapezoidal area formed by the time and the sensing value in the sampling time interval is used as the moving angle in a certain time, which is detailed as follows: assuming that the angular velocity values of a plurality of sampling times (t1, t2, t 3.). are (w (t1), w (t2), w (t3),. so, the moving angle between t1 and t2 can be obtained as a trapezoidal area in which the w (t1) value is an upper base, the w (t2) value is a lower base, and the interval between t2 and t1 is high, and the area is (upper base + lower base) × high/2 ═ w (t1) + w (t2)) × (t2-t 1)/2; and by analogy, the angle values in the time interval of t2 and t3 are obtained to obtain the change process of the trunk rotation angle in the whole motion process.
In step 3-5), the trainee with higher sport technical level can make the operation performance reach a certain degree of precision under the long-time repeated training, and the action shows that the rhythm is more stable and consistent (Southard & Miracle, 1993); in the step 3-5), the trainees are averagely divided into two groups according to the stroke ages to be trained: the group A is the one which is trained for less than 5 years, the group B is the one which is trained for more than 5 years, and the relationship between the training stroke age and the action stability is analyzed; the correlation coefficient concept is used to measure the similarity of the sensed values of the trainer in the adjacent rower periods, and determine whether the rower movement has consistency.
Verification of the results of the examples was performed:
descriptive statistics and confidence analysis of experimental parameters:
the statistical result measured by the trainer wearing the sensor is shown as table one; the inter-class correlation (ICC) is performed by using the time and number of strokes obtained by repeated measurement, the reliability of the measured data is examined, and the results in the table I show that the inter-class correlation of each parameter of the two tests is between.965 and.988 (ICC >.70) no matter the stroke frequency, the stroke width, the stroke speed and the stroke efficiency, which shows that the data measured by the research tool has high reliability.
Table one summary table of descriptive statistics and intra-group correlation coefficients (n ═ 11)
Figure BDA0001717139600000081
Analysis of the effect of wearing sensors on swimming time and frequency:
the results of this study are shown in table two. In the experimental process, whether the sensor is worn or not is not obviously different from the time of finishing 50 meters and the number of times of scratching hands, and the trainer does not influence the athletic performance by wearing the sensor, so the result measured by the sensor is adopted in the research, and the actual performance can be specifically deduced.
Table two summary table of swimming time and hand-stroke times t test result with or without sensor (n ═ 11)
Figure BDA0001717139600000082
Analyzing the scratching frequency, scratching width, scratching speed and scratching efficiency:
taking 50 m fast swimming of the D trainer as an example, FIG. 5 is a continuous waveform diagram generated by the Y-axis value of the acceleration sensor, and the analysis result is explained below by taking the diagram as an example.
Average dividing frequency: the 50 meter swiftly-swimming Y-axis action time of the D trainer is 37.5 seconds, which results in 16 rower cycles, i.e. 32 rowers, so the average stroke frequency of the D trainer is 32/(37.5/60) times 51.20 times/min.
(II) average dividing: from the above results, since the D trainer used 32 rowers at 50 m, the average swath was 50/32-1.56 m/swath.
(III) average stroking speed: the stroke rate is obtained from the product of the stroke frequency and the stroke width, and the stroke rate of the D trainer is 51.20 times/min × 1.56 m/min, 79.87 m/min, 1.33 m/s.
(IV) water-rowing efficiency: the product of stroke speed and stroke width is the stroke efficiency, and the efficiency value of the trainer D is 1.33 m/s multiplied by 1.56 m/s which is 2.07 m/s 2/s.
Analysis of head rotation:
FIG. 6 shows the result of the angular rate data measured by the palm gyroscope of the trainer A after integration, and from FIG. 2, the change of the trainer A in the roll of the body can be found, the roll angle of the body is about 64-67 degrees when the right side is ventilated, and the roll angle of the body is about 60-61 degrees on the non-ventilated side; due to the degree of asymmetry between the two hand strokes, and the guidance of the ventilation action, the ventilation side is usually larger than the non-ventilation side. In the case of up-and-down fluctuation, the up-and-down fluctuation angle of the trainer A is between 12 degrees and-6 degrees, and the up-and-down fluctuation of the body is caused because the arms grab water and stroke downwards to generate lift force, the stable and small fluctuation is the characteristic of an excellent trainer, and when the up-and-down fluctuation is too large, whether the stroke track and the depth are proper or not is reviewed. In the aspect of left-right swinging, the Z axis of the trainer A shows a tendency of increasing angles, which represents the phenomenon of rightward deviation in the advancing process, the left-right swinging can cause the body to deviate rightward due to wrong arm water inlet position, too outer left-hand water stroke path or unstable relation with the core muscle group, and after the body swims for a certain distance, reverse adjustment must be carried out, so that the swimming distance can be lengthened, redundant actions are generated, and the efficiency is reduced.
Analyzing swimming posture actions:
FIG. 7 is a graph showing the three-axis acceleration pattern of the person training A before and after correction of the action of the swimming. As can be seen from the action waveform before correction in FIG. 3, the X-axis of player A in the water-catching stage shows two downward waveforms, which represent that the water-catching path is biased to the left, so that the subsequent segment needs to be corrected to wipe out, and after comparing with the underwater image, the hand-wiping process of player A is confirmed to be the same. In addition, in the Y-axis and Z-axis sensing waveforms of player a, it can be found that transient surges occur in each of the paddling stages, which means that extra movements are generated in the paddling process of player a, and the paddling process is not smooth enough. After the analyzed data is provided for the coach as feedback and action correction, the A player obviously improves the action loss, the waveform after the action correction is shown in the right diagram of fig. 7, the three axial waveform diagrams are smoother, the noise is reduced, and the problem of action continuity is improved. The water pushing stage is further enhanced in the distribution of the action ratio; the ratio of the player's action time to the single rower cycle time at each stage is detailed in table three:
Figure BDA0001717139600000101
the result after the player corrects the action can be obviously seen from the change of the time ratio, the water catching stage is reduced from 30.8% to 26.5%, the water pushing stage is reduced from 24.1% to 22.3%, the arm lifting stage is reduced from 29.7% to 25.1%, the recovery stage is increased from 15.4% to 26.1%, the time ratio of the action in water is shortened in action correction, the water pushing process is stronger and more powerful to generate forward pushing force, and the relaxing opportunity is obtained by more gentle recovery action; the single rower cycle time is extended from 1.82 seconds to 2.15 seconds, the single rower frequency is modified from 32.96 times/minute to 27.91 times/minute, slowing the rower and putting the force in the right place will help to improve rower efficiency. The analysis of the motion waveform is provided to the trainer to be used as a basis for optimizing the motion of the trainer.
Compared with the prior art, the embodiment uses the mobile carrier with built-in sensor and waterproof function as a tool, is applied to collect the action change situation of the swimmer in water, and the related experiments show that the sensor wearing does not affect the swimming time and frequency, and the data measured by the research tool has extremely high reliability, and contributes to the following swimming analysis modes: 1) analyzing the period and the action time of the action waveform to obtain the information of the stroke frequency of the player, obtaining the average stroke width by the relationship between the action period and the length of the swimming pool, and obtaining parameters such as stroke speed, stroke efficiency and the like after operation; 2) the angular velocity data measured by the gyroscope sensor fixed on the head is subjected to integral operation to obtain the rotating angle of the trunk of the swimmer; 3) after the information measured by the palm acceleration sensor is converted into action waveforms, the swimming gestures can be analyzed according to the obtained action waveforms and used as the reference for correcting actions; 4) evaluating the action stability of the player by using the correlation coefficient of the previous action period and the later action period, and displaying that the action stability of the research object tends to be stable along with the increase of the swimming age; 5) the sensor and the bone conduction technology are combined, so that the action parameter information can be quickly and effectively obtained, and the action efficiency of the player can be checked in real time and the action correction can be assisted.
What has been described above are merely some embodiments of the present invention. It will be apparent to those skilled in the art that various changes and modifications can be made without departing from the inventive concept thereof, and these changes and modifications can be made without departing from the spirit and scope of the invention.

Claims (4)

1. A swimming posture analysis method based on a sensor is characterized in that: the method comprises the following steps:
1) the palm data acquisition device including sensor module is worn in the palm to the training person, and sensor module includes: the three-axis accelerometer, the gyroscope sensor and the wireless communication module are arranged on the base;
2) the trainer wears the bone conduction sensor device on the head;
3) and (3) analysis:
3-1) analysis of the influence of whether the sensor is worn or not on the swimming time and frequency of 50 meters: in order to know whether the trainer affects the completion time or the number of times of scratching the hand due to wearing the sensor, the trainer is respectively measured whether the sensor is worn or not, and the measurement result is analyzed by a paired sample tcalibration so as to inspect the influence of the sensor on the completion time and the number of times of scratching the hand;
3-2) analyzing the scratching frequency, scratching width, scratching speed and scratching efficiency: the following information can be obtained from the rower cycle analyzed by the patterned sensing data:
average dividing frequency: the number of times of scratching hands completed in unit time is indicated, and the average scratching frequency can be obtained according to the relation between the cycle number and the time axis in the oscillogram, wherein the unit is times/minute;
average drawing: the distance of advancing in a single stroke cycle is expressed in the unit of meter/time; calculating the relationship between the number of times of the rower's action period and the length of the swimming pool to obtain the average rowing information;
average dividing speed: the distance traveled in unit time is obtained by multiplying the stroke frequency and the stroke width, and the unit is meter/second;
the efficiency of paddling: the efficiency of arm stroke is defined as the product of two parameters of stroke speed and stroke width, and the unit is meter2Per second/time;
3-3) analysis of torso rotations: the gyro sensor fixed on the head is used for obtaining the rotation angle of the trunk during swimming, and the change condition of the X axis of the gyro represents the fluctuation of the body on the water surface, the Y axis represents the rolling of the body, and the Z axis represents the left-right swinging of the body according to the axial diagram;
3-4) analyzing swimming stroke actions: referring to the agility swimming gesture decomposition mode provided by Ohgi2002, the agility swimming action is simplified into four action stages of water grabbing, water pushing, arm lifting, returning and the like, and the main change conditions corresponding to the three-axis accelerometer in each action stage are analyzed as follows:
stage I water catching: the palm grips water outwards, and the X axis presents a negative value;
stage II water pushing: after the water grabbing is finished, pushing water to the thigh side, and inverting the X axis to be a positive value; the Y axis exhibits a negative value;
stage III lifting: after water is pushed, the lifting arm discharges water, and the Y axis and the Z axis are quickly thrown backwards to present a larger negative value;
and (3) recovering in a stage IV: the X axis has no great change, the Y axis has a negative value and the Z axis has a positive value;
analyzing the waveform state of the sensing data of the triaxial accelerometer according to the obtained characteristic relationship, judging the detailed action composition of the swimming trainer, and obtaining the detailed process of the rower in water;
3-5) analysis of motion stability:
the sensing value of a single rower cycle is represented by a random variable, different rower cycles are represented by different random variables, and the correlation coefficient of the random variables of adjacent rower cycles is calculated; suppose that two adjacent rowers act with random variables UtAnd VtTo represent, and UtAnd VtThe time length of (1) is TU and TV, since there are twoThe time lengths of the rowers may be different, and the quantity of the captured sensing data is different; before the correlation coefficient calculation is performed, the normalization procedure is performed, and the details thereof are as follows:
time normalization procedure: lengthening TU and TV to a preset time T by UtFor example, first calculate for UtThe drawing multiplying power of (1) is p ═ TTU; when the time axis is elongated, the sensing data located on the original time axis also needs to be correspondingly changed, so that t is the original UtA sensing data is sampled at a time point t, and the elongated position is set to t × p for the time point t, and the sensing data at the time point t is changed to u/p2；
Sensing value frequency normalization procedure: due to UtAnd VtThe amount of sensing data represented is not the same, and in this procedure we use the concept of interpolation to insert sensing points such that the random variable U istAnd VtThe number of sensing data in (1) is the same; assuming that the T time is divided into n equal parts, at any point T of the n equal parts, the time T is adjacent to the time T on the left and right sides of the time axism-1And tmThe normalized sensing data am-1And amTo obtain the interpolated sensing value of the fixed bisector t, the formula is as follows:
Figure FDA0002268774290000021
calculation of correlation coefficient program: the random variable U after the two normalization processes are executedtAnd VtRespectively expressed by U and V, the time lengths of the two random variables are both T, and the number of the sensed data is n, the expected values and standard deviations are respectively U and V, SU and SV, and then the covariance S of U and V can be calculatedUVComprises the following steps:
Figure FDA0002268774290000022
by using the definition of the correlation coefficient, the correlation coefficient r can be obtained as:
Figure FDA0002268774290000023
wherein the final r value will be between-1 and 1, the closer to 1 represents the trend of two random variables U and V being similar, i.e. the stability of the rower's motion in two cycles is higher;
4) and carrying out timely voice feedback of analysis data by utilizing the bone conduction sensor device.
2. The sensor-based swimming stroke analysis method according to claim 1, wherein: in the step 3-1), enough rest exists between the two measurements so as to avoid the influence of the trainer on the measurement result due to fatigue.
3. The sensor-based swimming stroke analysis method according to claim 2, wherein: in step 3-3), the angular velocity value of each unit time is integrated to obtain a progressive angle, and since the sampling value of the gyro sensor is not a continuous function, a trapezoidal area formed by the time and the sensing value in the sampling time interval is used as the moving angle in a certain time, which is detailed as follows: assume that a plurality of sampling times (t) are known1,t2,t3,..) the angular velocity values are (w (t)1),w(t2),w(t3) ,..), and can therefore determine t1And t2The moving angle between w (t)1) The numerical value is used as the upper base, w (t)2) Numerical values as the bottom, (t)2-t1) A ladder-shaped area with a high interval, which is (upper bottom + lower bottom) x high/2 ═ w (t)1)+w(t2))×(t2-t1) 2; and analogize to obtain t2And t3The angle value in the time interval obtains the change process of the rotating angle of the trunk in the whole movement process.
4. A sensor-based swimming stroke analysis method according to claim 3, characterized in that: in the step 3-5), a trainer with higher sport technical level can make the operation performance reach a certain degree of accuracy under the long-time repeated training, and the rhythm of the action display is more stable and consistent; in the step 3-5), the trainees are averagely divided into two groups according to the stroke ages to be trained: the group A is the one which is trained for less than 5 years, the group B is the one which is trained for more than 5 years, and the relationship between the training stroke age and the action stability is analyzed; the correlation coefficient concept is used to measure the similarity of the sensed values of the trainer in the adjacent rower periods, and determine whether the rower movement has consistency.
CN201810714914.2A 2018-07-03 2018-07-03 Swimming posture analysis device and method based on sensor Active CN108452504B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810714914.2A CN108452504B (en) 2018-07-03 2018-07-03 Swimming posture analysis device and method based on sensor

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810714914.2A CN108452504B (en) 2018-07-03 2018-07-03 Swimming posture analysis device and method based on sensor

Publications (2)

Publication Number Publication Date
CN108452504A CN108452504A (en) 2018-08-28
CN108452504B true CN108452504B (en) 2020-04-14

Family

ID=63216302

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810714914.2A Active CN108452504B (en) 2018-07-03 2018-07-03 Swimming posture analysis device and method based on sensor

Country Status (1)

Country Link
CN (1) CN108452504B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109529295B (en) * 2018-10-29 2021-01-19 叶强 Exercise posture correction trainer and training method
TW202031321A (en) 2019-02-25 2020-09-01 財團法人工業技術研究院 Swimming posture correction method and swimming posture correction system
CN111603746A (en) * 2019-02-25 2020-09-01 财团法人工业技术研究院 Swimming posture correction method and system
CN111167100B (en) * 2019-06-23 2023-11-14 广西北海智涵生物科技开发有限公司 Intelligent swimming learning ring
IT201900014436A1 (en) 2019-08-08 2021-02-08 Iesteam Srl System for the automatic detection of chronometric performances in a swimming pool
CN112076456A (en) * 2020-08-07 2020-12-15 楚林溪 Swimming training method and system, terminal device, swimming goggles and readable storage medium
CN114602155B (en) * 2022-05-11 2023-02-21 荣耀终端有限公司 Swimming information statistical method, computer-readable storage medium and electronic device
CN116808553B (en) * 2023-08-04 2023-12-29 国家体育总局体育科学研究所 System for quantitatively researching specific technical level of athletic swimmers

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105934262A (en) * 2013-12-20 2016-09-07 X米特里克斯运动有限公司 Wearable detector for the detection of parameters related to a physical activity
CN106073707A (en) * 2013-06-03 2016-11-09 飞比特公司 The use of the gyroscope in individual's body-building tracking device and by bike activity

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2008032315A1 (en) * 2006-09-11 2008-03-20 Eldad Shemesh Training and instructing support device
US20120009553A1 (en) * 2010-07-07 2012-01-12 Ben-Tal Eran Apparatus for assisting swimming training

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106073707A (en) * 2013-06-03 2016-11-09 飞比特公司 The use of the gyroscope in individual's body-building tracking device and by bike activity
CN105934262A (en) * 2013-12-20 2016-09-07 X米特里克斯运动有限公司 Wearable detector for the detection of parameters related to a physical activity

Also Published As

Publication number Publication date
CN108452504A (en) 2018-08-28

Similar Documents

Publication Publication Date Title
CN108452504B (en) Swimming posture analysis device and method based on sensor
US9414784B1 (en) Movement assessment apparatus and a method for providing biofeedback using the same
US9173596B1 (en) Movement assessment apparatus and a method for providing biofeedback using the same
Wang et al. Using wearable sensors to capture posture of the human lumbar spine in competitive swimming
US20140229135A1 (en) Motion analysis apparatus and motion analysis method
RU2013134980A (en) DEVICE AND METHOD FOR GOLF SWING ANALYSIS
Chun et al. A sensor-aided self coaching model for uncocking improvement in golf swing
US20220161117A1 (en) Framework for recording and analysis of movement skills
Sanders et al. Reliability of three-dimensional linear kinematics and kinetics of swimming derived from digitized video at 25 and 50 Hz with 10 and 5 frame extensions to the 4th order Butterworth smoothing window
AlFadly et al. Influence Technique Training Data sensor (Triton Wear) To improve biomechanical variables for some stages Performance and achievement 50m freestyle youth
Srivastava et al. Efficient characterization of tennis shots and game analysis using wearable sensors data
Glazier et al. Inter-and intra-individual movement variability in the golf swing
Lintmeijer et al. Real-time feedback on mechanical power output: facilitating crew rowers’ compliance with prescribed training intensity
CN109529295B (en) Exercise posture correction trainer and training method
Engel et al. Intra-cyclic analysis of the butterfly swimming technique using an inertial measurement unit
KR101257690B1 (en) Swing Analysis Apparatus attached on Golf Club
US11504595B2 (en) Dry swim simulation system
US11461905B2 (en) Determining a kinematic sequence
Dhinesh et al. Tennis serve correction using a performance improvement platform
US20150335946A1 (en) Tennis training system
Skublewska-Paszkowska et al. Analysis of rowing based on handle trajectory
Atack The biomechanics of rugby place kicking
JP4361231B2 (en) Electronic measuring equipment
Cabrera-Arellano et al. iPaddle: An Intelligent Swimming Training System
JP2019208690A (en) Evaluation method and evaluation system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant