CN105912142A - Step recording and behavior identification method based on acceleration sensor - Google Patents

Step recording and behavior identification method based on acceleration sensor Download PDF

Info

Publication number
CN105912142A
CN105912142A CN201610082327.7A CN201610082327A CN105912142A CN 105912142 A CN105912142 A CN 105912142A CN 201610082327 A CN201610082327 A CN 201610082327A CN 105912142 A CN105912142 A CN 105912142A
Authority
CN
China
Prior art keywords
acceleration
human body
waveform
axis
steps
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.)
Granted
Application number
CN201610082327.7A
Other languages
Chinese (zh)
Other versions
CN105912142B (en
Inventor
黄伟
李建
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ai Kangweida Intelligent Medical Science And Technology Ltd Of Shenzhen
Original Assignee
Ai Kangweida Intelligent Medical Science And Technology Ltd Of Shenzhen
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 Ai Kangweida Intelligent Medical Science And Technology Ltd Of Shenzhen filed Critical Ai Kangweida Intelligent Medical Science And Technology Ltd Of Shenzhen
Priority to CN201610082327.7A priority Critical patent/CN105912142B/en
Publication of CN105912142A publication Critical patent/CN105912142A/en
Application granted granted Critical
Publication of CN105912142B publication Critical patent/CN105912142B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/03—Arrangements for converting the position or the displacement of a member into a coded form
    • G06F3/033—Pointing devices displaced or positioned by the user, e.g. mice, trackballs, pens or joysticks; Accessories therefor
    • G06F3/0346—Pointing devices displaced or positioned by the user, e.g. mice, trackballs, pens or joysticks; Accessories therefor with detection of the device orientation or free movement in a 3D space, e.g. 3D mice, 6-DOF [six degrees of freedom] pointers using gyroscopes, accelerometers or tilt-sensors
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C22/00—Measuring distance traversed on the ground by vehicles, persons, animals or other moving solid bodies, e.g. using odometers, using pedometers
    • G01C22/006—Pedometers
    • A—HUMAN NECESSITIES
    • A43—FOOTWEAR
    • A43B—CHARACTERISTIC FEATURES OF FOOTWEAR; PARTS OF FOOTWEAR
    • A43B3/00—Footwear characterised by the shape or the use
    • A43B3/34—Footwear characterised by the shape or the use with electrical or electronic arrangements
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/1123—Discriminating type of movement, e.g. walking or running
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/6802—Sensor mounted on worn items
    • A61B5/6804—Garments; Clothes
    • A61B5/6807—Footwear
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2560/00—Constructional details of operational features of apparatus; Accessories for medical measuring apparatus
    • A61B2560/02—Operational features
    • A61B2560/029—Operational features adapted for auto-initiation
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2562/00—Details of sensors; Constructional details of sensor housings or probes; Accessories for sensors
    • A61B2562/02—Details of sensors specially adapted for in-vivo measurements
    • A61B2562/0219—Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2562/00—Details of sensors; Constructional details of sensor housings or probes; Accessories for sensors
    • A61B2562/02—Details of sensors specially adapted for in-vivo measurements
    • A61B2562/0247—Pressure sensors
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/1116—Determining posture transitions
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/1118—Determining activity level
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/112—Gait analysis
    • 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
    • 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
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/30—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Public Health (AREA)
  • Surgery (AREA)
  • Pathology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Veterinary Medicine (AREA)
  • Physiology (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Remote Sensing (AREA)
  • Psychiatry (AREA)
  • Signal Processing (AREA)
  • General Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Dentistry (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Human Computer Interaction (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

The invention discloses a step recording and behavior identification method based on an acceleration sensor. The method is characterized in that the placing method of the sensor is that the forward direction of a foot is the positive direction of an X-axis, the leftward direction is the positive direction of a Y-axis, and a foot lifting direction is the negative direction of a Z-axis. Since when a human body moves, the displacement on the X-axis direction is relatively large, and acceleration changes are relatively obvious and have obvious periodicity, motion step number can be identified and motion behaviors can be identified just by setting rational threshold values. By smoothing filtering and Kalman filtering, waveform is smoother and errors are reduced, so a system can accurately record steps in real time. The method can accurately distinguish various motions of a person. The method can calculate step number of various motions in real time. In addition, the method does not have very high computing capacity on a system, and a mobile phone or a common single-chip microcomputer (MCU) in common configuration can complete operation.

Description

Step recording and behavior identification method based on acceleration sensor
Technical Field
The invention belongs to a step recording and behavior identification method based on an acceleration sensor.
Background
Early motion recognition was primarily visually based, recognizing the type of motion of a person given a sequence of images or a video segment. The vision-based method has the advantages of natural interaction, rich extracted characteristic information and the like, but the method also has some limitations in practical application and needs to overcome many problems. Such as lighting conditions in the environment, the position of the person in front of the camera, the size of the field, etc. The sensor has the advantages of low price, convenient carrying, no limitation of places and the like, along with the development of the devices, the motion recognition is brought into a new research field, the defects of the traditional motion recognition method based on vision in practical application are supplemented, and the application of the motion recognition in daily life is promoted. This technique has been used in rehabilitation status monitoring for patients with behavioral disorders, preventive monitoring for sudden illness in the elderly, and the like. Common sensors include an acceleration sensor, a gyroscope, a microphone and the like, some devices with built-in sensors such as an AppleiPhone, a Nintendo Wiimote and the like are developed, and the development of the wireless devices enables wide-range interactive application such as intelligent home and mixed reality application.
For motion recognition using acceleration sensors, the main problems are three: firstly, the method is used for quickly and automatically segmenting the acceleration signal output by the sensor so as to achieve the purpose of online motion segmentation and prepare for subsequent online identification; secondly, establishing an effective classification model to achieve the purpose of efficiently and accurately classifying and identifying the motion; thirdly, a proper method is adopted to identify the movement ending time so as to improve the interaction. The invention takes the three problems as basic starting points, analyzes key problems in the motion recognition process, solves the main technical problems mentioned above and realizes an efficient online motion recognition system.
For the problem of acceleration signal segmentation, many research works are to segment the sensor signals manually as a training and testing database. The load of signal processing is reduced, the data is more ideal, the influence of the data is eliminated, and the performance of the analysis and identification algorithm can be contrasted. In practical application, however, the manual method has poor mutual inductance, and is inconvenient to operate and apply, so that signals need to be segmented online; for the selection of classification models, most of research and corresponding systems at the present stage adopt a dynamic time warping algorithm (DTW) and a Hidden Markov Model (HMM), the DTW algorithm requires less training data, and can dynamically update matched templates. The operation speed of the algorithm is obviously slowed down along with the increase of the length of time sequence data to be recognized and the number of templates, the HMM method uses one state to represent the current action, but many systemic actions are complex and cannot be sufficiently represented by only one state, so two or more state variables are required to represent the actions, the invention adopts the Fused HMM method, solves the problem that a single HMM cannot simultaneously model two time sequence sequences with correlation, has good description capacity for the systemic action with the interaction process, and when one HMM information is lost, the other HMM still can normally work, thereby increasing the robustness of the algorithm; for the problem of performing motion recognition in advance, the current main processing method is to call the recognition process after a motion is completed, and in some applications, the user experience is reduced due to the delay. The invention adopts an autoregressive prediction model, uses known frame data to predict unknown data, and can start the recognition process before the movement is finished by analyzing the predicted data, thereby achieving the effect of advanced recognition.
Disclosure of Invention
The invention aims to provide a step recording and behavior identification method based on an acceleration sensor.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a step recording and behavior identification method based on an acceleration sensor is characterized in that the sensor placement mode adopted by the method is as follows: the forward direction of the feet is the positive direction of an X axis, the leftward direction is the positive direction of a Y axis, and the foot lifting direction is the negative direction of a Z axis; therefore, the number of steps of the movement can be identified as long as a reasonable threshold value is set;
the acceleration of the human body is larger than a threshold value Ax when the acceleration of an X axis is in motion by collecting a large number of data samples, when the acceleration is changed from less than Ax to more than Ax, then the acceleration is changed from more than Ax to correspond to the actions of lifting and falling the foot of the human body, namely the human body is recognized to move by one step, because some errors exist in the sensors, the acceleration of a plurality of points possibly appears in one step and is wandering around Ax, the condition of a plurality of steps can be calculated in a corresponding mode, in order to eliminate the condition, the step number of the human body moving in one second can not exceed 5 steps according to the maximum motion speed calculation of the human body, the sampling rate of the sensors is set to be 25Hz, the step number calculated in 25 sampling points can not exceed 5 steps, because the sensors are only placed in one shoe, when the human body moves by 5 steps, one foot moves by 3 steps at most, the number of steps calculated in 1 second cannot exceed 3 steps. Therefore, the step number of the person who moves by at least more than 8 sampling points can be calculated, and the step number which is calculated due to errors is removed according to the rule, so that the step number is accurately calculated, and the specific steps are as follows:
and accurately calculating the step number, which comprises the following specific steps:
the method comprises the following steps: the acquired acceleration sensor data is sent to a mobile phone or a general single chip Microcomputer (MCU) in real time through the intelligent shoe;
step two: carrying out smooth filtering and Kalman filtering on the acquired data to make the waveform smoother and reduce errors;
step three: analyzing the data after smooth filtering to calculate the step number of the movement;
step four: cutting the data after Kalman filtering to obtain the waveform of each step, analyzing the characteristic value of the waveform, and confirming the motion state of the human body;
step five: the step numbers of various motion states of the human body can be analyzed through the fusion of the two data;
by analyzing the acceleration waveform of the human body during movement, the waveform of each movement has corresponding periodicity, and the waveforms of different movements in one period are different, so that each movement can be distinguished by distinguishing the characteristic values of the waveforms.
Preferably, the intelligent shoe in the first step of the step-recording and behavior recognition method based on the acceleration sensor is a shoe which can collect acceleration information in the human body movement process and send the acceleration information to a mobile phone in real time through Bluetooth or transmit the acceleration information to a general single chip Microcomputer (MCU).
Further, preferably, in the third step of the step recording and behavior identification method based on the acceleration sensor, after the acceleration values of the X, Y, and Z axes are obtained at the mobile phone terminal, the original data is copied into two parts, one part is filtered through smoothing, and the other part is filtered through kalman filtering to eliminate errors.
Further, preferably, in step four of the acceleration sensor-based step recording and behavior recognition method, the smoothing filtering is performed by using a simple averaging method, and in order to obtain an average brightness value of neighboring pixel points, the smoothed filtered data is used to calculate the number of steps of the motion.
Further, preferably, the data after kalman filtering in step five of the acceleration sensor-based step recording and behavior recognition method shows that different acceleration values of each moving human body reflect the intensity of the movement to some extent, so that the magnitude of the acceleration can be used to distinguish walking, fast walking and running, and the calculation formula of the acceleration is as follows:
wherein a is the resultant acceleration, ax，ay，azRespectively calculating the average value a 'of the combined acceleration in a period for the accelerations of the X axis, the Y axis and the Z axis measured by the sensor, and distinguishing walking, fast walking and running according to the magnitude of the a'; after walking and running are distinguished, further analysis is carried out on the basis, the characteristic value of the waveform is extracted, and the waveform is classified according to the characteristic value, so that the motion state of the human body can be confirmed;
extracting characteristic values, namely calculating the average value, the average difference, the quartering difference, the dispersion coefficient, the skewness coefficient and the like of the waveform in a period as the characteristic values of the waveform;
various motions can be accurately distinguished by determining a reasonable threshold value through sampling and counting actual motions.
After the scheme is adopted, the waveform is smoother and the error is reduced by means of smooth filtering and Kalman filtering, so that the system can accurately record steps in real time; meanwhile, various motions of people can be accurately distinguished; secondly, the step numbers of various motions can be calculated in real time; thirdly, the calculation requirement on the system is not very high, and the calculation can be finished by a mobile phone or a general single chip Microcomputer (MCU) which is configured commonly.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
The present invention will be described in detail below with reference to the accompanying drawings so that the above advantages of the present invention will be more apparent. Wherein,
FIG. 1 is a schematic diagram of acceleration along the X-axis during movement in an acceleration sensor-based pace and behavior recognition method of the present invention;
FIG. 2 is a waveform diagram of acceleration during walking after Kalman filtering based on the step counting and behavior identification method of the acceleration sensor;
FIG. 3 is a schematic diagram showing the acceleration waveform of the forefoot during running on the ground after Kalman filtering according to the step-counting and behavior recognition method of the acceleration sensor; (ii) a
FIG. 4 is a schematic diagram of a waveform of acceleration while ascending stairs after Kalman filtering based on the step-counting and behavior recognition method of the acceleration sensor according to the present invention;
FIG. 5 is a schematic diagram of a waveform of acceleration during descent of a staircase after Kalman filtering based on the step-counting and behavior recognition method of the acceleration sensor according to the present invention;
FIG. 6 is a schematic diagram of the acceleration waveform during full sole touchdown running after Kalman filtering based on the step-counting and behavior recognition method of the acceleration sensor according to the present invention;
FIG. 7 is a schematic diagram of a waveform of acceleration during a heel strike run after Kalman filtering based on a step-counting and behavior recognition method of an acceleration sensor according to the present invention;
FIG. 8 is a waveform diagram of fast-walking acceleration after Kalman filtering according to the method for step counting and behavior recognition based on acceleration sensors;
FIG. 9 is a flow chart of a method for acceleration sensor based pacing and behavior recognition according to the present invention.
Detailed Description
The following detailed description will be made with reference to fig. 1 to 9 and examples, so that how to apply the technical means to solve the technical problems and achieve the technical effects can be fully understood and implemented. It should be noted that, as long as there is no conflict, the embodiments and the features of the embodiments of the present invention may be combined with each other, and the technical solutions formed are within the scope of the present invention.
After a mobile phone end acquires acceleration values of X, Y and Z axes, data errors are large due to the fact that sampling rate, measurement noise and the like can have certain influences on data of the sensor, filtering needs to be conducted on original data, the original data are copied into two parts, one part is filtered through smoothing, and the other part is filtered through Kalman filtering to eliminate errors.
The smoothing filtering in the spatial domain is generally performed by a simple averaging method, that is, an average luminance value of neighboring pixel points is obtained. The size of the neighborhood is directly related to the smoothing effect, the larger the neighborhood is, the better the smoothing effect is, but the larger the neighborhood is, the larger the edge information loss is due to the fact that the neighborhood is too large, so that the output image becomes fuzzy, and the waveform has certain hysteresis due to smooth filtering, and the human motion posture cannot be reflected in real time. However, he can distinguish the number of steps of the human body movement well, and the data after smooth filtering can be used for calculating the number of steps of the movement.
The sensor placement mode adopted by the algorithm is as follows: the forward direction of the foot is the positive direction of an X axis, the leftward direction is the positive direction of a Y axis, and the foot lifting direction is the negative direction of a Z axis. When the human body moves, the displacement in the X-axis direction is large, and the acceleration change is obvious and has obvious periodicity (as shown in figure 1). The number of steps of the movement can be identified as long as a reasonable threshold is set.
The acceleration of the human body is larger than a threshold value (set as Ax) when the acceleration of the X axis is in motion by collecting a large number of data samples, when the acceleration is changed from less than Ax to more than Ax, then the acceleration is changed from more than Ax to correspond to the foot lifting and falling actions of the human body, namely the human body is identified to move by one step, because some errors exist in the sensors, the acceleration of a plurality of points can be caused to wander around Ax in one step, the condition of more steps can be calculated by the above mode, in order to eliminate the condition, the number of steps of the human body moving in one second can not exceed 5 steps according to the maximum motion speed calculation of the human body, the sampling rate of the sensors is set to be 25Hz, the number of steps calculated in 25 sampling points can not exceed 5 steps, because the sensors are only placed in one shoe, when the human body moves by 5 steps, one foot moves by 3 steps at most, the number of steps calculated in 1 second cannot exceed 3 steps. Therefore, the number of sampling points which are at least more than 8 when the person moves for one step can be calculated, and the step number which is calculated due to error is removed according to the rule, so that the purpose of accurately calculating the step number is achieved.
The Kalman filtering inputs and outputs observation data through the system, and the algorithm for carrying out optimal estimation on the system state not only ensures the information of the waveform, but also makes the waveform very smooth, thereby providing convenience for extracting the characteristic value of the waveform. It is possible to distinguish the motion state according to the waveform characteristic value. The waveform after kalman filtering is shown in the following figure:
by analyzing the acceleration waveforms of the common movements, it can be seen that the waveform of each movement has periodicity, and the waveforms of different movements in one period are different, so that each movement can be distinguished by distinguishing the characteristic values of the waveforms.
The magnitude of the acceleration reflects to some extent the intensity of the movement, so that the magnitude of the acceleration can be used to distinguish walking, fast walking and running. The company for calculating the resultant acceleration is as follows:
a=√(a_x^2+a_y^2+a_z^2 )
a, the resultant acceleration a _ X, a _ Y and a _ Z are respectively the acceleration of X axis, Y axis and Z axis measured by the sensor
And (4) calculating the average value a 'of the combined acceleration in a period, and distinguishing walking, fast walking and running according to the magnitude of a'. After walking and running are distinguished, the characteristic values of the waveforms are extracted through further analysis on the basis, and the waveforms are classified according to the characteristic values, so that the motion state of the human body can be confirmed. Regarding the feature value extraction, the average value, the average difference, the quartering difference, the dispersion coefficient, the skewness coefficient and the like of the waveform in one period are calculated to be used as the feature value of the waveform. Various motions can be accurately distinguished by determining a reasonable threshold value through sampling and counting actual motions.
After the scheme is adopted, the waveform is smoother and the error is reduced by means of smooth filtering and Kalman filtering, so that the system can accurately record steps in real time; meanwhile, various motions of people can be accurately distinguished; secondly, the step numbers of various motions can be calculated in real time; thirdly, the calculation requirement on the system is not very high, and the calculation can be finished by a mobile phone or a general single chip Microcomputer (MCU) which is configured commonly.

Claims (5)

1. A step recording and behavior identification method based on an acceleration sensor is characterized in that the sensor placement mode adopted by the method is as follows: the forward direction of the feet is the positive direction of an X axis, the leftward direction is the positive direction of a Y axis, and the foot lifting direction is the negative direction of a Z axis; therefore, the number of steps of the movement can be identified as long as a reasonable threshold value is set;
the acceleration of the human body is larger than a threshold value Ax when the acceleration of an X axis is in motion by collecting a large number of data samples, when the acceleration is changed from less than Ax to more than Ax, then the acceleration is changed from more than Ax to correspond to the actions of lifting and falling the foot of the human body, namely the human body is recognized to move by one step, because some errors exist in the sensors, the acceleration of a plurality of points possibly appears in one step and is wandering around Ax, the condition of a plurality of steps can be calculated in a corresponding mode, in order to eliminate the condition, the step number of the human body moving in one second can not exceed 5 steps according to the maximum motion speed calculation of the human body, the sampling rate of the sensors is set to be 25Hz, the step number calculated in 25 sampling points can not exceed 5 steps, because the sensors are only placed in one shoe, when the human body moves by 5 steps, one foot moves by 3 steps at most, therefore, the calculated step number within 1 second cannot exceed 3 steps, so that the step number of the person who moves by one step is calculated to be at least more than 8 sampling points, and the step number which is calculated more due to errors is removed according to the rule, so that the step number is accurately calculated, and the specific steps are as follows:
the method comprises the following steps: the acquired acceleration is transmitted to a mobile phone or a general single chip Microcomputer (MCU) in real time through the intelligent shoe;
step two: carrying out smooth filtering and Kalman filtering on the acquired data to make the waveform smoother and reduce errors;
step three: analyzing the data after smooth filtering to calculate the step number of the movement;
step four: cutting the data after Kalman filtering to obtain the waveform of each step, analyzing the characteristic value of the waveform, and confirming the motion state of the human body;
step five: the step numbers of various motion states of the human body can be analyzed through the fusion of the two data;
by analyzing the acceleration waveform of the human body during movement, the waveform of each movement has corresponding periodicity, and the waveforms of different movements in one period are different, so that each movement can be distinguished by distinguishing the characteristic values of the waveforms.
2. The method for step recording and behavior recognition based on the acceleration sensor according to claim 1, wherein the intelligent shoe in the first step is a shoe capable of collecting acceleration information of a human body in a motion process and sending the acceleration information to a mobile phone in real time through Bluetooth, or processing data on a general single chip Microcomputer (MCU).
3. The acceleration sensor-based step recording and behavior recognition method according to claim 1, wherein in the third step, after the processor obtains the acceleration values of the X, Y and Z axes, the original data is copied into two parts, one part is subjected to smoothing filtering, and the other part is subjected to kalman filtering to eliminate errors.
4. The acceleration-sensor-based pace and behavior recognition method of claim 1, wherein in step four, the smoothing filtering is performed by a simple averaging method, and the smoothed and filtered data is used to calculate the number of steps of the motion for obtaining the average brightness value of the neighboring primitive points.
5. The acceleration sensor-based pace and action recognition method of claim 1, wherein the kalman filtered data in the fifth step shows that the different acceleration values of each moving human body reflect the intensity of the movement to some extent, so that the magnitude of the acceleration can be used to distinguish walking, fast walking and running, and the calculation formula of the acceleration is as follows:
wherein a is the resultant acceleration, ax，ay，azRespectively calculating the average value a 'of the combined acceleration in a period for the accelerations of the X axis, the Y axis and the Z axis measured by the sensor, and distinguishing walking, fast walking and running according to the magnitude of the a'; after walking and running are distinguished, further analysis is carried out on the basis, the characteristic value of the waveform is extracted, and the waveform is classified according to the characteristic value, so that the motion state of the human body can be confirmed;
extracting characteristic values, namely calculating the average value, the average difference, the quartering difference, the dispersion coefficient, the skewness coefficient and the like of the waveform in a period as the characteristic values of the waveform;
various motions can be accurately distinguished by determining a reasonable threshold value through sampling and counting actual motions.
CN201610082327.7A 2016-02-05 2016-02-05 A kind of note step and Activity recognition method based on acceleration sensor Active CN105912142B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610082327.7A CN105912142B (en) 2016-02-05 2016-02-05 A kind of note step and Activity recognition method based on acceleration sensor

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610082327.7A CN105912142B (en) 2016-02-05 2016-02-05 A kind of note step and Activity recognition method based on acceleration sensor

Publications (2)

Publication Number Publication Date
CN105912142A true CN105912142A (en) 2016-08-31
CN105912142B CN105912142B (en) 2019-06-11

Family

ID=56744401

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610082327.7A Active CN105912142B (en) 2016-02-05 2016-02-05 A kind of note step and Activity recognition method based on acceleration sensor

Country Status (1)

Country Link
CN (1) CN105912142B (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106562508A (en) * 2016-10-19 2017-04-19 泉州迪特工业产品设计有限公司 Intelligent shoes for child and realization method therefor
CN106653058A (en) * 2016-10-28 2017-05-10 中国科学院计算技术研究所 Double-channel step detection method
CN106723612A (en) * 2016-11-21 2017-05-31 歌尔股份有限公司 A kind of step counting system
CN107277222A (en) * 2016-12-20 2017-10-20 浙江从泰网络科技有限公司 User behavior state judging method based on mobile phone built-in sensors
CN107343789A (en) * 2017-05-17 2017-11-14 浙江利尔达物联网技术有限公司 A kind of step motion recognition method based on 3-axis acceleration sensor
CN107747950A (en) * 2017-09-28 2018-03-02 上海惠芽信息技术有限公司 Step recording method and device
CN108680181A (en) * 2018-04-23 2018-10-19 Oppo广东移动通信有限公司 Wireless headset, step-recording method and Related product based on headset detection
CN109029492A (en) * 2018-10-12 2018-12-18 潍坊歌尔电子有限公司 A kind of step-recording method, device and wrist step counting equipment
CN109949543A (en) * 2019-04-18 2019-06-28 西安建筑科技大学 A kind of multifunctional shoe and the remotely intelligently monitoring method based on pressure sensitive
CN110313917A (en) * 2018-03-28 2019-10-11 财团法人交大思源基金会 It is the tumble sensor-based system and its method for judging benchmark with activities of daily life
CN114259720A (en) * 2020-09-15 2022-04-01 荣耀终端有限公司 Action recognition method and device, terminal equipment and motion monitoring system
CN114655224A (en) * 2022-03-21 2022-06-24 潍柴动力股份有限公司 Road gradient estimation method, electronic device and storage medium

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008262522A (en) * 2007-04-11 2008-10-30 Aichi Micro Intelligent Corp Pedometer
CN102278998A (en) * 2010-03-25 2011-12-14 精工电子有限公司 Electronic apparatus and program
CN102297701A (en) * 2010-06-22 2011-12-28 雅马哈株式会社 Pedometer
JP5176047B2 (en) * 2012-04-09 2013-04-03 アイチ・マイクロ・インテリジェント株式会社 Pedometer
CN103727954A (en) * 2013-12-27 2014-04-16 北京超思电子技术股份有限公司 Pedometer
US20140188431A1 (en) * 2012-11-01 2014-07-03 Hti Ip, Llc Method and system for determining whether steps have occurred
US20160001131A1 (en) * 2014-07-03 2016-01-07 Katarzyna Radecka Accurate Step Counting Pedometer for Children, Adults and Elderly

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008262522A (en) * 2007-04-11 2008-10-30 Aichi Micro Intelligent Corp Pedometer
CN102278998A (en) * 2010-03-25 2011-12-14 精工电子有限公司 Electronic apparatus and program
CN102297701A (en) * 2010-06-22 2011-12-28 雅马哈株式会社 Pedometer
JP5176047B2 (en) * 2012-04-09 2013-04-03 アイチ・マイクロ・インテリジェント株式会社 Pedometer
US20140188431A1 (en) * 2012-11-01 2014-07-03 Hti Ip, Llc Method and system for determining whether steps have occurred
CN103727954A (en) * 2013-12-27 2014-04-16 北京超思电子技术股份有限公司 Pedometer
US20160001131A1 (en) * 2014-07-03 2016-01-07 Katarzyna Radecka Accurate Step Counting Pedometer for Children, Adults and Elderly

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106562508A (en) * 2016-10-19 2017-04-19 泉州迪特工业产品设计有限公司 Intelligent shoes for child and realization method therefor
CN106653058B (en) * 2016-10-28 2020-03-17 中国科学院计算技术研究所 Dual-track-based step detection method
CN106653058A (en) * 2016-10-28 2017-05-10 中国科学院计算技术研究所 Double-channel step detection method
CN106723612A (en) * 2016-11-21 2017-05-31 歌尔股份有限公司 A kind of step counting system
CN107277222A (en) * 2016-12-20 2017-10-20 浙江从泰网络科技有限公司 User behavior state judging method based on mobile phone built-in sensors
CN107277222B (en) * 2016-12-20 2020-12-15 浙江斑智科技有限公司 User behavior state judgment method based on built-in sensor of mobile phone
CN107343789A (en) * 2017-05-17 2017-11-14 浙江利尔达物联网技术有限公司 A kind of step motion recognition method based on 3-axis acceleration sensor
CN107343789B (en) * 2017-05-17 2020-04-14 浙江利尔达物芯科技有限公司 Step motion identification method based on three-axis acceleration sensor
CN107747950A (en) * 2017-09-28 2018-03-02 上海惠芽信息技术有限公司 Step recording method and device
CN110313917A (en) * 2018-03-28 2019-10-11 财团法人交大思源基金会 It is the tumble sensor-based system and its method for judging benchmark with activities of daily life
CN110313917B (en) * 2018-03-28 2022-04-26 财团法人交大思源基金会 Fall-down sensing system and method using daily life movement as judgment reference
CN108680181A (en) * 2018-04-23 2018-10-19 Oppo广东移动通信有限公司 Wireless headset, step-recording method and Related product based on headset detection
CN109029492A (en) * 2018-10-12 2018-12-18 潍坊歌尔电子有限公司 A kind of step-recording method, device and wrist step counting equipment
CN109029492B (en) * 2018-10-12 2021-09-03 潍坊歌尔电子有限公司 Step counting method and device and wrist step counting equipment
CN109949543A (en) * 2019-04-18 2019-06-28 西安建筑科技大学 A kind of multifunctional shoe and the remotely intelligently monitoring method based on pressure sensitive
CN114259720A (en) * 2020-09-15 2022-04-01 荣耀终端有限公司 Action recognition method and device, terminal equipment and motion monitoring system
CN114655224A (en) * 2022-03-21 2022-06-24 潍柴动力股份有限公司 Road gradient estimation method, electronic device and storage medium

Also Published As

Publication number Publication date
CN105912142B (en) 2019-06-11

Similar Documents

Publication Publication Date Title
CN105912142B (en) A kind of note step and Activity recognition method based on acceleration sensor
CN106123911A (en) A kind of based on acceleration sensor with the step recording method of angular-rate sensor
CN109522793B (en) Method for detecting and identifying abnormal behaviors of multiple persons based on machine vision
Wang et al. Fall detection based on dual-channel feature integration
Jalal et al. Depth maps-based human segmentation and action recognition using full-body plus body color cues via recognizer engine
US10810414B2 (en) Movement monitoring system
US11047706B2 (en) Pedometer with accelerometer and foot motion distinguishing method
JP5986656B2 (en) Taekwondomeuse recognition and ascending screening device based on human skeleton using depth camera and method
CN113164098A (en) Human gait analysis system and method
CN106096662B (en) Human motion state identification based on acceleration transducer
CN103530894B (en) A kind of video object method for tracing based on multiple dimensioned piece of rarefaction representation and system thereof
KR101159962B1 (en) Facial Expression Recognition Interaction Method between Mobile Machine and Human
US20200034739A1 (en) Method and device for estimating user's physical condition
JP2012518236A (en) Method and system for gesture recognition
CN110738154A (en) pedestrian falling detection method based on human body posture estimation
CN108021888B (en) Fall detection method
CN108958482B (en) Similarity action recognition device and method based on convolutional neural network
CN102707806A (en) Motion recognition method based on acceleration sensor
KR101307984B1 (en) Method of robust main body parts estimation using descriptor and machine learning for pose recognition
CN111428666A (en) Intelligent family accompanying robot system and method based on rapid face detection
CN115346272A (en) Real-time tumble detection method based on depth image sequence
CN104331705B (en) Automatic detection method for gait cycle through fusion of spatiotemporal information
CN111144167A (en) Gait information identification optimization method, system and storage medium
WO2018014432A1 (en) Voice application triggering control method, device and terminal
CN107688828A (en) A kind of bus degree of crowding estimating and measuring method based on mobile phone sensor

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant