WO2017156834A1 - 测量卡路里消耗和监测手部姿势识别的智慧手套及估算的方法和系统 - Google Patents

测量卡路里消耗和监测手部姿势识别的智慧手套及估算的方法和系统 Download PDF

Info

Publication number
WO2017156834A1
WO2017156834A1 PCT/CN2016/080734 CN2016080734W WO2017156834A1 WO 2017156834 A1 WO2017156834 A1 WO 2017156834A1 CN 2016080734 W CN2016080734 W CN 2016080734W WO 2017156834 A1 WO2017156834 A1 WO 2017156834A1
Authority
WO
WIPO (PCT)
Prior art keywords
module
data
glove
calorie consumption
motion
Prior art date
Application number
PCT/CN2016/080734
Other languages
English (en)
French (fr)
Inventor
伍楷舜
邹永攀
王丹
吴金咏
Original Assignee
深圳大学
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 深圳大学 filed Critical 深圳大学
Publication of WO2017156834A1 publication Critical patent/WO2017156834A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A41—WEARING APPAREL
    • A41D—OUTERWEAR; PROTECTIVE GARMENTS; ACCESSORIES
    • A41D19/00—Gloves
    • 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
    • A—HUMAN NECESSITIES
    • A41—WEARING APPAREL
    • A41D—OUTERWEAR; PROTECTIVE GARMENTS; ACCESSORIES
    • A41D19/00—Gloves
    • A41D19/0024—Gloves with accessories
    • A41D19/0027—Measuring instruments, e.g. watch, thermometer
    • 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
    • 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
    • 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
    • A63B2220/50—Force related parameters
    • A63B2220/51—Force
    • 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
    • A63B2220/80—Special sensors, transducers or devices therefor
    • A63B2220/83—Special sensors, transducers or devices therefor characterised by the position of the sensor
    • A63B2220/833—Sensors arranged on the exercise apparatus or sports implement
    • 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
    • A63B2230/75—Measuring physiological parameters of the user calorie expenditure

Definitions

  • the invention relates to an intelligent monitoring and improving device for fitness actions, in particular to a method and system for calorie consumption and hand posture recognition.
  • the human body In the daily fitness process, especially during strength training, the human body will consume a large amount of calories. In order to help the nutritional supplement and diet after training, do not accurately estimate the calorie consumption of the human body during training; at the same time, correct The posture is a necessary condition for any training program. This is because the correct posture helps to alleviate or even eliminate any potential training injuries and helps the trainer's health.
  • the present invention provides a system for calorie consumption and hand gesture recognition, which solves the problem of cumbersome energy consumption of the human body during the exercise process and the addition of additional equipment in the prior art.
  • a smart glove for measuring calorie consumption and monitoring hand gesture recognition comprising: a processor module, a storage module, a communication module, a sensing module, an early warning prompt module, a display module, a power module, Switch module and cloud server, the storage module, communication
  • the module, the sensing module, the warning prompt module, the display module, the power module, the switch module and the cloud server are respectively connected to the processor module, wherein the sensing module comprises a pressure sensor and an IMU inertia unit, the pressure sensor and the IMU inertia unit Respectively responsible for sensing the amount of stress and IMU data of the hand during the user's fitness process, and collecting source data for user strength training, and transmitting the source data to the processor module; the processor module is responsible for the source Pre-processing of the data; the storage module is responsible for storing the result of the processor pre-processing and the processing result returned from the cloud server; the communication module is responsible for transmitting the source data and the pre-processed data to
  • the invention also provides a method for estimating calorie consumption and monitoring hand gesture recognition, using the smart glove for measuring calorie consumption and monitoring hand gesture recognition, and comprising the following steps: step S1, using a membrane pressure sensor and The IMU inertia unit collects the original data during the strength training; in step S2, the original data is transmitted to the processor module for processing; and in step S3, the received pressure sensor and the IMU inertial unit data are subjected to ETL analysis, and the ETL process is passed.
  • the ETL technology extracts, transposes, loads, and delivers the data obtained in step S2; and, in step S4, estimates the calories burned, establishes a classification model through the support vector machine, and uses the source data beyond the strength training as the target action. Class, the other as a non-target action class; then determine whether the threshold of strength training is exceeded, and if so, a reminder signal is sent, otherwise the data is transmitted to the mobile device for recording, and the dietary recipe, nutritional mix, and training plan are recommended.
  • step S1 the user opens the entire system through a button or a switch, and one or more pressure sensors are disposed on the sensing module of the smart glove.
  • step S2 As a further improvement of the present invention, the following steps are included in step S2:
  • Step S21 collecting pressure data and motion posture data through the pressure sensor and the IMU inertia unit, and performing sliding average filtering on the obtained acceleration sensor, the magnetic sensor, and the gyroscope data to eliminate background noise based on the wearable computing technology, the sliding
  • the sliding window width of the averaging filter is adjusted according to the actual situation of the user (for example, the sliding window width of the sliding averaging filter is 7);
  • Step S22 an accurate estimation of the glove attitude is implemented by using a complementary filter for reading the acceleration sensor and the gyroscope, and the complementary filter is used to combine the estimated value of the attitude angle of the acceleration sensor with the attitude of the glove and the estimated attitude of the gyroscope to the attitude angle of the glove;
  • Step S23 after estimating the attitude angle of the glove during the movement by using a complementary filter, the projection component of the gravity in the moving direction of the smart glove is removed according to the mechanical principle, and the acceleration value generated by the movement of the smart glove itself is extracted. Achieve acceleration calibration;
  • Step S24 obtaining the calibrated acceleration data, and obtaining the speed value of the wisdom glove movement by using one integral method
  • Step S25 after integrating the speed value of the wisdom glove movement, calibrating the speed of the wisdom glove movement according to the movement feature of the wisdom glove starting from the stationary state and ending in the stationary state;
  • step S26 the obtained speed value is integrated once, thereby obtaining the displacement magnitude of the wisdom glove movement; in step S27, combined with the above result, the trajectory of the fitness movement is reconstructed, thereby obtaining the movement trajectory of the hand during the user's fitness.
  • step S4 comprises the following steps:
  • Step S41A combined with the displacement value obtained in step S2, combined with the reading of the film pressure sensor, through the processing of step S3, calculates the work done by the glove movement, and calculates the calorie consumption according to the relationship model between the established work and the calorie consumption;
  • Step S42B recording the calorie consumption, and reminding the meal matching and nutritional balance
  • Step S42A Pre-establishing a high-dimensional feature model based on a statistical learning theory, wherein the high-dimensional feature model uses an abnormal pattern of information changes due to various posture deviation actions in the set space as a training sample;
  • Step S41B mapping the abnormal mode to a high-dimensional feature model of a type of support vector machine, separating the target action class, and determining whether the motion posture deviates, and if so, issuing a warning signal.
  • the present invention again provides a system for estimating calorie consumption and monitoring hand gesture recognition, employing the method of estimating calorie consumption and monitoring hand gesture recognition, and including the following modules:
  • a signal acquisition and calculation module for collecting signals of the movement state of the smart glove and evaluating the motion state information for preliminary calculation
  • An abnormality detecting module configured to identify, by using an anomaly detection algorithm, whether the signal is abnormal
  • the action judging module is used for classifying the target action class and other action classes to distinguish the abnormal mode caused by the strength training threshold as the target action class, and determining whether the posture of the smart glove is deviated;
  • an alarm module configured to issue an alert signal when it is determined that the posture deviation occurs.
  • the signal acquisition and calculation module includes the following units:
  • Inductive acquisition unit for turning on the entire system and collecting motion data.
  • the collected motion data includes strength training.
  • the motion data in the direction and the three-dimensional coordinates in all directions;
  • a data processing unit configured to average the force and motion information in each direction, and use the average value as the motion state information
  • the smoothing unit performs ETL analysis on the data obtained by the data processing unit, and smoothes the motion state information by a moving average method.
  • the abnormality detecting module includes the following units:
  • An abnormality calculating unit configured to perform data segmentation on the time series of the motion state information to obtain a subsequence, and calculate local abnormal data of the subsequence;
  • the abnormal output unit is configured to output the sub-sequence as an abnormal mode when the abnormal data is greater than or equal to a preset threshold.
  • the action judging module includes the following units:
  • a model unit for establishing a high-dimensional feature model based on a statistical learning theory preset wherein the high-dimensional feature model uses an abnormal pattern in a set space due to changes in motion state information due to various human actions as a training sample;
  • the motion recognition unit is configured to map the abnormal pattern output by the abnormal output unit to a high-dimensional feature model of a type of support vector machine, and separate the target action class.
  • a feedback module is further included for feeding back response information for the hand posture warning signal, and adjusting a high-dimensional feature model of a type of support vector machine.
  • the beneficial effects of the invention in the process of strength training, the detection accuracy of the detected action is 84%-94%, the false positive rate is low, and the measurement accuracy of calories is as high as 90%, which can realize the deviation of the posture of the opponent.
  • the alarm signal is sent out, and the self-learning function of the system is used to deal with the false alarm situation, thereby further reducing the false alarm rate; the invention performs the calorie measurement and the hand posture recognition work based on the existing sensor technology, and adopts efficient data. Processing methods and algorithm models can be widely applied in strength training, providing a good reference for fitness enthusiasts, and can provide post-meal matching and nutritional balance recommendations.
  • FIG. 1 is a block diagram of a smart glove according to an embodiment of the present invention.
  • Figure 2 is a schematic diagram of the operation of another embodiment of the present invention.
  • FIG. 3 is a schematic diagram of a workflow of another embodiment of the present invention.
  • FIG. 4 is a block diagram of functional modules of still another embodiment of the present invention.
  • a smart glove for measuring calorie consumption and monitoring hand gesture recognition comprising: a processor module, a storage module, a communication module, a sensing module, an early warning prompt module, a display module, a power module, a switch module and a cloud server, wherein the storage module
  • the communication module, the sensing module, the warning prompt module, the display module, the power module, the switch module and the cloud server are respectively connected to the processor module
  • the sensing module comprises a pressure sensor and an IMU inertia unit, the pressure sensor and the IMU
  • the inertia unit is respectively responsible for sensing the amount of pressure of the hand and the IMU data during the fitness process of the user, and collecting the source data for the user strength training, and transmitting the source data to the processor module
  • the processor module is responsible for Pre-processing of the source data
  • the storage module is responsible for storing the result of the processor pre-processing and the processing result returned from the cloud server
  • the communication module is responsible for transmitting the source data and the pre-processe
  • the warning prompt module is responsible for the abnormality of the fitness (such as when the user action is not standard, the exercise is excessive or the force is improper), the user is alerted; the display module is responsible for the user's fitness process The results of the type of action, calories burned, and posture standard are displayed.
  • the invention also provides a method for estimating calorie consumption and monitoring hand posture recognition, adopting the smart glove for measuring calorie consumption and monitoring hand posture recognition, and comprising the following steps: step S1, using a film of a pressure sensor
  • the pressure sensor and the IMU inertia unit collect source data during strength training; in step S2, the source data is transmitted to the processor module for processing; and in step S3, the received pressure sensor and the data acquired by the IMU inertia unit are subjected to an ETL process, Performing an ETL process, which is a process of extracting, transposing, loading, and delivering the data obtained by step S2 by the ETL technology; and, in step S4, calculating the calories burned, and establishing a classification model by using the support vector machine,
  • the source data beyond the strength training is taken as the target action class, and the other is used as the non-target action class; then it is judged whether the threshold of the strength training is exceeded, and if so, a reminder signal is issued, if otherwise, the data is transmitted
  • step S1 the user opens the entire system through a button or a switch, and one or more pressure sensors are disposed on the sensing module of the smart glove.
  • step S2 The following steps are included in step S2:
  • Step S21 collecting pressure data and motion posture data through the pressure sensor and the IMU inertia unit, and performing sliding average filtering on the obtained acceleration sensor, the magnetic sensor, and the gyroscope data to eliminate background noise based on the wearable computing technology, the sliding
  • the sliding window width of the averaging filter is 7;
  • Step S22 an accurate estimation of the glove attitude is implemented by using a complementary filter for reading the acceleration sensor and the gyroscope, and the complementary filter is used to combine the estimated value of the attitude angle of the acceleration sensor with the attitude of the glove and the estimated attitude of the gyroscope to the attitude angle of the glove;
  • Step S23 after estimating the attitude angle of the glove during the movement by using a complementary filter, the projection component of the gravity in the moving direction of the smart glove is removed according to the mechanical principle, and the acceleration value generated by the movement of the smart glove itself is extracted. Achieve acceleration calibration;
  • Step S24 obtaining the calibrated acceleration data, and obtaining the speed value of the wisdom glove movement by using one integral method
  • Step S25 after integrating the speed value of the wisdom glove movement, calibrating the speed of the wisdom glove movement according to the movement feature of the wisdom glove starting from the stationary state and ending in the stationary state;
  • step S26 the obtained speed value is integrated once, thereby obtaining the displacement magnitude of the wisdom glove movement; in step S27, combined with the above result, the trajectory of the fitness movement is reconstructed, thereby obtaining the movement trajectory of the hand during the user's fitness.
  • the step S4 includes the following steps:
  • Step S41A combined with the displacement value obtained in step S2, combined with the reading of the film pressure sensor, through the processing of step S3, calculates the work done by the glove movement, and calculates the calorie consumption according to the relationship model between the established work and the calorie consumption;
  • Step S42B recording the calorie consumption, and reminding the meal matching and nutritional balance
  • Step S42A Pre-establishing a high-dimensional feature model based on a statistical learning theory, wherein the high-dimensional feature model uses an abnormal pattern of information changes due to various posture deviation actions in the set space as a training sample;
  • Step S41B mapping the abnormal mode to a high-dimensional feature model of a type of support vector machine, separating the target action class, and determining whether the motion posture deviates, and if so, issuing a warning signal.
  • the present invention again provides a system for estimating calorie consumption and monitoring hand gesture recognition, employing the method of estimating calorie consumption and monitoring hand gesture recognition, and including the following modules:
  • a signal acquisition and calculation module for collecting signals of the movement state of the smart glove and evaluating the motion state information for preliminary calculation
  • An abnormality detecting module configured to identify, by using an anomaly detection algorithm, whether the signal is abnormal
  • the action judging module is used for classifying the target action class and other action classes to distinguish the abnormal mode caused by the strength training threshold as the target action class, and determining whether the posture of the smart glove is deviated;
  • an alarm module configured to issue an alert signal when it is determined that the posture deviation occurs.
  • the signal acquisition and calculation module includes the following units:
  • An inductive acquisition unit for turning on the entire system and collecting motion data, and the collected motion data includes motion data in the direction of strength training and motion data in all directions of three-dimensional coordinates;
  • a data processing unit configured to average the force and motion information in each direction, and use the average value as the motion state information
  • the smoothing unit performs ETL analysis on the data obtained by the data processing unit, and smoothes the motion state information by a moving average method.
  • the abnormality detecting module includes the following units:
  • An abnormality calculating unit configured to perform data segmentation on the time series of the motion state information to obtain a subsequence, and calculate local abnormal data of the subsequence;
  • the abnormal output unit is configured to output the sub-sequence as an abnormal mode when the abnormal data is greater than or equal to a preset threshold.
  • the action judging module includes the following units:
  • a model unit for establishing a high-dimensional feature model based on a statistical learning theory preset wherein the high-dimensional feature model uses an abnormal pattern in a set space due to changes in motion state information due to various human actions as a training sample;
  • the motion recognition unit is configured to map the abnormal pattern output by the abnormal output unit to a high-dimensional feature model of a type of support vector machine, and separate the target action class.
  • a feedback module is further included for feeding back response information for the hand posture warning signal, and adjusting a high-dimensional feature model of a type of support vector machine.
  • the present example provides a smart glove for measuring calorie consumption and monitoring hand gesture recognition, including: a processor module, a storage module, a communication module, a sensing module, an early warning prompt module, a display module, a power module, and a switch.
  • the module and the cloud server, the storage module, the communication module, the sensing module, the warning prompt module, the display module, the power module, the switch module and the cloud server are respectively connected to the processor module, wherein the sensing module comprises a pressure sensor and an IMU
  • the sensing module comprises a pressure sensor and an IMU
  • the inertia unit, the pressure sensor and the IMU inertia unit are respectively responsible for sensing the amount of pressure and IMU data of the hand during the fitness process of the user, and collecting the source data for the user strength training, and transmitting the source data to the processor.
  • the processor module is responsible for preprocessing the source data
  • the storage module is responsible for storing the result of the processor preprocessing And the processing result returned from the cloud server
  • the communication module is responsible for transmitting the source data and the preprocessed data to the cloud server for further data processing and analysis by the cloud server, and is responsible for transmitting the result of the cloud server processing back.
  • the warning prompt module is responsible for alerting the user when the user moves in a non-standard, excessive exercise or stress situation
  • the display module is responsible for the type of action and consumption of the user during the fitness process The results of calorie and posture standardization are shown.
  • the smart glove described in this example may also be referred to as a glove for short;
  • the sensing module of the smart glove is provided with one or two or more pressure sensors, the pressure sensor is disposed at the hand grasping force, and the IMU inertia unit is disposed at
  • the wrist is convenient for the user to use;
  • the storage module includes a built-in large-capacity memory and an external memory interface, which is convenient for storing and calling the user's motion data;
  • the source data is respectively sensing the user's fitness process through the pressure sensor and the IMU inertia unit.
  • the hand in the middle is subjected to the amount of pressure and the collected data corresponding to the IMU data.
  • all the postures are first made in accordance with the standard posture.
  • the standard posture can be guided by the instructional video or the coach, and then the user's source data and the motion track of each movement are extracted through the sensing module. And save as a reference; then, when the user performs the fitness exercise, the user's fitness hand movement track is compared with the reference trajectory at this time, for example, by comparing the similarity comparison manners, the user's posture during exercise can be determined.
  • the standard is the degree of similarity with the reference trajectory; when the difference between the user's calorie consumption and the hand posture and the reference data reaches a certain range, that is, when the preset threshold is exceeded
  • the warning prompt module emits an alarm prompt such as sound, light or vibration; the threshold is preset according to user requirements, for example, can be set to 5.
  • this example also provides a method for estimating calorie consumption and monitoring hand gesture recognition, which uses the wisdom glove described in Embodiment 1, and includes the following steps:
  • Step S1 using the membrane pressure sensor of the pressure sensor and the IMU inertia unit to collect source data during strength training;
  • Step S2 the source data is transmitted to the processor module for processing
  • Step S3 performing the ETL process by performing the ETL process on the data obtained by the received pressure sensor and the IMU inertia unit, wherein the ETL process extracts, transposes, loads, and delivers the data obtained in step S2 by the ETL technology. process;
  • step S4 calculating the calories burned, establishing a classification model by using the support vector machine, using the source data exceeding the strength training as the target action class, and the other as the non-target action class; and then determining whether it is super The threshold for strength training, if yes, a reminder signal, if otherwise, the data is transmitted to the mobile device for recording, and the dietary recipe, nutritional mix and training plan are recommended.
  • the TL process refers to the extract, transform and load of the data, that is, the process of extracting, transposing, loading and delivering the data, which refers to an important step of preprocessing the data before the data processing;
  • the mechanical work output by the human body during the movement can be linked to the calories consumed by the human body in the process through a conversion factor, and the work is calculated by combining the readings of the load cell and the reconstructed motion trajectory.
  • Step S4 transmits the data to the mobile device, records, recommends the meal recipe, nutrition mix and training plan, and the threshold can be customized according to the user's needs, such as when the barbell is used, by the preset force
  • the size, called the threshold signals if it exceeds this pre-set size.
  • the processor module receives the pressure sensor and the IMU inertia unit to measure the required motion data, and establishes the relationship between the motion signal and the strength training action, and only needs to use a simple sensor to pass the calorie of the detected person.
  • the recognition of the consumption and the hand posture determines whether the subject has a deviation of the hand posture and gives an alarm, which reduces the dependence on the cumbersome measuring device, and greatly improves the correct rate of the hand posture recognition; In the middle, the calorie consumption and meal recommendation can be displayed through the mobile device terminal.
  • the number of pressure sensors of the smart glove is one or two, and the number of central processors included in the processor module is one, as shown in FIG.
  • Two types of sensors pressure sensor and IMU inertia unit. These two types of sensors transmit the motion data to the central processor through complementary filtering, exponential averaging and ETL, and then according to the set calculation model and machine learning method. Identification of the posture.
  • step S1 the user preferably opens the entire system by means of a button or a switch, which may be a button and a switch implemented by a touch sensor, which may be provided in the IMU inertia unit accessory so that For user operations.
  • a button or a switch which may be a button and a switch implemented by a touch sensor, which may be provided in the IMU inertia unit accessory so that For user operations.
  • step S2 the abnormality detection algorithm is used to identify the abnormality of the acquired motion information is a time series abnormality detection algorithm based on the local anomaly factor, and the step S2 includes the following steps:
  • Step S21 collecting pressure data and motion posture data through the pressure sensor and the IMU inertia unit, and sliding the obtained data of the acceleration sensor, the magnetic sensor and the gyroscope based on the wearable computing technology Mean filtering to eliminate background noise, the sliding window width of the sliding mean filter is 7;
  • Step S22 an accurate estimation of the glove attitude is implemented by using a complementary filter for reading the acceleration sensor and the gyroscope, and the complementary filter is used to combine the estimated value of the attitude angle of the acceleration sensor with the attitude of the glove and the estimated attitude of the gyroscope to the attitude angle of the glove;
  • Step S23 after estimating the attitude angle of the glove during the movement by using a complementary filter, the projection component of the gravity in the moving direction of the smart glove is removed according to the mechanical principle, and the acceleration value generated by the movement of the smart glove itself is extracted. Achieve acceleration calibration;
  • Step S24 obtaining the calibrated acceleration data, and obtaining the speed value of the wisdom glove movement by using one integral method
  • Step S25 after integrating the speed value of the wisdom glove movement, calibrating the speed of the wisdom glove movement according to the movement feature of the wisdom glove starting from the stationary state and ending in the stationary state;
  • step S26 the obtained speed value is integrated once, thereby obtaining the displacement magnitude of the wisdom glove movement.
  • step S22 the analysis processing of the data is realized by the complementary filter, and the attitude of the object in space can be combined by the acceleration and the angular velocity to achieve accurate estimation. Therefore, the step S22 is combined with the acceleration sensor to measure the glove. An estimate of the attitude angle and an estimate of the attitude angle of the glove measured by the gyroscope, thereby obtaining an accurate estimate of the glove posture.
  • step S23 the acceleration value generated by the movement of the smart glove itself is extracted, that is, the acceleration component of the wisdom glove in three directions of the three-dimensional coordinates is filtered out, which can be realized by complementary filtering or Kalman filtering. .
  • step 24 for less speed drift after integration, the integral is only for the acceleration value corresponding to the glove movement.
  • the exponential mean value (EMA) is used as the judgment whether the glove acceleration value corresponds to the stationary state or the motion state of the glove;
  • the index mean value in the sliding window is greater than the threshold value, the acceleration in the window is determined to correspond to the motion state of the glove; otherwise, it is determined to correspond to the stationary state of the glove; when the judgment process is completed, the integral is only applied to the corresponding state.
  • the acceleration value of the glove in motion state, thereby obtaining the corresponding speed value, and the acceleration value corresponding to the glove static state is regarded as a zero value; the threshold value is combined with the sensor, the algorithm and the user requirement in the data processing process. It is determined comprehensively that this threshold will be different under different circumstances.
  • the movement state of the glove is judged by using the exponential average value (EMA) described in the step S24, and when it is determined to be stationary, the corresponding speed value is forcibly set to 0.
  • EMA exponential average value
  • the example also includes a high-dimensional feature model for adjusting a type of support vector machine for calorie consumption and hand gesture recognition, and the step S4 includes the following steps:
  • Step S41A combined with the displacement value obtained in step S2, combined with the reading of the film pressure sensor, after the step
  • the processing of step S3 calculates the work done by the glove movement, and calculates the calorie consumption according to the relationship model between the established work and the calorie consumption;
  • Step S42B recording the calorie consumption, and reminding the meal matching and nutritional balance
  • Step S42A Pre-establishing a high-dimensional feature model based on a statistical learning theory, wherein the high-dimensional feature model uses an abnormal pattern of information changes due to various posture deviation actions in the set space as a training sample;
  • Step S41B mapping the abnormal mode to a high-dimensional feature model of a type of support vector machine, separating the target action class, and determining whether the motion posture deviates, and if so, issuing a warning signal.
  • the relationship model is that the mechanical work outputted by the human body during the movement is linked to the calories consumed by the human body in the process by the conversion factor, and the work is calculated by combining the readings of the load cell and the reconstructed motion trajectory, thereby calculating The calories burned out then distinguish the target action class from the other action classes.
  • the model is constructed based on the relationship between work and calories in the glove posture, so that the calculation and analysis can be performed.
  • the work state can be expressed by the motion state, which shows the start time of the motion, the end time of the motion, and the instantaneous speed of the motion; Not all internal energy can be used for work, and part of the internal energy can be lost due to heat consumption and blood flow.
  • the internal energy of the human body can be successfully converted according to a certain ratio.
  • the model relationship between work and calorie consumption is constructed.
  • Mechanical work is measured in wearable computing, usually by measuring the motion trajectory to calculate the work done by the human body. Further, measuring motion trajectories is accomplished using some common sensors such as acceleration sensors and gravity sensors.
  • this example also provides a system for estimating calorie consumption and monitoring hand gesture recognition, and adopts the method for estimating calorie consumption and monitoring hand gesture recognition described in Embodiment 2, and includes the following modules: signal The obtaining and calculating module 41 is configured to collect signals of the movement state of the smart glove, and evaluate the motion state information for preliminary calculation;
  • the abnormality detecting module 42 is configured to identify, by using an abnormality detecting algorithm, whether the signal changes abnormally;
  • the action judging module 43 is configured to distinguish the target action class from the other action class, and use the abnormal mode caused by the strength training threshold as the target action class, and determine whether the posture of the smart glove is deviated. ;
  • the alarm module 44 is configured to issue an alert signal when it is determined that the posture deviation occurs.
  • the signal acquisition and calculation module 41 of this example includes the following units:
  • the inductive acquisition unit 411 is configured to open the entire system and collect motion data, and the collected motion data includes motion data in a strength training direction and motion data in three directions of three-dimensional coordinates;
  • the data processing unit 412 is configured to average the force and motion information in each direction, and use the average value as the motion state information
  • the smoothing unit 413 performs ETL analysis on the data obtained by the data processing unit 412, and smoothes the motion state information by a moving average method.
  • the three-dimensional coordinates of the inductive acquisition unit 411 are the x-axis, the y-axis, and the z-axis that determine the motion state of the object in the stereoscopic space.
  • the abnormality detecting module 42 of this example includes the following units:
  • the abnormality calculating unit 421 is configured to perform data segmentation on the time series of the motion state information to obtain a subsequence, and calculate local abnormal data of the subsequence;
  • the abnormal output unit 422 is configured to output the sub-sequence as an abnormal mode when the abnormal data is greater than or equal to a preset threshold.
  • the present example utilizes a time series anomaly detection algorithm, which can be separated by a more accurate detection standard, and separates the time series corresponding to the standard and abnormal postures, thereby eliminating the abnormal patterns caused by the two common human motions.
  • the abnormality detecting module 42 adopts a multi-dimensional spatial data discriminating algorithm based on a support vector machine.
  • the abnormality detecting module 42 performs the posture abnormality detecting principle as follows: first, all the postures are performed several times according to a standard posture, and the standard posture can be Refer to the instructional video or the coach to guide, then extract the user's source data and the motion track of each action, and save as a reference; then, when the user performs the fitness exercise, the user's fitness hand movement track and reference at this time The trajectory is compared, for example, by comparing the similarity comparison methods, it is possible to determine whether the posture of the user's movement is standard, and the degree of standardization, which is the degree of similarity with the reference trajectory; the acquired data is over time
  • the generated that is, the time series model, in order to facilitate data analysis and processing, this example divides the acquired data, and the subsequence refers to intercepting a piece of data acquired by the sensing module and passing through the ETL process, as for the sub
  • the action judging module 43 in this example includes the following units:
  • the model unit 431 is configured to establish a high-dimensional feature model based on a statistical learning theory preset, wherein the high-dimensional feature model uses an abnormal pattern in the set space due to changes in motion state information due to various human actions as a training sample;
  • the action recognition unit 432 is configured to map the abnormal mode output by the abnormality output unit 422 to a high-dimensional feature model of a type of support vector machine, and separate the target action class.
  • the example also preferably includes a feedback module 45 for feeding back response information for the hand posture warning signal and adjusting a high dimensional feature model of a class of support vector machines.
  • the data obtained in this example is generated over time, that is, the time series model.
  • this example divides the acquired data, and the subsequence refers to intercepting a segment obtained by the sensing module and passing through
  • the data of the ETL process as for the length of the subsequence and the amount of data, can be customized according to the processing capability of the system; the local anomaly data is calculated in the subsequence according to the method of the support vector machine, and some of the data are obviously related to this sub-sequence.
  • the other data of the sequence has deviated data; the preset threshold is a preset threshold, which is comprehensively determined in the data processing process in combination with sensors, algorithms, and user requirements. In different cases, the threshold will be different.
  • the partial motion action will output a corresponding abnormal mode because a significant change in the motion state information is detected. Then, these abnormal patterns will enter the action analysis to determine what kind of action the abnormal mode belongs to.
  • this example uses a multi-class support vector machine (multi-class SVM) that extracts features from the exception mode; multi-class SVM is an extension.
  • Support vector machine algorithm in multi-class SVM, all samples are divided into target classes and other classes; in order to solve the problem of nonlinear classification, the input samples are mapped into a high-dimensional image.
  • the abnormal mode of the non-standard posture is regarded as the target action class
  • the abnormal mode of the other actions is regarded as the other action class.
  • the abnormal mode of non-standard actions has also been mapped to a high-dimensional image in advance.
  • the multi-class SVM judgment it is possible to separate the abnormal action from the abnormal mode outputted in the previous step, and determine what kind of action has occurred depending on the abnormal mode output and the selected model.
  • This example also preferably includes a high-dimensional feature model for adjusting and improving a class of support vector machines, and provides system feedback that can optimize detection and decision algorithms. If the alert is not turned off in time, the system will send help information to others via other devices associated with the signal, such as sending instant messages or short messages through third-party applications. Letter and other help.

Landscapes

  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Physical Education & Sports Medicine (AREA)
  • Engineering & Computer Science (AREA)
  • Textile Engineering (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Alarm Systems (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

一种卡路里消耗和手部姿势识别的智慧手套，包括：处理器模块、存储模块、通信模块、感知模块、预警提示模块、显示模块、电源模块、开关模块和云服务器，存储模块、通信模块、感知模块、预警提示模块、显示模块、电源模块、开关模块和云服务器分别与处理器模块相连接。还公开了采用该智慧手套估算卡路里消耗和监测手部姿势识别的方法和系统。

Description

测量卡路里消耗和监测手部姿势识别的智慧手套及估算的方法和系统 技术领域
本发明涉及健身动作的智能监测与改进装置，尤其涉及一种卡路里消耗和手部姿势识别的方法及系统。
背景技术
如今，随着健康生活的理念越来越深入人心，越来越多的健身爱好者希望能够测量出参加力量训练时候的卡路里的消耗及之后的饮食调节，然而，传统的获取运动过程中人体的能量消耗不仅程序繁琐，且还需要额外的设备，因此我们急切需要找到一种能够方便并有效的检测力量训练时的卡路里的消耗和动作识别的方法。
日常健身过程中，尤其是力量训练时，人体将消耗大量的卡路里，为了有助于训练后的营养补充和膳食搭配，又不要较准确地估算出人体在训练过程中的卡路里消耗；同时，正确的姿势是任何训练项目的必要条件。这是因为正确的姿势有助于减轻甚至消除任何潜在的训练伤病，有助于训练者的身体健康。
为了实现对健身时动作的准备检测和卡路里的消耗，人们提出了利用跑步机，自行车，跑鞋来检测卡路里的消耗的方法，然而采用这些方法搭建的检测系统本身都存在着种种不足，这些系统都采取了特定运动的计算模块来检测卡路里的消耗，在利用这些特定的计算模块进行卡路里的计算的时候，测量模块并不是可移动的，不能够真正地做到可穿戴，在目前的研究和工业生产中，并没有一种设备能够实现上述的功能，后者现有的方法由于模块的特殊性并不能够广泛地应用到这个场景。
发明内容
为了解决现有技术的问题，本发明提供一种卡路里消耗和手部姿势识别的系统，解决现有技术中获取运动过程中人体的能量消耗程序繁琐以及增加额外设备的问题。
本发明是通过以下技术方案实现的:一种测量卡路里消耗和监测手部姿势识别的智慧手套，包括：处理器模块、存储模块、通信模块、感知模块、预警提示模块、显示模块、电源模块、开关模块和云服务器，所述存储模块、通信 模块、感知模块、预警提示模块、显示模块、电源模块、开关模块和云服务器分别与处理器模块相连接，其中，所述感知模块包括压力传感器和IMU惯性单元，所述压力传感器和IMU惯性单元分别用于负责感应用户健身过程中的手部受压力量大小和IMU数据，进而针对用户力量训练收集源数据，并将所述源数据传送给处理器模块；所述处理器模块负责对源始数据的预处理；所述存储模块负责存储处理器预处理的结果以及从云服务器传回的处理结果；所述通信模块负责将源数据、预处理后的数据传至云服务器，供云服务器作进一步的数据处理和分析，并负责将云服务器处理的结果传回至智慧手套；所述预警提示模块则负责健身异常情况时给用户发出警示；所述显示模块负责将用户健身过程中的动作类型、消耗卡路里和姿势标准度的结果显示出来。
本发明同时提供了一种估算卡路里消耗和监测手部姿势识别的方法，采用了所述的测量卡路里消耗和监测手部姿势识别的智慧手套，并包括以下步骤:步骤S1，利用薄膜压力传感器和IMU惯性单元来采集力量训练时的原始数据；步骤S2，将原始数据传输到处理器模块进行处理；步骤S3，将接收到的压力传感器和IMU惯性单元数据进行ETL分析，所述ETL过程为通过ETL技术对步骤S2所得到的数据进行抽取、转置、加载和交付的过程；以及，步骤S4，估算出消耗的卡路里，通过支持向量机建立分类模型，将超出力量训练的源数据作为目标动作类，其他的作为非目标动作类；然后判断是否超出力量训练的阈值，若是则发出提醒信号，若否则将数据传输到可移动设备进行记录，并推荐膳食食谱、营养搭配及训练计划。
作为本发明的进一步改进：在步骤S1中，用户通过按钮或开关将整个系统打开，智慧手套的感知模块上设置有一个或两个以上的压力传感器。
作为本发明的进一步改进：在步骤S2中包括以下步骤：
步骤S21，通过压力传感器和IMU惯性单元采集压力数据和运动姿态数据，基于可穿戴计算技术，将获得的加速度传感器、磁传感器和陀螺仪的数据进行滑动均值滤波，以消除背景噪声，所述滑动均值滤波器的滑动窗口宽度根据用户实际情况进行调整(如所述滑动均值滤波器的滑动窗口宽度为7)；
步骤S22，对加速度传感器和陀螺仪的读数采用互补滤波器的方式实现手套姿态的准确估计，该互补滤波器用于结合加速度传感器对于手套姿态角的估计值和陀螺仪对于手套姿态角的估计值；
步骤S23，采用互补滤波器的方式估计出手套在运动过程中的姿态角后，根据力学原理消除掉重力在智慧手套运动方向上的投射分量，提取出由于智慧手套自身运动所产生的加速度值，实现加速度校准；
步骤S24，获取经过校准后的加速度数据，采用一次积分的方式获得智慧手套运动的速度值；
步骤S25，积分得到智慧手套运动的速度值之后，根据智慧手套起始于静止状态并终止于静止状态的运动特征，对智慧手套运动的速度进行校准；
步骤S26，对所得速度值进行一次积分，从而得到智慧手套运动的位移大小；步骤S27，结合上述结果，对健身动作的轨迹进行重构，从而得到用户健身过程中手部的运动轨迹。
作为本发明的进一步改进：所述步骤S4包括以下步骤：
步骤S41A，结合步骤S2所得到的位移值，结合薄膜压力传感器的读数，经过步骤S3的处理，计算出手套运动所做的功，按照建立的功与卡路里消耗的关系模型计算卡路里消耗；
步骤S42B，记录卡路里的消耗，并进行膳食搭配和营养均衡的提醒；
步骤S42A，基于统计学习理论预先建立高维特征模型，所述高维特征模型以设定空间内由于各项姿态偏离动作导致的信息变化的异常模式作为训练样本；
步骤S41B，将异常模式映射至一类支持向量机的高维特征模型中，分离出目标动作类，并判断出运动姿态是否发生偏离，若是则发出警示信号。
本发明再次提供了一种估算卡路里消耗和监测手部姿势识别的系统，采用了所述的估算卡路里消耗和监测手部姿势识别的方法，并包括以下模块:
信号获取与计算模块，用于收集智慧手套运动状态时的信号，并评估运动状态信息加以初步计算；
异常检测模块，用于通过异常检测算法识别信号是否变化的异常；
动作判断模块，用于将目标动作类和其他动作类进行区分的一类支持向量机，以超出力量训练阈值所导致的异常模式作为目标动作类，并判断智慧手套的运动姿势是否发生姿势偏离；
以及，警报模块，用于当判断发生姿态偏离时发出警示信号。
作为本发明的进一步改进：所述信号获取与计算模块包括以下单元：
感应采集单元，用于开启整个系统并收集运动数据，收集的运动数据包括力量训 练方向上的力和三维坐标各个方向上的运动数据；
数据处理单元，用于对每一个方向上的力和运动信息求得平均值，将此平均值作为运动状态信息；
平滑单元，针对数据处理单元所得到的数据进行ETL分析，并通过滑动平均方法对运动状态信息进行平滑处理。
作为本发明的进一步改进：所述异常检测模块包括以下单元：
异常计算单元，用于对运动状态信息的时间序列实施数据分割得到子序列，计算子序列的局部异常数据；
异常输出单元，用于当所述的异常数据大于或等于预设阈值时，将子序列作为异常模式输出。
作为本发明的进一步改进：所述动作判断模块包括以下单元：
建立模型单元，用于基于统计学习理论预设建立高维特征模型，所述高维特征模型以设定空间内由于各项人体动作导致运动状态信息变化的异常模式作为训练样本；
动作识别单元，用于将异常输出单元所输出的异常模式映射至一类支持向量机的高维特征模型中，分离出目标动作类。
作为本发明的进一步改进：还包括反馈模块，用于反馈针对手部姿势警报信号的响应信息，调整一类支持向量机的高维特征模型。
本发明的有益效果：在进行力量训练过程中，被检测动作的检测准确率为84％-94％，误报率低，以及卡路里的测量准确率高达90％，能够实现对手部姿势偏离判断后发出警报信号，并利用系统的自学习功能处理误报情况，进一步降低误报率；本发明在现有的传感器技术的基础上，进行卡路里的测量和手部姿势识别的工作，采用高效的数据处理方法和算法模型，可广泛地应用力量训练中，为健身爱好者提供良好的参考，并能够提供后期膳食搭配和营养均衡推荐。
附图说明
图1是本发明一种实施例的智慧手套的模块示意图；
图2是本发明另一种实施例的工作原理图；
图3是本发明另一种实施例的工作流程示意图；
图4是本发明再一种实施例的功能模块框图。
具体实施方式
下面结合附图和实施例对本发明作进一步的描述。
一种测量卡路里消耗和监测手部姿势识别的智慧手套，包括：处理器模块、存储模块、通信模块、感知模块、预警提示模块、显示模块、电源模块、开关模块和云服务器，所述存储模块、通信模块、感知模块、预警提示模块、显示模块、电源模块、开关模块和云服务器分别与处理器模块相连接，其中，所述感知模块包括压力传感器和IMU惯性单元，所述压力传感器和IMU惯性单元分别用于负责感应用户健身过程中的手部受压力量大小和IMU数据，进而针对用户力量训练收集源数据，并将所述源数据传送给处理器模块；所述处理器模块负责对源始数据的预处理；所述存储模块负责存储处理器预处理的结果以及从云服务器传回的处理结果；所述通信模块负责将源数据、预处理后的数据传至云服务器，供云服务器作进一步的数据处理和分析，并负责将云服务器处理的结果传回至智慧手套；所述预警提示模块则负责健身异常情况(如当用户动作不标准、运动超量或者受力不当)时给用户发出警示；所述显示模块负责将用户健身过程中的动作类型、消耗卡路里和姿势标准度的结果显示出来。
本发明同时提供了一种估算卡路里消耗和监测手部姿势识别的方法，采用了所述的测量卡路里消耗和监测手部姿势识别的智慧手套，并包括以下步骤:步骤S1，利用压力传感器的薄膜压力传感器和IMU惯性单元来搜集力量训练时的源数据；步骤S2，将源数据传输到处理器模块进行处理；步骤S3，将接收到的压力传感器和IMU惯性单元所获取的数据进行ETL过程，进行ETL分析，所述ETL过程为通过ETL技术对步骤S2所得到的数据进行抽取、转置、加载和交付的过程；以及，步骤S4，计算出消耗的卡路里，通过支持向量机建立分类模型，将超出力量训练的源数据作为目标动作类，其他的作为非目标动作类；然后判断是否超出力量训练的阈值，若是则发出提醒信号，若否则将数据传输到可移动设备进行记录，并推荐膳食食谱、营养搭配及训练计划。
在步骤S1中，用户通过按钮或开关将整个系统打开，智慧手套的感知模块上设置有一个或两个以上的压力传感器。
在步骤S2中包括以下步骤：
步骤S21，通过压力传感器和IMU惯性单元采集压力数据和运动姿态数据，基于可穿戴计算技术，将获得的加速度传感器、磁传感器和陀螺仪的数据进行滑动均值滤波，以消除背景噪声，所述滑动均值滤波器的滑动窗口宽度为7；
步骤S22，对加速度传感器和陀螺仪的读数采用互补滤波器的方式实现手套姿态的准确估计，该互补滤波器用于结合加速度传感器对于手套姿态角的估计值和陀螺仪对于手套姿态角的估计值；
步骤S23，采用互补滤波器的方式估计出手套在运动过程中的姿态角后，根据力学原理消除掉重力在智慧手套运动方向上的投射分量，提取出由于智慧手套自身运动所产生的加速度值，实现加速度校准；
步骤S24，获取经过校准后的加速度数据，采用一次积分的方式获得智慧手套运动的速度值；
步骤S25，积分得到智慧手套运动的速度值之后，根据智慧手套起始于静止状态并终止于静止状态的运动特征，对智慧手套运动的速度进行校准；
步骤S26，对所得速度值进行一次积分，从而得到智慧手套运动的位移大小；步骤S27，结合上述结果，对健身动作的轨迹进行重构，从而得到用户健身过程中手部的运动轨迹。
所述步骤S4包括以下步骤：
步骤S41A，结合步骤S2所得到的位移值，结合薄膜压力传感器的读数，经过步骤S3的处理，计算出手套运动所做的功，按照建立的功与卡路里消耗的关系模型计算卡路里消耗；
步骤S42B，记录卡路里的消耗，并进行膳食搭配和营养均衡的提醒；
步骤S42A，基于统计学习理论预先建立高维特征模型，所述高维特征模型以设定空间内由于各项姿态偏离动作导致的信息变化的异常模式作为训练样本；
步骤S41B，将异常模式映射至一类支持向量机的高维特征模型中，分离出目标动作类，并判断出运动姿态是否发生偏离，若是则发出警示信号。
本发明再次提供了一种估算卡路里消耗和监测手部姿势识别的系统，采用了所述的估算卡路里消耗和监测手部姿势识别的方法，并包括以下模块:
信号获取与计算模块，用于收集智慧手套运动状态时的信号，并评估运动状态信息加以初步计算；
异常检测模块，用于通过异常检测算法识别信号是否变化的异常；
动作判断模块，用于将目标动作类和其他动作类进行区分的一类支持向量机，以超出力量训练阈值所导致的异常模式作为目标动作类，并判断智慧手套的运动姿势是否发生姿势偏离；
以及，警报模块，用于当判断发生姿态偏离时发出警示信号。
所述信号获取与计算模块包括以下单元：
感应采集单元，用于开启整个系统并收集运动数据，收集的运动数据包括力量训练方向上的力和三维坐标各个方向上的运动数据；
数据处理单元，用于对每一个方向上的力和运动信息求得平均值，将此平均值作为运动状态信息；
平滑单元，针对数据处理单元所得到的数据进行ETL分析，并通过滑动平均方法对运动状态信息进行平滑处理。
所述异常检测模块包括以下单元：
异常计算单元，用于对运动状态信息的时间序列实施数据分割得到子序列，计算子序列的局部异常数据；
异常输出单元，用于当所述的异常数据大于或等于预设阈值时，将子序列作为异常模式输出。
所述动作判断模块包括以下单元：
建立模型单元，用于基于统计学习理论预设建立高维特征模型，所述高维特征模型以设定空间内由于各项人体动作导致运动状态信息变化的异常模式作为训练样本；
动作识别单元，用于将异常输出单元所输出的异常模式映射至一类支持向量机的高维特征模型中，分离出目标动作类。
还包括反馈模块，用于反馈针对手部姿势警报信号的响应信息，调整一类支持向量机的高维特征模型。
实施例1：
如图1所示，本例提供一种测量卡路里消耗和监测手部姿势识别的智慧手套，包括：处理器模块、存储模块、通信模块、感知模块、预警提示模块、显示模块、电源模块、开关模块和云服务器，所述存储模块、通信模块、感知模块、预警提示模块、显示模块、电源模块、开关模块和云服务器分别与处理器模块相连接，其中，所述感知模块包括压力传感器和IMU惯性单元，所述压力传感器和IMU惯性单元分别用于负责感应用户健身过程中的手部受压力量大小和IMU数据，进而针对用户力量训练收集源数据，并将所述源数据传送给处理器模块；所述处理器模块负责对源始数据的预处理；所述存储模块负责存储处理器预处理的结果 以及从云服务器传回的处理结果；所述通信模块负责将源数据、预处理后的数据传至云服务器，供云服务器作进一步的数据处理和分析，并负责将云服务器处理的结果传回至智慧手套；所述预警提示模块则负责当用户动作不标准、运动超量或者受力不当中任意一种情况时给用户发出警示；所述显示模块负责将用户健身过程中的动作类型、消耗卡路里和姿势标准度的结果显示出来。
本例所述的智慧手套也可简称为手套；所述智慧手套的感知模块上设置有一个或两个以上的压力传感器，所述压力传感器设置在手掌握力处，所述IMU惯性单元设置在手腕处，便于客户使用方便；所述存储模块包括内置大容量存储器和外置的存储器接口，便于用户运动数据的存储和调用；所述源数据为通过压力传感器和IMU惯性单元分别感应用户健身过程中的手部受压力量大小和IMU数据所对应的采集数据。
本例的使用过程，首先按照标准的姿势将所有姿势做几遍，所述标准的姿势可以参考教学视频或者教练在一旁指导，然后通过感知模块提取出用户的源数据和每一个动作的运动轨迹，并保存作为参照；然后，当用户进行健身运动时，将用户此时的健身手部运动轨迹与参照轨迹进行对比，比如通过相似度比较的方式进行对比，即可判断出用户运动时的姿势是否标准，以及标准程度，所述标准程度也就是与参考轨迹的相似程度；当用户卡路里消耗和手部姿势与参照数据之间的差值达到一定范围，也就是超过了预设的阀值时，所述预警提示模块发出声、光或震动等报警提示；所述阀值根据用户需求进行预先设定，例如，可设置为5。
实施例2：
如图2和图3所示，本例还提供一种估算卡路里消耗和监测手部姿势识别的方法，采用了实施例1所述的智慧手套，并包括以下步骤：
步骤S1，利用压力传感器的薄膜压力传感器和IMU惯性单元来搜集力量训练时的源数据；
步骤S2，将源数据传输到处理器模块进行处理；
步骤S3，将接收到的压力传感器和IMU惯性单元所获取的数据进行ETL过程，进行ETL分析，所述ETL过程为通过ETL技术对步骤S2所得到的数据进行抽取、转置、加载和交付的过程；
以及，步骤S4，计算出消耗的卡路里，通过支持向量机建立分类模型，将超出力量训练的源数据作为目标动作类，其他的作为非目标动作类；然后判断是否超 出力量训练的阈值，若是则发出提醒信号，若否则将数据传输到可移动设备进行记录，并推荐膳食食谱、营养搭配及训练计划。
所述步骤S3中，TL过程指数据的extract、Transform和load，即对数据进行抽取、转置、加载和交付的过程，是指数据处理之前对数据进行预处理的重要步骤；步骤S4中，根据设置的计算模型，人体在运动过程中对外输出的机械功可以通过一个转化因子与人体在这一过程中消耗的卡路里联系起来，结合测力传感器的读数和重构所得运动轨迹来计算功，从而计算出消耗的卡路里和将目标动作类和其他动作类进行区分的一类支持向量机；将超出力量训练作为目标动作类，并判断是否超出力量训练的阈值，若是，则发出提醒信号；本步骤S4将数据传输到可移动设备，进行记录，推荐膳食食谱、营养搭配及训练计划,；所述阀值可以根据用户的需求进行自定义设置，如在使用杠铃时，由预先设置的力的大小，称之为阈值，如果超过这个预先设置的大小，则发出信号。
在实际应用中，通过处理器模块接收压力传感器和IMU惯性单元测量出所需的运动数据，建立出运动信号和力量训练动作的关系，只需要使用简单的传感器即能够通过被检测者的卡路里的消耗和手部姿势的识别，判断出被检测者是否发生手部姿势的偏离并进行报警，减少了对笨重的测量设备的依赖，将大大地提高手部姿势识别的正确率；在特定的运动中，可通过可移动设备终端显示出卡路里消耗量及膳食搭配推荐。
在本例中，所述智慧手套的压力传感器数量为一个或者两个以上，所述处理器模块所包括的中央处理器的数目为一个，如附图1所示，被检测的智慧手套上携带两类传感器：压力传感器和IMU惯性单元，这两类传感器将收集到运动数据经过互补滤波、指数均值和ETL等过程传输到中央处理器中，再根据设置的计算模型和机器学习的方法进行手部姿势的识别。
本例在步骤S1中，用户优选通过按钮或开关将整个系统打开，所述按钮和开关可以是通过触摸感应器来实现的按钮和开关，所述按钮和开关可以设置在IMU惯性单元附件，以便于用户操作。
本例所述步骤S2利用异常检测算法识别所获取的运动信息的异常是基于局部异常因子的时间序列异常检测算法，步骤S2中包括以下步骤：
步骤S21，通过压力传感器和IMU惯性单元采集压力数据和运动姿态数据，基于可穿戴计算技术，将获得的加速度传感器、磁传感器和陀螺仪的数据进行滑动 均值滤波，以消除背景噪声，所述滑动均值滤波器的滑动窗口宽度为7；
步骤S22，对加速度传感器和陀螺仪的读数采用互补滤波器的方式实现手套姿态的准确估计，该互补滤波器用于结合加速度传感器对于手套姿态角的估计值和陀螺仪对于手套姿态角的估计值；
步骤S23，采用互补滤波器的方式估计出手套在运动过程中的姿态角后，根据力学原理消除掉重力在智慧手套运动方向上的投射分量，提取出由于智慧手套自身运动所产生的加速度值，实现加速度校准；
步骤S24，获取经过校准后的加速度数据，采用一次积分的方式获得智慧手套运动的速度值；
步骤S25，积分得到智慧手套运动的速度值之后，根据智慧手套起始于静止状态并终止于静止状态的运动特征，对智慧手套运动的速度进行校准；
步骤S26，对所得速度值进行一次积分，从而得到智慧手套运动的位移大小。
步骤S22中，通过互补滤波器实现数据的分析处理，物体在空间中的姿态，可由加速度和角速度结合相应进而达到准确估计的目的，因此，所述步骤S22通过互补滤波器结合加速度传感器测量的手套姿态角的估计值和陀螺仪测量的手套姿态角的估计值，进而得到手套姿势的准确估计。步骤S23中，提取出由于智慧手套自身运动所产生的加速度值，也就是要滤除掉智慧手套在三维坐标的三个方向上的加速度分量，这一点可以通过互补滤波或卡曼滤波等方式实现。步骤24中，为了较少积分后的速度漂移，积分只针对对应于手套运动时的加速度值，为此，指数均值(EMA)被采用作为判断手套加速度值对应于手套的静止状态还是运动状态；当滑动窗口内指数均值大于阈值时，该窗口内的加速度被判断为对应于手套的运动状态；反之，则被判断为对应于手套的静止状态；当判断过程完成后，积分只被作用于对应手套运动状态下的加速度值，从而获得相应的速度值，而对应于手套静止状态下的加速度值则被视为零值；所述阈值是在数据处理过程中，结合传感器、算法以及用户需求进行综合确定，不同的情况下，此阈值会有所不同。所述步骤S25中，采用步骤S24所述的指数均值(EMA)判断手套的运动状态，当判定为静止时，则将对应的速度值强制设置为0。
本例还包括用于卡路里消耗和手部姿势识别的调整一类支持向量机的高维特征模型，所述步骤S4包括以下步骤：
步骤S41A，结合步骤S2所得到的位移值，结合薄膜压力传感器的读数，经过步 骤S3的处理，计算出手套运动所做的功，按照建立的功与卡路里消耗的关系模型计算卡路里消耗；
步骤S42B，记录卡路里的消耗，并进行膳食搭配和营养均衡的提醒；
步骤S42A，基于统计学习理论预先建立高维特征模型，所述高维特征模型以设定空间内由于各项姿态偏离动作导致的信息变化的异常模式作为训练样本；
步骤S41B，将异常模式映射至一类支持向量机的高维特征模型中，分离出目标动作类，并判断出运动姿态是否发生偏离，若是则发出警示信号。
所述关系模型为根据人体在运动过程中对外输出的机械功通过转化因子与人体在这一过程中消耗的卡路里联系起来，结合测力传感器的读数和重构所得运动轨迹来计算功，从而计算出消耗的卡路里进而将目标动作类和其他动作类进行区分。
本例根据手套姿势做功与卡路里之间的关系来构建出模型，从而来进行计算分析，功与能量之间可用运动状态来表示，体现出运动开始时刻、运动结束时刻和运动的瞬时速度；因为不是所有的内能用来做功，一部分内能由于热量消耗，血液流动而损耗，人体的内能按照一定的比率转化成功。能量流失的计量按照以下的公式来计算：人体运动过程中对外输出的机械功＝卡路里消耗转为机械功的效率*人体运动过程中消耗高的卡路里。
基于机器学习的相关知识，构建出做功与卡路里消耗之间的模型关系。可穿戴计算中测量机械功，通常是测量运动轨迹从而计算出来人体所做的功。进一步，测量运动轨迹是用一些常见的传感器例如加速度传感器和重力传感器来实现。
实施例3：
如图4所示，本例还提供一种估算卡路里消耗和监测手部姿势识别的系统，采用了实施例2所述的估算卡路里消耗和监测手部姿势识别的方法，并包括以下模块：信号获取与计算模块41，用于收集智慧手套运动状态时的信号，并评估运动状态信息加以初步计算；
异常检测模块42，用于通过异常检测算法识别信号是否变化的异常；
动作判断模块43，用于将目标动作类和其他动作类进行区分的一类支持向量机，以超出力量训练阈值所导致的异常模式作为目标动作类，并判断智慧手套的运动姿势是否发生姿势偏离；
以及，警报模块44，用于当判断发生姿态偏离时发出警示信号。
本例所述信号获取与计算模块41包括以下单元：
感应采集单元411，用于开启整个系统并收集运动数据，收集的运动数据包括力量训练方向上的力和三维坐标各个方向上的运动数据；
数据处理单元412，用于对每一个方向上的力和运动信息求得平均值，将此平均值作为运动状态信息；
平滑单元413，针对数据处理单元412所得到的数据进行ETL分析，并通过滑动平均方法对运动状态信息进行平滑处理。
所述感应采集单元411的三维坐标为确定物体在立体空间上的运动状态的x轴、y轴和z轴。
本例所述异常检测模块42包括以下单元：
异常计算单元421，用于对运动状态信息的时间序列实施数据分割得到子序列，计算子序列的局部异常数据；
异常输出单元422，用于当所述的异常数据大于或等于预设阈值时，将子序列作为异常模式输出。
优选地，本例利用时间序列异常检测算法，可通过更为精确的检测标准限定，将标准和异常两种姿势所对应的时间序列分离出去，排除此两种常见的人体动作造成的异常模式。
所述异常检测模块42采用的是基于支持向量机的多维空间数据判别算法，该异常检测模块42进行姿势异常检测原理为：首先按照标准的姿势将所有姿势做几遍，所述标准的姿势可以参考教学视频或者教练在一旁指导，然后提取出用户的源数据和每一个动作的运动轨迹，并保存作为参照；然后，当用户进行健身运动时，将用户此时的健身手部运动轨迹与参照轨迹进行对比，比如通过相似度比较的方式进行对比，即可判断出用户运动时的姿势是否标准，以及标准程度，所述标准程度也就是与参考轨迹的相似程度；获取的数据是随着时间而产生的，也就是时间序列模型，为便于实现数据分析和处理，本例对获取的数据进行分割，而子序列指的是截取一段由感知模块获取并经过ETL过程的数据，至于子序列的长度和数据量，根据系统的处理能力可以自定义设置；所述局部异常数据是指在子序列中，按照支持向量机的方法计算，部分数据明显与此子序列的其他数据有偏差的数据；所述预设阀值为预先设置的阀值，是在数据处理过程中，结合传 感器、算法以及用户需求进行综合确定，不同的情况下，此阈值会有所不同。
本例所述动作判断模块43包括以下单元：
建立模型单元431，用于基于统计学习理论预设建立高维特征模型，所述高维特征模型以设定空间内由于各项人体动作导致运动状态信息变化的异常模式作为训练样本；
动作识别单元432，用于将异常输出单元422所输出的异常模式映射至一类支持向量机的高维特征模型中，分离出目标动作类。
本例还优选包括反馈模块45，用于反馈针对手部姿势警报信号的响应信息，调整一类支持向量机的高维特征模型。
本例获取的数据是随着时间而产生的，也就是时间序列模型，为便于实现数据分析和处理，本例对获取的数据进行分割，而子序列指的是截取一段由感知模块获取并经过ETL过程的数据，至于子序列的长度和数据量，根据系统的处理能力可以自定义设置；所述局部异常数据是指在子序列中，按照支持向量机的方法计算，部分数据明显与此子序列的其他数据有偏差的数据；所述预设阀值为预先设置的阀值，是在数据处理过程中，结合传感器、算法以及用户需求进行综合确定，不同的情况下，此阈值会有所不同。
在完成异常检测模块42后，部分运动动作将因为造成运动状态信息的明显变化被检测到而输出对应的异常模式。然后，对这些异常模式将进入动作分析，从而判断异常模式是属于何种动作。为了从这些模式中区别出异常动作，本例使用了从异常模式中所提取过的特征的多类支持向量机(multi-class Support Vector Machine,multi-class SVM)；多类SVM是一种扩展的支持向量机算法，在多类SVM，所有的样本分为目标类和其他类；为了解决非线性分类的问题，将输入样本映射成一个高维图像。在本例中，非标准姿势的异常模式被视作目标动作类，而其他动作的异常模式被视作为其他动作类。非标准动作的异常模式亦已经被事先映射成一个高维图像。通过利用多类SVM判断，可以从上一步骤输出的异常模式中分离出异常动作来，依赖于由输出的异常模式和选定的模型，从而决定发生了何种动作。
本例还优选包括用于调整完善一类支持向量机的高维特征模型，提供可以优化检测和决策算法的系统反馈。如果警报没有及时被关闭，系统则会通过信号关联的其他设备向他人发出求助信息，比如通过第三方应用发送即时信息或短 信等求助。
以上内容是结合具体实现方式对本发明做的进一步阐述，不应认定本发明的具体实现只局限于以上说明。对于本技术领域的技术人员而言，在不脱离本发明构思的前提下，还可以做出若干简单推演或替换，均应视为有本发明所提交的权利要求确定的保护范围之内。

Claims (10)

  1. 一种测量卡路里消耗和监测手部姿势识别的智慧手套，其特征在于，包括：处理器模块、存储模块、通信模块、感知模块、预警提示模块、显示模块、电源模块、开关模块和云服务器，所述存储模块、通信模块、感知模块、预警提示模块、显示模块、电源模块、开关模块和云服务器分别与处理器模块相连接，其中，所述感知模块包括压力传感器和IMU惯性单元，所述压力传感器和IMU惯性单元分别用于负责感应用户健身过程中的手部受压力量大小和IMU数据，进而针对用户力量训练收集源数据，并将所述源数据传送给处理器模块；所述处理器模块负责对源始数据的预处理；所述存储模块负责存储处理器预处理的结果以及从云服务器传回的处理结果；所述通信模块负责将源数据、预处理后的数据传至云服务器，供云服务器作进一步的数据处理和分析，并负责将云服务器处理的结果传回至智慧手套；所述预警提示模块则负责健身异常情况时给用户发出警示；所述显示模块负责将用户健身过程中的动作类型、消耗卡路里和姿势标准度的结果显示出来。
  2. 一种估算卡路里消耗和监测手部姿势识别的方法，其特征在于，采用了权利要求1所述的测量卡路里消耗和监测手部姿势识别的智慧手套，并包括以下步骤:步骤S1，利用压力传感器的薄膜压力传感器和IMU惯性单元来搜集力量训练时的源数据；步骤S2，将源数据传输到处理器模块进行处理；步骤S3，将接收到的压力传感器和IMU惯性单元所获取的数据进行ETL过程，进行ETL分析，所述ETL过程为通过ETL技术对步骤S2所得到的数据进行抽取、转置、加载和交付的过程；以及，步骤S4，计算出消耗的卡路里，通过支持向量机建立分类模型，将超出力量训练的源数据作为目标动作类，其他的作为非目标动作类；然后判断是否超出力量训练的阈值，若是则发出提醒信号，若否则将数据传输到移动设备进行记录，并推荐膳食食谱、营养搭配及训练计划。
  3. 根据权利要求2所述的估算卡路里消耗和监测手部姿势识别的方法，其特征在于，在步骤S1中，用户通过按钮或开关将整个系统打开，智慧手套的感知模块上设置有一个或两个以上的压力传感器。
  4. 根据权利要求2所述的估算卡路里消耗和监测手部姿势识别的方法，其特征在于，在步骤S2中包括以下步骤：
    步骤S21，通过压力传感器和IMU惯性单元采集压力数据和运动姿态数据，将 获得的加速度传感器、磁传感器和陀螺仪的数据进行滑动均值滤波，以消除背景噪声，所述滑动均值滤波器的滑动窗口宽度根据用户实际情况进行调整；
    步骤S22，对加速度传感器和陀螺仪的读数采用互补滤波器的方式实现手套姿态的准确估计，该互补滤波器用于结合加速度传感器对于手套姿态角的估计值和陀螺仪对于手套姿态角的估计值；
    步骤S23，采用互补滤波器的方式估计出手套在运动过程中的姿态角后，根据力学原理消除掉重力在智慧手套运动方向上的投射分量，提取出由于智慧手套自身运动所产生的加速度值，实现加速度校准；
    步骤S24，获取经过校准后的加速度数据，采用一次积分的方式获得智慧手套运动的速度值；
    步骤S25，积分得到智慧手套运动的速度值之后，根据健身动作起始于静止状态并终止于静止状态的运动特征，对智慧手套运动的速度进行校准；
    步骤S26，对所得速度值进行一次积分，从而得到智慧手套运动的位移大小；
    步骤S27，结合上述结果，对健身动作的轨迹进行重构，从而得到用户健身过程中手部的运动轨迹。
  5. 根据权利要求2所述的估算卡路里消耗和监测手部姿势识别的方法，其特征在于，所述步骤S4包括以下步骤：
    步骤S41A，结合步骤S2所得到的位移值，结合薄膜压力传感器的读数，经过步骤S3的处理，计算出手套运动所做的功，按照建立的功与卡路里消耗的关系模型计算卡路里消耗；
    步骤S42B，记录卡路里的消耗，并进行膳食搭配和营养均衡的提醒；
    步骤S42A，基于统计学习理论预先建立高维特征模型，所述高维特征模型以设定空间内由于各项姿态偏离动作导致的信息变化的异常模式作为训练样本；
    步骤S41B，将异常模式映射至一类支持向量机的高维特征模型中，分离出目标动作类，并判断出运动姿态是否发生偏离，若是则发出警示信号。
  6. 一种估算卡路里消耗和监测手部姿势识别的系统，其特征在于，采用了权利要求2至5任意一项所述的卡路里消耗和手部姿势识别的方法，并包括以下模块:信号获取与计算模块，用于收集智慧手套运动状态时的信号，并评估运动状态信息加以初步计算；
    异常检测模块，用于通过异常检测算法识别信号是否变化的异常；
    动作判断模块，用于将目标动作类和其他动作类进行区分的一类支持向量机，以超出力量训练阈值所导致的异常模式作为目标动作类，并判断智慧手套的运动姿势是否发生姿势偏离；
    以及，警报模块，用于当判断发生姿态偏离时发出警示信号。
  7. 根据权利要求6所述的估算卡路里消耗和监测手部姿势识别的系统，其特征在于，所述信号获取与计算模块包括以下单元：
    感应采集单元，用于开启整个系统并收集运动数据，收集的运动数据包括力量训练方向上的力和三维坐标各个方向上的运动数据；
    数据处理单元，用于对每一个方向上的力和运动信息求得平均值，将此平均值作为运动状态信息；
    平滑单元，针对数据处理单元所得到的数据进行ETL分析，并通过滑动平均方法对运动状态信息进行平滑处理。
  8. 根据权利要求6所述的估算卡路里消耗和监测手部姿势识别的系统，其特征在于，所述异常检测模块包括以下单元：
    异常计算单元，用于对运动状态信息的时间序列实施数据分割得到子序列，计算子序列的局部异常数据；
    异常输出单元，用于当所述的异常数据大于或等于预设阈值时，将子序列作为异常模式输出。
  9. 根据权利要求8所述的估算卡路里消耗和监测手部姿势识别的系统，其特征在于，所述动作判断模块包括以下单元：
    建立模型单元，用于基于统计学习理论预设建立高维特征模型，所述高维特征模型以设定空间内由于各项人体动作导致运动状态信息变化的异常模式作为训练样本；
    动作识别单元，用于将异常输出单元所输出的异常模式映射至一类支持向量机的高维特征模型中，分离出目标动作类。
  10. 根据权利要求6所述的估算卡路里消耗和监测手部姿势识别的系统，其特征在于，还包括反馈模块，用于反馈针对手部姿势警报信号的响应信息，调整多分类支持向量机的高维特征模型。
PCT/CN2016/080734 2016-03-18 2016-04-29 测量卡路里消耗和监测手部姿势识别的智慧手套及估算的方法和系统 WO2017156834A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610155558.6 2016-03-18
CN201610155558.6A CN105833508A (zh) 2016-03-18 2016-03-18 测量卡路里消耗和监测手部姿势识别的智慧手套及估算的方法和系统

Publications (1)

Publication Number Publication Date
WO2017156834A1 true WO2017156834A1 (zh) 2017-09-21

Family

ID=56587457

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/080734 WO2017156834A1 (zh) 2016-03-18 2016-04-29 测量卡路里消耗和监测手部姿势识别的智慧手套及估算的方法和系统

Country Status (2)

Country Link
CN (1) CN105833508A (zh)
WO (1) WO2017156834A1 (zh)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108647942A (zh) * 2018-07-27 2018-10-12 重庆爱鸽者智能科技有限公司 一种基于通用移动终端的分布式赛鸽执裁系统
CN110941223A (zh) * 2019-12-24 2020-03-31 湖南交科智能科技有限公司 一种智慧路锥系统
CN111696646A (zh) * 2019-03-13 2020-09-22 青岛海尔电冰箱有限公司 用于冰箱的推荐菜谱的方法和装置
CN112580587A (zh) * 2020-12-29 2021-03-30 西安精密机械研究所 基于振动噪声信号的骨关节受损信息评估系统及评估方法
CN112698288A (zh) * 2020-11-17 2021-04-23 芜湖美的厨卫电器制造有限公司 用于识别姿态的方法、装置、处理器、热水器及监测系统
CN112762895A (zh) * 2020-10-30 2021-05-07 四川写正智能科技有限公司 一种基于传感器判定读写姿态的方法
CN114886387A (zh) * 2022-07-11 2022-08-12 深圳市奋达智能技术有限公司 基于压感的走跑运动卡路里计算方法、系统及存储介质
CN115487477A (zh) * 2022-08-11 2022-12-20 安徽中飞星梦体育文化发展有限公司 一种可适用不同人群的匍匐隧道及其系统
CN117883076A (zh) * 2024-01-23 2024-04-16 北京邦尼营策科技有限公司 一种基于大数据的人体运动能量消耗监测系统及方法
CN118642600A (zh) * 2024-08-15 2024-09-13 安徽大学 一种基于直角双平面电容传感阵列的手势识别方法和系统

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107049324B (zh) * 2016-11-23 2019-09-17 深圳大学 一种肢体运动姿态的判断方法及装置
CN108270658B (zh) * 2017-01-02 2021-04-06 郑洪� 信息推送方法、推送装置及推送系统
CN109646902A (zh) * 2017-10-10 2019-04-19 齐新旺 一种基于识别设备的健身运动计量方法
CN107875621B (zh) * 2017-11-28 2019-08-27 杨佰振 一种自动化智能数字运动健身管理系统
CN109011508A (zh) * 2018-07-30 2018-12-18 三星电子（中国）研发中心 一种智能教练系统及方法
CN109830278B (zh) * 2019-01-15 2021-01-15 深圳和而泰智能控制股份有限公司 无氧运动健身推荐方法、装置、无氧运动设备及存储介质
CN110660411B (zh) * 2019-09-17 2021-11-02 北京声智科技有限公司 基于语音识别的健身安全提示方法、装置、设备及介质
CN113952700B (zh) * 2021-11-22 2022-12-09 深圳市天鹏宇科技有限公司 一种智能健身交互系统以及智能健身远程指导系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020194668A1 (en) * 2001-05-24 2002-12-26 Seung-Woo Kwon Functional golf gloves
CN102698425A (zh) * 2012-05-28 2012-10-03 吴江科沃客机械设备有限公司 一种运动手套
CN202751769U (zh) * 2012-05-31 2013-02-27 靖守禹 多功能拳套
US20140309547A1 (en) * 2013-04-12 2014-10-16 Michael Linderman Athletic Glove Providing Feedback Regarding Grip to a Wearer
CN104888444A (zh) * 2015-05-20 2015-09-09 深圳大学 一种卡路里消耗和手部姿势识别的智慧手套、方法及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020194668A1 (en) * 2001-05-24 2002-12-26 Seung-Woo Kwon Functional golf gloves
CN102698425A (zh) * 2012-05-28 2012-10-03 吴江科沃客机械设备有限公司 一种运动手套
CN202751769U (zh) * 2012-05-31 2013-02-27 靖守禹 多功能拳套
US20140309547A1 (en) * 2013-04-12 2014-10-16 Michael Linderman Athletic Glove Providing Feedback Regarding Grip to a Wearer
CN104888444A (zh) * 2015-05-20 2015-09-09 深圳大学 一种卡路里消耗和手部姿势识别的智慧手套、方法及系统

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108647942B (zh) * 2018-07-27 2024-03-15 重庆爱鸽者智能科技有限公司 一种基于通用移动终端的分布式赛鸽执裁系统
CN108647942A (zh) * 2018-07-27 2018-10-12 重庆爱鸽者智能科技有限公司 一种基于通用移动终端的分布式赛鸽执裁系统
CN111696646A (zh) * 2019-03-13 2020-09-22 青岛海尔电冰箱有限公司 用于冰箱的推荐菜谱的方法和装置
CN110941223A (zh) * 2019-12-24 2020-03-31 湖南交科智能科技有限公司 一种智慧路锥系统
CN112762895A (zh) * 2020-10-30 2021-05-07 四川写正智能科技有限公司 一种基于传感器判定读写姿态的方法
CN112762895B (zh) * 2020-10-30 2022-07-15 四川写正智能科技有限公司 一种基于传感器判定读写姿态的方法
CN112698288A (zh) * 2020-11-17 2021-04-23 芜湖美的厨卫电器制造有限公司 用于识别姿态的方法、装置、处理器、热水器及监测系统
CN112580587A (zh) * 2020-12-29 2021-03-30 西安精密机械研究所 基于振动噪声信号的骨关节受损信息评估系统及评估方法
CN114886387B (zh) * 2022-07-11 2023-02-14 深圳市奋达智能技术有限公司 基于压感的走跑运动卡路里计算方法、系统及存储介质
CN114886387A (zh) * 2022-07-11 2022-08-12 深圳市奋达智能技术有限公司 基于压感的走跑运动卡路里计算方法、系统及存储介质
CN115487477A (zh) * 2022-08-11 2022-12-20 安徽中飞星梦体育文化发展有限公司 一种可适用不同人群的匍匐隧道及其系统
CN117883076A (zh) * 2024-01-23 2024-04-16 北京邦尼营策科技有限公司 一种基于大数据的人体运动能量消耗监测系统及方法
CN118642600A (zh) * 2024-08-15 2024-09-13 安徽大学 一种基于直角双平面电容传感阵列的手势识别方法和系统

Also Published As

Publication number Publication date
CN105833508A (zh) 2016-08-10

Similar Documents

Publication Publication Date Title
WO2017156834A1 (zh) 测量卡路里消耗和监测手部姿势识别的智慧手套及估算的方法和系统
WO2017156835A1 (zh) 健身姿势识别、评估、预警和强度估算的智能方法及系统
CN104888444B (zh) 一种卡路里消耗和手部姿势识别的智慧手套、方法及系统
US10765347B2 (en) Gait analysis device and computer program product
US20210319585A1 (en) Method and system for gaze estimation
Jensen et al. Classification of kinematic swimming data with emphasis on resource consumption
KR101426750B1 (ko) 열화상을 이용한 심박수 측정 시스템
Huynh et al. Fall detection system using combination accelerometer and gyroscope
TW201443834A (zh) 平衡監測與跌倒預警方法及裝置
CN107320108B (zh) 一种关节活动度测量方法
CN103543834A (zh) 一种手势识别装置与方法
US20210059569A1 (en) Fall risk evaluation method, fall risk evaluation device, and non-transitory computer-readable recording medium in which fall risk evaluation program is recorded
CN104473648A (zh) 一种结合生理参数监测的人体摔倒预警和检测方法
CN106725495A (zh) 一种跌倒检测方法、装置及系统
US20240057946A1 (en) Sarcopenia evaluation method, sarcopenia evaluation device, and non-transitory computer-readable recording medium in which sarcopenia evaluation program is recorded
Liu et al. Automatic fall risk detection based on imbalanced data
Mollaret et al. Perceiving user's intention-for-interaction: A probabilistic multimodal data fusion scheme
CN117594181A (zh) 基于数据检测的健康报告生成方法和系统
US11482046B2 (en) Action-estimating device
CN103870814A (zh) 基于智能相机的非接触式实时眼动识别方法
Liao et al. A vision-based walking posture analysis system without markers
Paulo et al. Human gait pattern changes detection system: A multimodal vision-based and novelty detection learning approach
Pramanick et al. Defatigue: Online non-intrusive fatigue detection by a robot co-worker
Huynh et al. Detection of activities daily living and falls using combination accelerometer and gyroscope
CN108784720B (zh) 基于肌张力传感器的痉挛检测的控制系统及其检测方法

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16893997

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 18.01.2019)

122 Ep: pct application non-entry in european phase

Ref document number: 16893997

Country of ref document: EP

Kind code of ref document: A1