CN109603142A - A kind of the identification method of counting and its device of dumbbell exercise - Google Patents
A kind of the identification method of counting and its device of dumbbell exercise Download PDFInfo
- Publication number
- CN109603142A CN109603142A CN201811426205.0A CN201811426205A CN109603142A CN 109603142 A CN109603142 A CN 109603142A CN 201811426205 A CN201811426205 A CN 201811426205A CN 109603142 A CN109603142 A CN 109603142A
- Authority
- CN
- China
- Prior art keywords
- acceleration
- dumbbell
- signal
- module
- exercise
- 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.)
- Pending
Links
Classifications
-
- 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
- A63B71/0669—Score-keepers or score display devices
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B21/00—Exercising apparatus for developing or strengthening the muscles or joints of the body by working against a counterforce, with or without measuring devices
- A63B21/06—User-manipulated weights
- A63B21/072—Dumb-bells, bar-bells or the like, e.g. weight discs having an integral peripheral handle
- A63B21/0726—Dumb bells, i.e. with a central bar to be held by a single hand, and with weights at the ends
-
- 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
-
- 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
Landscapes
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Physical Education & Sports Medicine (AREA)
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Orthopedic Medicine & Surgery (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a kind of identifications of dumbbell exercise and method of counting and its device.The hardware of device includes 3-axis acceleration sensor module, micro controller module, display module, buzzer module, power module, key module totally six modules.Whether using the variation characteristic of the prominent acceleration signal of difference method, judging automatically further according to the threshold value pre-set is once really dumbbell exercise signal, and the starting point moment of automatic identification dumbbell exercise and end point moment.Acceleration of gravity signal is isolated using FIR low-pass filtering, then subtracts acceleration of gravity signal from original signal, obtains acceleration of motion signal, then acceleration of motion signal is normalized, extracts 12 characteristic values.Finally dumbbell exercise type is identified using decision tree-KNN algorithm, as long as correctly recognizing an effective exercise, buzzer rings a sound just to be prompted, while corresponding to counting and the sum of dumbbell exercise type in liquid crystal screen display.
Description
Technical field
The invention belongs to sports fitness equipment scopes, are related to the data sampling and processing of dumbbell exercise acceleration signal
And the identification field of type of sports.
Background technique
Action recognition is a kind of important technology for carrying out human-computer interaction, and the technology of action recognition is broadly divided into two kinds: a kind of
It is the human action identification of view-based access control model sensor, one is the human action identifications based on wearable motion sensor.Before
Person acquires image by imaging sensor, and is analyzed and processed to image, thus identification maneuver type, Theory comparison at
Ripe, there are many applications, but processing data volume is big, and power consumption is high, and image acquiring sensor and processor volume are big, not portable;
The latter is then analyzed and processed again to data by acquiring the signals such as acceleration, angular speed in motion process, is identified
Type of action and posture.
Motion sensor is mostly based on to the identification of dumbbell exercise, during doing dumbbell exercise by acquiring user of service
The signals such as acceleration, angular speed identify dumbbell exercise according to the variation characteristic of acceleration or angular speed.Recent domestic pair
The research of intelligent Dumbbell is more and more, constantly develops towards human-computer interaction, multi-functional, inexpensive, miniaturization direction.
Summary of the invention
It is an object of the invention to: a kind of automatic testing method and device are provided for dumbbell exercise, acquisition is moved through in real time
Acceleration signal in journey judges automatically athletic posture, shows measurement result in time, and passes through buzzer and remind user of service's meter
It counts successfully.The identification of dumbbell exercise provided by the invention and method of counting count that accurate, installation cost is low, are very suitable to commonly use
Family daily exercise uses.
The technical solution adopted in the present invention is as follows:
1. the identification method of counting and its device of a kind of dumbbell exercise, the hardware of device include 3-axis acceleration sensor
Module, micro controller module, display module, buzzer module, power module, key module totally six modules, modules
Function are as follows:
(1) the 3-axis acceleration signal of sensor module continuous acquisition dumbbell, and data are transmitted to micro controller module;
(2) micro controller module handles collected 3-axis acceleration signal data, is analyzed, and identifies dumbbell exercise
Type;
(3) micro controller module is every successfully identify an effective exercise after, meter corresponding to the type of sports that will identify that
Numerical value and total count value add 1, and so that buzzer module is rung a sound and prompted;
(4) display module uses liquid crystal display, the count value and total count value of circulation five kinds of type of sports of display;
(5) power module is powered to whole device, real-time detection cell voltage and electricity, when brownout, not enough power supply
When, indicator light flashing alarm is prompted, and is then automatically shut power off, is protected battery, avoid overdischarge;
(6) key module includes power on/off key, restarts counting button, and power button and closing key are same press
Key, long-pressing shutdown key apparatus can just shut down under open state.
2. the identification of dumbbell exercise is comprised the steps of: with method of counting
Step 1, by the acceleration in 3-axis acceleration sensor module acquisition dumbbell three directions of x, y, z in three dimensions
Signal is spent, the acceleration information in three directions is obtained;
Step 2 pre-processes the acceleration information in three directions, using the variation of the prominent acceleration of difference method
Characteristic, whether be primary really dumbbell exercise signal, and automatic identification is mute if judging automatically further according to the threshold value pre-set
The starting point moment and end point moment of fluid motion;If acceleration change amount is less than threshold value, this acquisition signal is considered as
Noise interferences and filter out, do not deal with, jump to step 9;If acceleration change amount is more than or equal to threshold value, from
The dynamic starting point and end point for judging dumbbell exercise, subsequently into step 3;
Step 3 carries out FIR low-pass filtering treatment to the acceleration signal in three directions respectively, obtains the weight in three directions
Power acceleration signal;
The acceleration signal in three directions is subtracted corresponding acceleration of gravity signal respectively, is divided by step 4
From acceleration of motion signal later;
Step 5, using the largest motion acceleration value in three directions as normalization factor, respectively to the fortune in three directions
Dynamic acceleration signal carries out global normalization's processing;
Step 6, the moving acceleration data after being normalized respectively to three directions seek energy, maximum value, minimum value and
12 characteristic values are obtained in absolute value mean value;
Step 7, using K arest neighbors (k-Nearest Neighbor, be abbreviated as KNN) sorting algorithm and decision Tree algorithms
The method blended, by the characteristic data set of 12 characteristic values and the five kinds of dumbbell exercises kept in advance obtained in step 6
It is compared, identifies type of sports;Five kinds of dumbbell exercise types include sitting posture dumbbell curl, one arm dumbbell of bending row the boat, sitting posture
Dumbbell pushes away shoulder, upright dumbbell alternative curl, upright dumbbell front raise;If successful match jumps to step 8;If five kinds
All it fails to match for dumbbell exercise type, then jumps to step 9;
Step 8, count value and total count value corresponding to the type of sports by successful match add 1, jump to step 10;
Step 9, abandons this acceleration information, and five kinds of dumbbell exercise type counts values and total count value remain unchanged;
Step 10, return step one continue acquisition and analysis next time.
3. in step 2, using the real dumbbell exercise of difference method automatic identification and dumbbell exercise starting point and knot
Beam spot the steps include:
Step 2.1, the acceleration signal of three axis of x, y, z is indicated with (x, y, z), remembers the value s (t) of t moment accelerometer
=(x (t), y (t), z (t)) is the acceleration signal that dumbbell exercise generates, using calculus of differences formula meter shown in formula (1)
Calculate the acceleration difference value Δ s of t momentt;
Δst=| x (t)-x (t-1) |+| y (t)-y (t-1) |+| z (t)-z (t-1) | (1)
Step 2.2, the difference mean value of N number of acceleration sampled data after t moment is calculated according to formula (2)
Step 2.3, two threshold value Th of initial time point punctum at the end of are respectively setstart、Thend, wherein starting threshold
Value, which is greater than, terminates threshold value;
Step 2.4, successively more each moment tsAcceleration difference mean valueWith threshold value ThstartIfThen
Think that primary effective dumbbell exercise has begun, and remembers that start time point is ts, enter step 2.5;Otherwise it is assumed that not finding
Starting point jumps to step 2.7;
Step 2.5, successively compare tsEach moment t after momenteAcceleration difference mean valueWith threshold value ThendIfThen think that primary effective dumbbell exercise is over, and punctum is t at the end of notee, enter step 2.6;Otherwise,
Think not find end point, jumps to step 2.7;
Step 2.6, with start time point for tsPunctum is t at the end ofeBetween data as a dumbbell exercise period
Data, enter step 2.7;
Step 2.7, the automatic identification process of starting point and end point is completed.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of device.
Fig. 2 is the circuit diagram of the micro controller module in example.
Fig. 3 is the circuit diagram of the acceleration sensor module in example.
Fig. 4 is the circuit diagram of the buzzer module in example.
Fig. 5 is the circuit diagram of the display module in example.
Fig. 6 is the circuit diagram of the power module in example.
Fig. 7 is the circuit diagram of the key module in example.
Fig. 8 is the algorithm flow chart for carrying out dumbbell action recognition in example using decision tree-KNN.
Fig. 9 is the workflow schematic diagram that the device in example carries out movement identification and counts.
Specific embodiment
In order to which more clearly description is of the invention, a specific embodiment of the invention work is further retouched in detail below
It states.
1. the identification method of counting and its device of a kind of dumbbell exercise, the hardware of device include 3-axis acceleration sensor
Module, micro controller module, display module, buzzer module, power module, key module totally six modules, as shown in Figure 1,
The function of each module are as follows:
(1) ADXL345 3-axis acceleration sensor is used, with the 3-axis acceleration of the frequency continuous acquisition dumbbell of 400Hz
Signal, and data are transmitted to micro controller module by IIC, micro controller module uses the chip of ARM, and concrete model is
The circuit connection principle of STM32F103C8T6, microcontroller and acceleration transducer is as shown in Figures 2 and 3;
(2) micro controller module handles collected 3-axis acceleration signal data, is analyzed, and identifies dumbbell exercise
Type;
(3) micro controller module is every successfully identify an effective exercise after, meter corresponding to the type of sports that will identify that
Numerical value and total count value add 1, and so that buzzer module is rung a sound and prompted, and 5V active buzzer, buzzing are used in example
The circuit diagram of device module is as shown in Figure 4;
(4) display module uses LCD1602 liquid crystal display, the count value and tale of circulation five kinds of type of sports of display
Value, the circuit diagram of display module are as shown in Figure 5;
(5) power module uses rechargeable 3.7V lithium battery, and power module is responsible for powering to whole device, real
When detection cell voltage and electricity, when brownout, not enough power supply, indicator light flashing alarm is prompted, then automatic to close
Power supply is closed, battery is protected, avoids overdischarge, the circuit diagram of power module is as shown in Figure 6;
(6) key module includes power on/off key, restarts counting button, and power button and closing key are same press
Key, long-pressing shutdown key apparatus can just shut down under open state, and the circuit diagram of key module is as shown in Figure 7.
2. the identification of dumbbell exercise is comprised the steps of: with method of counting
Step 1, by the acceleration in 3-axis acceleration sensor module acquisition dumbbell three directions of x, y, z in three dimensions
Signal is spent, the acceleration information in three directions is obtained;
Step 2 pre-processes the acceleration information in three directions, using the variation of the prominent acceleration of difference method
Characteristic, whether be primary really dumbbell exercise signal, and automatic identification is mute if judging automatically further according to the threshold value pre-set
The starting point moment and end point moment of fluid motion;If acceleration change amount is less than threshold value, this acquisition signal is considered as
Noise interferences and filter out, do not deal with, jump to step 9;If acceleration change amount is more than or equal to threshold value, from
The dynamic starting point and end point for judging dumbbell exercise, subsequently into step 3;
Step 3 carries out FIR low-pass filtering treatment to the acceleration signal in three directions respectively, obtains the weight in three directions
Power acceleration signal;
The acceleration signal in three directions is subtracted corresponding acceleration of gravity signal respectively, is divided by step 4
From acceleration of motion signal later;
Step 5, using the largest motion acceleration value in three directions as normalization factor, respectively to the fortune in three directions
Dynamic acceleration signal carries out global normalization's processing;
Step 6, the moving acceleration data after being normalized respectively to three directions seek energy, maximum value, minimum value and
12 characteristic values are obtained in absolute value mean value;
Step 7, using K arest neighbors (k-Nearest Neighbor, be abbreviated as KNN) sorting algorithm and decision Tree algorithms
The method blended, by the characteristic data set of 12 characteristic values and the five kinds of dumbbell exercises kept in advance obtained in step 6
It is compared, identifies type of sports;Five kinds of dumbbell exercise types include sitting posture dumbbell curl, one arm dumbbell of bending row the boat, sitting posture
Dumbbell pushes away shoulder, upright dumbbell alternative curl, upright dumbbell front raise;If successful match jumps to step 8;If five kinds
All it fails to match for dumbbell exercise type, then jumps to step 9;The flow chart of decision tree-KNN algorithm is as shown in Figure 8.
Step 8, count value and total count value corresponding to the type of sports by successful match add 1, jump to step 10;
Step 9, abandons this acceleration information, and five kinds of dumbbell exercise type counts values and total count value remain unchanged;
Step 10, return step one continue acquisition and analysis next time.
Whole device carries out movement identification and the workflow of counting is as shown in Figure 9.
3. in step 2, using difference method come the real dumbbell exercise of automatic identification and dumbbell exercise starting point and
End point the steps include:
Step 2.1, the acceleration signal of three axis of x, y, z is indicated with (x, y, z), remembers the value s (t) of t moment accelerometer
=(x (t), y (t), z (t)) is the acceleration signal that dumbbell exercise generates, using calculus of differences formula meter shown in formula (1)
Calculate the acceleration difference value Δ s of t momentt;
Δst=| x (t)-x (t-1) |+| y (t)-y (t-1) |+| z (t)-z (t-1) | (1)
Step 2.2, the difference mean value of N number of acceleration sampled data after t moment is calculated according to formula (2)
Step 2.3, two threshold value Th of initial time point punctum at the end of are respectively setstart、Thend, wherein starting threshold
Value, which is greater than, terminates threshold value;
Step 2.4, successively more each moment tsAcceleration difference mean valueWith threshold value ThstartIfThen
Think that primary effective dumbbell exercise has begun, and remembers that start time point is ts, enter step 2.5;Otherwise it is assumed that not finding
Starting point jumps to step 2.7;
Step 2.5, successively compare tsEach moment t after momenteAcceleration difference mean valueWith threshold value ThendIfThen think that primary effective dumbbell exercise is over, and punctum is t at the end of notee, enter step 2.6;Otherwise,
Think not find end point, jumps to step 2.7;
Step 2.6, with start time point for tsPunctum is t at the end ofeBetween data as a dumbbell exercise period
Data, enter step 2.7;
Step 2.7, the automatic identification process of starting point and end point is completed.
4. in step 3, the example design FIR low pass filter of one 25 rank, to the effective action signal recognized
It is filtered denoising;If h [26] is low-pass filtering coefficient array, M is the data length in an effective exercise period, x
[M], y [M], z [M] are respectively the x being truncated to, and y, z 3-axis acceleration array, then FIR low-pass filtering algorithm is as follows:
Input: x [M], y [M], z [M]
Step 3.1:i=0;
Step 3.2:x [i]=h [0] * x [i]+h [1] * x [i-1]+...+h [25] * x [i-25],
Y [i]=h [0] * y [i]+h [1] * y [i-1]+...+h [25] * y [i-25],
Z [i]=h [0] * z [i]+h [1] * z [i-1]+...+h [25] * z [i-25];
Step 3.3:i++;
Step 3.4: if i < M goes to step 3.2, otherwise transferring out;
Output: x [M], y [M], z [M].
5. in step 7, the energy of moving acceleration data, maximum value, the calculation formula of minimum value and absolute value mean value
Are as follows:
1) signal energy are as follows:
2) signal maximum and minimum value are respectively (wherein k=x, y, z):
3) the absolute value mean value of signal is (wherein L is signal length):
Claims (2)
1. the identification method of counting and its device of a kind of dumbbell exercise, which is characterized in that the hardware of device includes 3-axis acceleration
Sensor module, micro controller module, display module, buzzer module, power module, key module totally six modules, dress
It sets and is characterized in that:
(1) the 3-axis acceleration signal of sensor module continuous acquisition dumbbell, and data are transmitted to micro controller module;
(2) micro controller module handles collected 3-axis acceleration signal data, is analyzed, and identifies dumbbell exercise class
Type;
(3) micro controller module is every successfully identify an effective exercise after, count value corresponding to the type of sports that will identify that
And total count value adds 1, and so that buzzer module is rung a sound and prompted;
(4) display module uses liquid crystal display, the count value and total count value of circulation five kinds of type of sports of display;
(5) power module is powered to whole device, and real-time detection cell voltage and electricity refer to when brownout, not enough power supply
Show that lamp flashing alarm is prompted, then automatically shuts power off, protect battery, avoid overdischarge;
(6) key module includes power on/off key, restarts counting button, and power button and closing key are same key, are opened
Long-pressing shutdown key apparatus can just shut down under machine state;
The identification of dumbbell exercise is comprised the steps of: with method of counting
Step 1, by 3-axis acceleration sensor module acquisition dumbbell in three dimensions three directions of x, y, z acceleration believe
Number, obtain the acceleration information in three directions;
Step 2 pre-processes the acceleration information in three directions, and the variation characteristic of acceleration is protruded using difference method,
Whether judge automatically further according to the threshold value pre-set is primary really dumbbell exercise signal, and automatic identification dumbbell exercise
Starting point moment and terminate point moment;If acceleration change amount is less than threshold value, this acquisition signal is considered as noise and is done
It disturbs signal and filters out, do not deal with, jump to step 9;If acceleration change amount is more than or equal to threshold value, judge automatically
The starting point and end point of dumbbell exercise, subsequently into step 3;
Step 3 carries out FIR low-pass filtering treatment to the acceleration signal in three directions respectively, and the gravity for obtaining three directions adds
Speed signal;
The acceleration signal in three directions is subtracted corresponding acceleration of gravity signal respectively, was separated by step 4
Acceleration of motion signal afterwards;
Step 5 respectively adds the movement in three directions using the largest motion acceleration value in three directions as normalization factor
Speed signal carries out global normalization's processing;
Step 6, the moving acceleration data after normalizing respectively to three directions seek energy, maximum value, minimum value and absolutely
It is worth mean value, 12 characteristic values is obtained;
Step 7 is mutually melted using K arest neighbors (k-Nearest Neighbor, be abbreviated as KNN) sorting algorithm with decision Tree algorithms
The method of conjunction carries out 12 characteristic values obtained in step 6 and the characteristic data set for the five kinds of dumbbell exercises kept in advance
It compares, identifies type of sports;Five kinds of dumbbell exercise types include sitting posture dumbbell curl, one arm dumbbell of bending row the boat, sitting posture dumbbell
Push away shoulder, upright dumbbell alternative curl, upright dumbbell front raise;If successful match jumps to step 8;If five kinds of dumbbells
All it fails to match for type of sports, then jumps to step 9;
Step 8, count value and total count value corresponding to the type of sports by successful match add 1, jump to step 10;
Step 9, abandons this acceleration information, and five kinds of dumbbell exercise type counts values and total count value remain unchanged;
Step 10, return step one continue acquisition and analysis next time.
2. the identification method of counting and its device of a kind of dumbbell exercise according to claim 1, which is characterized in that the step
It is used to the difference method of the real dumbbell exercise of automatic identification and dumbbell exercise starting point and end point, step in rapid two
Are as follows:
Step 2.1, the acceleration signal of three axis of x, y, z is indicated with (x, y, z), remembers value s (t)=(x of t moment accelerometer
(t), y (t), z (t)) it is the acceleration signal that dumbbell exercise generates, when calculating t using calculus of differences formula shown in formula (1)
The acceleration difference value Δ s at quartert;
Δst=| x (t)-x (t-1) |+| y (t)-y (t-1) |+| z (t)-z (t-1) | (1)
Step 2.2, the difference mean value of N number of acceleration sampled data after t moment is calculated according to formula (2)
Step 2.3, two threshold value Th of initial time point punctum at the end of are respectively setstart、Thend, wherein it is big to start threshold value
In end threshold value;
Step 2.4, successively more each moment tsAcceleration difference mean valueWith threshold value ThstartIfThen think
Primary effective dumbbell exercise has begun, and remembers that start time point is ts, enter step 2.5;Otherwise it is assumed that not finding starting
Point, jumps to step 2.7;
Step 2.5, successively compare tsEach moment t after momenteAcceleration difference mean valueWith threshold value ThendIfThen think that primary effective dumbbell exercise is over, and punctum is t at the end of notee, enter step 2.6;Otherwise,
Think not find end point, jumps to step 2.7;
Step 2.6, with start time point for tsPunctum is t at the end ofeBetween number of the data as a dumbbell exercise period
According to entering step 2.7;
Step 2.7, the automatic identification process of starting point and end point is completed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811426205.0A CN109603142A (en) | 2018-11-27 | 2018-11-27 | A kind of the identification method of counting and its device of dumbbell exercise |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811426205.0A CN109603142A (en) | 2018-11-27 | 2018-11-27 | A kind of the identification method of counting and its device of dumbbell exercise |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109603142A true CN109603142A (en) | 2019-04-12 |
Family
ID=66004802
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811426205.0A Pending CN109603142A (en) | 2018-11-27 | 2018-11-27 | A kind of the identification method of counting and its device of dumbbell exercise |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109603142A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111780780A (en) * | 2020-06-16 | 2020-10-16 | 贵州省人民医院 | Step counting method and device based on filter bank |
CN112337043A (en) * | 2020-10-20 | 2021-02-09 | 随机漫步(上海)体育科技有限公司 | Method and device for motion recognition, method and device for motion evaluation and dumbbell |
CN114265656A (en) * | 2021-12-24 | 2022-04-01 | 四川千里倍益康医疗科技股份有限公司 | Fascia gun and display control method of display screen of fascia gun |
CN114534236A (en) * | 2022-03-24 | 2022-05-27 | 山西新和实业有限公司 | Intelligent dumbbell motion counting method based on motion sensor |
WO2023178594A1 (en) * | 2022-03-24 | 2023-09-28 | 广东高驰运动科技股份有限公司 | Action counting method and apparatus, device, and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2281612A1 (en) * | 2009-08-04 | 2011-02-09 | IDT Technology Limited | Apparatus and method for counting exercise repetitions |
CN103933722A (en) * | 2014-02-28 | 2014-07-23 | 永康市坤卓科技有限公司 | Bodybuilding dumbbell exercise detection device and method |
CN105588577A (en) * | 2014-10-23 | 2016-05-18 | 中国移动通信集团公司 | Detection method and detection apparatus for abnormal step counting in exercise monitoring device |
WO2016097746A1 (en) * | 2014-12-19 | 2016-06-23 | Mclaren Applied Technologies Limited | Biomechanical analysis |
CN205988512U (en) * | 2016-08-25 | 2017-03-01 | 南通怡达运动用品有限公司 | There is the dumbbell of action recognition function |
CN107233696A (en) * | 2017-06-27 | 2017-10-10 | 东北林业大学 | A kind of multifunctional dumbbell |
-
2018
- 2018-11-27 CN CN201811426205.0A patent/CN109603142A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2281612A1 (en) * | 2009-08-04 | 2011-02-09 | IDT Technology Limited | Apparatus and method for counting exercise repetitions |
CN103933722A (en) * | 2014-02-28 | 2014-07-23 | 永康市坤卓科技有限公司 | Bodybuilding dumbbell exercise detection device and method |
CN105588577A (en) * | 2014-10-23 | 2016-05-18 | 中国移动通信集团公司 | Detection method and detection apparatus for abnormal step counting in exercise monitoring device |
WO2016097746A1 (en) * | 2014-12-19 | 2016-06-23 | Mclaren Applied Technologies Limited | Biomechanical analysis |
CN205988512U (en) * | 2016-08-25 | 2017-03-01 | 南通怡达运动用品有限公司 | There is the dumbbell of action recognition function |
CN107233696A (en) * | 2017-06-27 | 2017-10-10 | 东北林业大学 | A kind of multifunctional dumbbell |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111780780A (en) * | 2020-06-16 | 2020-10-16 | 贵州省人民医院 | Step counting method and device based on filter bank |
CN112337043A (en) * | 2020-10-20 | 2021-02-09 | 随机漫步(上海)体育科技有限公司 | Method and device for motion recognition, method and device for motion evaluation and dumbbell |
CN114265656A (en) * | 2021-12-24 | 2022-04-01 | 四川千里倍益康医疗科技股份有限公司 | Fascia gun and display control method of display screen of fascia gun |
CN114534236A (en) * | 2022-03-24 | 2022-05-27 | 山西新和实业有限公司 | Intelligent dumbbell motion counting method based on motion sensor |
WO2023178594A1 (en) * | 2022-03-24 | 2023-09-28 | 广东高驰运动科技股份有限公司 | Action counting method and apparatus, device, and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109603142A (en) | A kind of the identification method of counting and its device of dumbbell exercise | |
CN106237604A (en) | Wearable device and the method utilizing its monitoring kinestate | |
KR102252269B1 (en) | Swimming analysis system and method | |
CN104135911B (en) | Activity classification in multi-axial cord movement monitoring device | |
CN105184325B (en) | Mobile intelligent terminal | |
CN104297519B (en) | Human motion gesture recognition method and mobile terminal | |
CN102302370A (en) | Method and device for detecting tumbling | |
CN107469326B (en) | Swimming monitoring method and device for wearable equipment and wearable equipment | |
Jensen et al. | Classification of kinematic swimming data with emphasis on resource consumption | |
CN102640085A (en) | System and method for recognizing gestures | |
CN108958482B (en) | Similarity action recognition device and method based on convolutional neural network | |
CN107105092A (en) | A kind of human body tumble recognition methods based on dynamic time warping | |
CN112587901B (en) | Swimming gesture recognition method, device, system and storage medium | |
CN110648478B (en) | Drowning detection system and method based on wireless communication | |
CN104586402A (en) | Feature extracting method for body activities | |
Ning et al. | Fall detection algorithm based on gradient boosting decision tree | |
Qu et al. | Convolutional neural network for human behavior recognition based on smart bracelet | |
CN105105757A (en) | Wearable human motion gesture track recording and assessment device | |
Fu et al. | Ping pong motion recognition based on smart watch | |
CN108720839B (en) | Human motion evaluation method and device based on motion energy collection | |
CN111803902A (en) | Swimming stroke identification method and device, wearable device and storage medium | |
CN109382827A (en) | A kind of robot system and its intelligent memory recognition methods | |
Wang et al. | Human motion phase segmentation based on three new features | |
CN110322662B (en) | Posture monitoring and falling alarm bracelet based on physiological and behavioral characteristics | |
Pipanmaekaporn et al. | Mining Acceleration Data for Smartphone-based Fall Detection |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190412 |
|
WD01 | Invention patent application deemed withdrawn after publication |