CN107773967B - Human motion characteristic analysis method based on smart watch data - Google Patents

Human motion characteristic analysis method based on smart watch data Download PDF

Info

Publication number
CN107773967B
CN107773967B CN201711015722.4A CN201711015722A CN107773967B CN 107773967 B CN107773967 B CN 107773967B CN 201711015722 A CN201711015722 A CN 201711015722A CN 107773967 B CN107773967 B CN 107773967B
Authority
CN
China
Prior art keywords
sampling
data
heart rate
calorie
state
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201711015722.4A
Other languages
Chinese (zh)
Other versions
CN107773967A (en
Inventor
赵露
康艳荣
龙源
邢桂东
郭丽莉
周冬林
楚川红
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.)
Institute of Forensic Science Ministry of Public Security PRC
Original Assignee
Institute of Forensic Science Ministry of Public Security PRC
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 Institute of Forensic Science Ministry of Public Security PRC filed Critical Institute of Forensic Science Ministry of Public Security PRC
Priority to CN201711015722.4A priority Critical patent/CN107773967B/en
Publication of CN107773967A publication Critical patent/CN107773967A/en
Application granted granted Critical
Publication of CN107773967B publication Critical patent/CN107773967B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B71/00—Games or sports accessories not covered in groups A63B1/00 - A63B69/00
    • A63B71/06—Indicating or scoring devices for games or players, or for other sports activities
    • A63B71/0619—Displays, user interfaces and indicating devices, specially adapted for sport equipment, e.g. display mounted on treadmills
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/024—Detecting, measuring or recording pulse rate or heart rate
    • 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
    • A61B5/681—Wristwatch-type devices
    • 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
    • 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
    • A61B5/7271—Specific aspects of physiological measurement analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/22—Indexing; Data structures therefor; Storage structures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2462—Approximate or statistical queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/28—Databases characterised by their database models, e.g. relational or object models
    • G06F16/284—Relational databases
    • 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
    • A63B2230/04—Measuring physiological parameters of the user heartbeat characteristics, e.g. ECG, blood pressure modulations
    • A63B2230/06—Measuring physiological parameters of the user heartbeat characteristics, e.g. ECG, blood pressure modulations heartbeat rate only
    • 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

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Theoretical Computer Science (AREA)
  • Physiology (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Databases & Information Systems (AREA)
  • Cardiology (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Psychiatry (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Signal Processing (AREA)
  • Software Systems (AREA)
  • Probability & Statistics with Applications (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Computational Linguistics (AREA)
  • Mathematical Physics (AREA)
  • Dentistry (AREA)
  • Fuzzy Systems (AREA)
  • Pulmonology (AREA)
  • Human Computer Interaction (AREA)
  • Physical Education & Sports Medicine (AREA)
  • Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

The invention relates to a human motion characteristic analysis method based on intelligent watch data, which is characterized by comprising the following steps of: 1) extracting and analyzing basic data and measurement data collected by the intelligent watch, wherein the measurement data comprises step number, calorie and heart rate; 2) dividing the measurement data in a certain measurement time period according to the step number change and the heart rate change to obtain the heart rate and calorie data of the sampling individual in the exercise state and normal state time periods respectively; 3) analyzing the human motion characteristic state based on the heart rate and calorie data of the motion state and the normal state time period to obtain the corresponding relation between the human motion state and the normal state and the change of the heart rate and the calorie. The method is simple and feasible, has accurate calculation result, and can be widely applied to the field of analyzing the human motion characteristics based on electronic data.

Description

Human motion characteristic analysis method based on smart watch data
Technical Field
The invention relates to the field of human motion characteristic analysis, in particular to a human motion characteristic analysis method based on intelligent watch data.
Background
As one type of wearable equipment, the intelligent watch has complete functions and convenient use, and plays an important role in aspects such as health monitoring. Two monitoring functions which are common in human activity characteristic monitoring of the smart watch are heart rate data monitoring and step counting data monitoring. In order to achieve stronger intelligence and more prominent user experience, various high-precision sensing devices including positioning, gravity, direction, acceleration and the like are widely applied to the smart watch system. A series of processes of dynamic capture and real-time capture of surrounding environment data information, data preprocessing, noise mutation error removal and dynamic movement space-time data storage are completed through a physical device in a specific frequency setting mode or a certain event triggering mode, and accurate human activity characteristic monitoring is achieved. The accurate monitoring of the human body activities plays an important role in the relevant fields of old people falling prevention, children loss prevention, young people movement, body building and the like, and is more and more widely concerned and applied in social life.
However, since smart watches are electronic products which are explosively increased in recent two years, there is no method for extracting and analyzing electronic data in smart watches, and there is no method for analyzing character motion characteristics based on data of such smart watches.
Disclosure of Invention
In view of the above problems, an object of the present invention is to provide a method for analyzing human motion characteristics based on smart watch data, which obtains a corresponding relationship between human motion characteristic states and heart rate and calorie variation by analyzing measurement data extracted from a smart watch.
In order to achieve the purpose, the invention adopts the following technical scheme: a human motion characteristic analysis method based on smart watch data is characterized by comprising the following steps: 1) extracting and analyzing basic data and measurement data of each sampling individual collected by the intelligent watch, wherein the measurement data comprises step number, calorie and heart rate; 2) dividing the measurement data in any measurement time period according to the step number change and the heart rate change to obtain the heart rate and calorie data of the sampling individual in the exercise state and normal state time periods in the measurement time period; 3) analyzing the human motion characteristic state based on the heart rate and calorie data of the motion state and the normal state time period to obtain the corresponding relation between the human motion state and the normal state and the change of the heart rate and the calorie.
In the step 1), the method for extracting and analyzing the basic data and the step number, the heart rate and the calorie measurement data collected by the intelligent watch comprises the following steps: 1.1) adopting an intelligent watch to collect basic data of each sampling individual, and simultaneously periodically collecting measurement data of each sampling individual; 1.2) periodically uploading basic data and measurement data acquired by each intelligent watch to a mobile terminal for storage; 1.3) extracting the basic data and the measurement data stored in each mobile terminal to a computer terminal for data processing.
In the step 1.1), the method for collecting basic data and measurement data by using the smart watch comprises the following steps: 1.1.1) determining the type of the intelligent watch and a sampling individual according to the experiment requirement; 1.1.2) determining the type of basic data and measurement data to be collected according to the determined type of the intelligent watch and the sampling individual; 1.1.3) the intelligent watch periodically collects the relevant measurement data of each sampling individual in normal daily activities and uploads the data to the mobile terminal.
In the step 2), the method for dividing the measurement data to obtain the heart rate and calorie data of the sampling individual in the exercise state and normal state time periods respectively comprises the following steps: 2.1) carrying out sum operation on the step numbers corresponding to all adjacent sampling nodes according to the time sequence to obtain a time period corresponding to the sampling node of which the sum operation result meets a preset step number change value; 2.2) carrying out sum operation on the heart rate data corresponding to all adjacent sampling nodes according to the time sequence to obtain a time period corresponding to the sampling node of which the sum operation result meets a preset heart rate change value; and 2.3) obtaining the time period of each sampling individual in the motion state according to the time period corresponding to the sampling node meeting the preset step number change value and the preset heart rate change value and the step number and the heart rate change corresponding to different motion types through the obtained sum operation, wherein other time periods are the time periods in the normal state.
In the step 2.1), the obtained sample point data of the sampling node satisfies the following formula:
Figure GDA0001981227940000021
wherein s isiRepresenting the number of steps, s, corresponding to the sampling node ii-1Represents the number of steps, S, corresponding to the sampling node i-1iRepresenting the sum of the steps between the sampling node i and the sampling node i-1; s0Indicating a motion condition preset value.
In the step 2.2), the obtained sample point data of the sampling node satisfies the following formula:
Figure GDA0001981227940000022
wherein h isiRepresenting the heart rate, h, corresponding to the sampling node ii-1Representing the heart rate, S, corresponding to the sampling node i-1iRepresenting the sum of the heart rates, H, between sampling node i and sampling node i-10Is a preset heart rate variation value.
In the step 3), the method for obtaining the corresponding relation between the human body motion state and the normal state and the heart rate and the calorie change comprises the following steps: 3.1) carrying out statistical analysis on the measurement data of all the sampling individuals to obtain the activity state analysis results of all the sampling individuals, including the average step number, the calorie value and the heart rate value of all the sampling individuals; 3.2) randomly extracting a plurality of sampling individuals, and analyzing the step number, the calorie and the heart rate change corresponding to the exercise state and the normal state before and after the exercise state in any measurement time period by adopting a nonparametric inspection method to obtain the difference characteristics of the step number, the calorie and the heart rate in the exercise state and the normal state; and 3.3) guiding and judging whether each sampling individual is in the exercise state or the normal state according to the difference characteristics of the step number, the calorie and the heart rate in the exercise state and the normal state.
In the step 3.2), the method for analyzing the data of each sampling individual by adopting a non-inspection method comprises the following steps:
3.2.1) assume that the measurement data of any sampling individual, i.e. the overall distribution of the sample records, is symmetric about the point θ, and at the same time assume that the overall distribution of the sample records is symmetric, thereby obtaining the test hypothesis H0：θ＝θ0Wherein, theta0To check the overall median;
3.2.2) calculating the difference z between the sample point corresponding to each sampling node and the median of the total inspection under the assumption of the inspection in the step 3.2.1)iThe calculation formula is as follows:
zi＝xi-θ0，i＝1,2,…,n，
wherein i is the sampling node ordinal number of the sampling unit, xiFor the sample point value corresponding to the sampling node, ziIs a sample point to theta0The distance of (d);
3.2.3) taking the absolute value of the difference obtained in the step 3.2.2), sequencing the sample points corresponding to each sampling node according to the absolute value of the difference, and calculating to obtain the rank R corresponding to each sample pointi；
3.2.4) comparing the difference z between each sample point and the median of the test population obtained in step 3.2.2)iAnd the rank R of each sample point obtained in step 3.2.3)iCalculating to obtain initial test statistic W+And W-；
Figure GDA0001981227940000031
Wherein
Figure GDA0001981227940000032
Figure GDA0001981227940000033
Wherein
Figure GDA0001981227940000034
3.2.5) judging whether the assumption is established in the step 3.2.1) according to the obtained initial test statistic, and further obtaining the total actual test statistic of the sample record of the sampling individual;
3.2.6) obtaining the difference result of the sample data of the sampling individual according to the obtained actual inspection statistic value.
Due to the adoption of the technical scheme, the invention has the following advantages: 1. according to the invention, the measurement data of the intelligent watch is detailed and accurate by extracting and analyzing the basic measurement data of the intelligent watch, and the method can be used in the field of character characteristic analysis. 2. The invention is based on the basic measurement data of the intelligent watch, carries out statistical analysis on the relationship between the heart rate and the calorie change under the exercise state and the non-exercise state (normal state) of the human body, has high accuracy of the analysis method, and can be used for judging the exercise state of the human body according to the heart rate and the calorie change. The invention can be widely applied to the field of evidence obtaining and analysis of electronic data.
Drawings
Fig. 1(a), (b) are stored smart watch data of a mobile phone in an embodiment of the present invention;
FIG. 2 is a graph showing the change in the number of steps of a sampled individual 1 within 48 hours according to an example of the present invention;
FIG. 3 is a graph of calorie consumption by a sampled individual 1 over a 48 hour period in an example of the present invention;
FIG. 4 shows heart rate variability of a sampled individual 1 over 48 hours in an example of the present invention;
FIG. 5 is a graph showing statistics of the number of pre-, mid-, and post-exercise steps of the sampled subject 2 according to an embodiment of the present invention;
FIG. 6 is a graph of heart rate statistics before, during, and after exercise for a sampled individual 2 in an embodiment of the present invention;
FIG. 7 is a graph showing statistics of the number of pre-, mid-, and post-exercise steps of a sampled individual 25 in accordance with an embodiment of the present invention;
FIG. 8 shows the heart rate statistics of a sampled individual 25 before, during, and after exercise in an embodiment of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and examples.
The invention provides a human motion characteristic analysis method based on intelligent watch data, which comprises the following steps of:
1) extracting and analyzing basic data and measurement data collected by the intelligent watch, wherein the measurement data comprises step number, calorie and heart rate;
2) dividing the measurement data in a certain measurement time period according to the step number change and the heart rate change to obtain the heart rate and calorie data of the sampling individual in the exercise state and normal state time periods respectively;
3) analyzing the human motion characteristic state based on the heart rate and calorie data of the motion state and the normal state time period to obtain the corresponding relation between the human motion state and the normal state and the change of the heart rate and the calorie.
In the step 1), the method for extracting and analyzing the basic data and the step number, the heart rate and the calorie measurement data collected by the intelligent watch comprises the following steps:
1.1) adopting an intelligent watch to collect basic data of each sampling individual, and simultaneously periodically collecting measurement data of each sampling individual;
1.2) periodically uploading basic data and measurement data acquired by each intelligent watch to a mobile terminal for storage;
1.3) extracting the basic data and the measurement data stored in each mobile terminal to a computer terminal for data processing.
In the step 1.1), the method for collecting basic data and measurement data by using the smart watch comprises the following steps:
1.1.1) determining the type of the intelligent watch and sampling individuals according to experimental requirements.
Because the intelligent watch has rich functions, different user groups exist according to different functions, for example, the child intelligent watch is suitable for primary and middle school students and children, and is mainly used for preventing loss; the intelligent watch for the old is suitable for the old with inconvenient movement or diseases, and mainly aims at preventing falling and health monitoring; the common intelligent watch is suitable for young people and has the functions of motion tracking, health monitoring, remote control, watch payment and the like. According to the invention, a common intelligent watch is adopted to collect experimental data, and a sampling individual is determined as a young person.
1.1.2) determining the type of basic data and measurement data to be collected according to the determined type of the intelligent watch and the sampling individual.
The basic measurement data types involved in the present invention include two types: firstly, sampling the basic data of an individual, and secondly, sampling the measurement data of the individual; wherein the basic data of the sampled individuals comprises the sex, the age, the height, the weight, the occupation and the total number of the sampled individuals; the measurement data of the sampled individuals includes the number of steps, calorie consumption, and heart rate.
1.1.3) as shown in fig. 1(a) and (b), the smart watch periodically collects the relevant measurement data of each sampling individual in normal daily activities and uploads the data to the mobile terminal.
The intelligent watch for the experiment is worn by the sampling individual to carry out normal daily movement, work, study, sleep and the like, and the intelligent watch uninterruptedly carries out basic measurement data acquisition in the process of wearing the sampling individual. In order to facilitate the comparison of experimental results, the sampled individuals manually record various states of the individuals in the experimental process, such as sleeping time, exercise types (such as running and body building) and no exercise time (such as sitting and learning).
In the step 1.3), the method for extracting and analyzing the measurement data collected in the smart watch includes the following steps:
1.3.1) connecting the mobile terminal with the measurement data with the computer terminal to ensure that the computer terminal is provided with a driving program and ensure complete data transmission;
1.3.2) transmitting basic measurement data in the mobile terminal to a computer terminal in an Android data file backup mode by adopting the existing professional data extraction tool;
1.3.3) analyzing the imported basic measurement data by adopting the existing SQLite database analysis tool;
1.3.4) carrying out time conversion on the basic measurement data analyzed by the database, namely converting the UNIX timestamp form analyzed by the database into UTC time;
1.3.5) importing the data after time conversion into an EXCEL table, and selecting data of different sampling individuals and different time intervals for different experiments to analyze. Results processing and descriptive Statistics of the data were performed by IBM SPSS Statistics 22.0, followed by results discussion.
In the step 2), the method for dividing the measurement data to obtain the heart rate and calorie data of the sampling individual in the exercise state and the normal state time period respectively comprises the following steps:
and 2.1) carrying out sum operation on the step numbers corresponding to all adjacent sampling nodes according to the time sequence to obtain a time period corresponding to the sampling node of which the sum operation result meets the preset step number change value.
The sampled data should satisfy the following equation:
Figure GDA0001981227940000051
wherein s isiRepresenting the number of steps, s, corresponding to the sampling node ii-1Represents the number of steps, S, corresponding to the sampling node i-1iRepresenting the sum of the steps between the sampling node i and the sampling node i-1; s0Representing a preset step number change value. The preset step change value is determined according to the acquisition frequency of the intelligent watch and the average number of steps per minute of an adult, and the value is 1000 in the invention. According to statistical data, the average adult walks between 60 and 100 steps per minute under normal conditions, and generally, the exercise time is considered to last for more than 10 minutes, the intelligent watch records the number of steps every 5 minutes, namely, the number of steps corresponding to each sampling node exceeds 500, and the average pace speed of the wearer within 5 minutes exceeds 100 steps per minute, namely, the wearer is in an exercise state. Thus, if two recordings exceed 1000, the wearer is considered to be in motion within the ten minutes.
And 2.2) carrying out sum operation on the heart rate data corresponding to all adjacent sampling nodes according to the time sequence to obtain a time period corresponding to the sampling node of which the sum operation result meets a preset heart rate change value.
The sampled data should satisfy the following equation:
Figure GDA0001981227940000061
wherein h isiRepresenting the heart rate, h, corresponding to the sampling node ii-1Representing the heart rate, S, corresponding to the sampling node i-1iRepresenting the sum of the heart rates, H, between sampling node i and sampling node i-10Is a preset heart rate variation value. The preset heart rate change value is determined according to the statistical result of the adult exercise state heart rate value and the acquisition frequency of the data of the intelligent watch, the statistical result of the adult exercise state heart rate value adopted in the invention is that the heart rate value is more than 100 for ten minutes, and the intelligent watch counts the data every 5 minutes, so that the invention H is characterized in that0A value of 200, i.e. a sum of two recordings of more than 200, indicates that the sampled individual enters a state of motion.
And 2.3) obtaining the time period of each sampling individual in the motion state according to the time period corresponding to the sampling node meeting the preset step number change value and the preset heart rate change value and the step number and the heart rate change corresponding to different motion types through the obtained sum operation, wherein other time periods are the time periods in the normal state.
M for motion state identification assumed to correspond to any sampling nodei＝Si+HiIndicates that there is
Figure GDA0001981227940000062
Considering that the types of sports are aerobic sports and anaerobic sports, the aerobic sports are mainly represented by running, fast walking, badminton and other sports, and the heart rate is improved while the step number is rapidly increased on the characteristic data of human body sports; while anaerobic exercise is represented by strength training, such exercise is characterized by insignificant rate of step growth, but significant changes in heart rate. Therefore, the exercise state and the normal state are distinguished according to the parameters of the step number change and the heart rate change. Then for any sampling node i, the motion status can be determined as shown in table 1 below:
TABLE 1 motion status identification and status relationship table
Si Hi Mi＝Si+Hi State determination
0 0 0 Exercise of sports
0 1 1 Exercise of sports
1 0 1 Exercise of sports
1 1 2 Is normal
Let T be all M of a sampling individuali<Time t of 2 hoursiIn consideration of the possibility of intermittent rest of the sampled individuals during a period of exercise, it is necessary to compare the exercise time intervals [ M ] recorded by the sampled individualsa,Mb](a<b) Then for all [ Ma,Mb](a<b) Satisfies M in the intervalj<2(a is less than or equal to j is less than or equal to b and a<b) At a time point tjE T, which is the time period of the sampling individual in motion state.
According to the analysis and judgment result of the recorded data of the intelligent watch, the data statistics and analysis are assisted, and the data statistics and analysis can be used as a basis for judging whether the movement recording time of the sampling individual is accurate or not.
In the step 3), the method for obtaining the corresponding relationship between the human body motion state and the normal state and the heart rate and the calorie change comprises the following steps:
3.1) carrying out statistical analysis on the measurement data of all the sampling individuals to obtain the activity state analysis results of all the sampling individuals, including the average step number, the calorie value and the heart rate value of all the sampling individuals;
and 3.2) randomly extracting a plurality of sampling individuals, and analyzing the step number, the calorie and the heart rate change corresponding to the exercise state and the front and back normal states of the exercise state of each sampling individual in a certain measurement time period by adopting a non-parameter detection method to obtain the difference characteristics of the step number, the calorie and the heart rate in the exercise state and the normal state.
And 3.3) guiding and judging whether each sampling individual is in the exercise state or the normal state according to the difference characteristics of the step number, the calorie and the heart rate in the exercise state and the normal state.
In the step 3.2), the method for analyzing the data of each sampling individual by adopting a non-inspection method comprises the following steps:
3.2.1) assuming that the total distribution of the sample records of the step number in the measurement data corresponding to a certain sampling individual is symmetrical about a point theta, and simultaneously assuming that the total distribution of the sample records is symmetrical, further obtaining a test hypothesis H0：θ＝θ0Wherein, theta0To check the overall median.
3.2.2) calculating the difference z between the sample point corresponding to each sampling node and the median of the total inspection under the assumption of the inspection in the step 3.2.1)iThe calculation formula is as follows:
zi＝xi-θ0，i＝1,2,…,n
wherein i is the sampling node ordinal number of the sampling unit, xiFor the sample point value corresponding to the sampling node, ziIs a sample point to theta0The distance of (c).
3.2.3) taking the absolute value of the difference obtained in the step 3.2.2), sequencing the sample points corresponding to each sampling node according to the absolute value of the difference, and calculating to obtain the rank R corresponding to each sample pointi. The calculation of the rank is an existing method, and the present invention is not described herein again.
3.2.4) comparing the difference z between each sample point and the median of the test population obtained in step 3.2.2)iAnd the rank R of each sample point obtained in step 3.2.3)iCalculating to obtain initial test statistic W+And W-。
Figure GDA0001981227940000071
Wherein
Figure GDA0001981227940000072
Figure GDA0001981227940000073
Wherein
Figure GDA0001981227940000074
3.2.5) judging whether the assumption is established in the step 3.2.1) according to the obtained initial test statistic, and further obtaining the total actual test statistic result of the sample record of the sampling individual.
If two initial test statistics W obtained in step 3.2.4)+And W-If the values are not equal or even differ greatly, the assumption of step 3.2.1) is not valid, and the actual test statistic result W is min (W)+,W-)。
3.2.6) obtaining the difference result of the sample record total of the sampling individuals according to the obtained test statistic value.
If n is very large to be approximated by a normal: and obtaining a value of a normal random variable Z related to W, and then checking a normal distribution table to obtain a P value representing the difference result. The data interpretation of the P values is shown in table 2 below:
TABLE 2 Difference results P values and interpretations
P value Probability of coincidence For testing hypothesis Statistical significance of
P>0.05 The probability of occurrence of the accident is more than 5 percent Fail to deny test hypothesis Two groups of differences have no significant meaning
P<0.05 The probability of occurence is less than 5 percent Can negate the test hypothesis Two groups of differences have significant meaning
P<0.01 The probability of occurence is less than 1% Can negate the test hypothesis The difference between the two is very significant
Example one
The intelligent watch adopted in the embodiment is a Moto 360 second-generation intelligent watch, the fashionable meta-watch disc appearance and the Android operating system provide rich functions, so that the intelligent watch becomes a sports intelligent watch popular in two years, the heart rate and the step recording function of the intelligent watch pass professional auditing, and a large amount of sports and health data can be generated and stored in the wearing process of a user, so that the intelligent watch is suitable for being used as a data acquisition tool in the embodiment. In this example, 25 young people between 20 and 30 years old were selected as sampling individuals, and experimental data was measured, and statistical information of the sampling individuals is shown in table 3 below.
TABLE 3 introduction of individual cases sampled
Sex Number of people Age (age) Height/cm Body weight/kg Occupation of the world
For male 12 25.5±2.3 174.5±5.8 73.5±8.4 Student's desk
Woman 13 25.7±1.0 165.6±7.4 54.5±8.2 Student's desk
The population of 25 sampled individuals was first counted in the order of experimental measurements, and the statistics of experimental data are shown in table 4:
TABLE 4 statistical table of the population of sampled individuals
Figure GDA0001981227940000081
Figure GDA0001981227940000091
The general situation of 25 sampled individuals can be understood approximately from table 2: the experimental period of each sampling individual is about one week, and the experimental design is met. For young people of 20-30 years old, the walking steps per day are about 2000-10000, and the male and female rules are not obvious; boys have higher overall caloric expenditure than girls; the average heart rate is substantially the same for both men and women.
As shown in fig. 2 to 4, the data of the number of steps, calories and heart rate of the sampled individual 1 in 48 hours clearly reflects the activity status of the sampled individual 1.
In conclusion, the smart watch has complete data storage, comprises a plurality of data such as step number, calorie and heart rate, automatically detects and records data every 5 minutes, has sufficient data volume, and can reflect the activity state of a user in a certain range. Day time: the step number changes along with the activity of the individual, the step number increases when the individual walks, and the step number is 0 when the individual is still; the calorie consumption value fluctuates along with the fluctuation of the calorie consumption value, and the calorie consumption value is slightly different under the influence of the physical condition of an individual; the heart rate has a wide variation range and changes along with the variation of the activity state. At night: the step number is 0; the calorie consumption trend and the heart rate fluctuation tend to be in a steady state from an unsteady state, and the numerical value is correspondingly reduced. Therefore, the measurement data of the smart watch can be used as the data for researching the character motion state characteristics.
In the experimental process, no heavy-load exercise, alcohol abuse and other adverse stimuli exist, and the sampled individuals perform data measurement with the daily life state as the standard. The experimental records of the sampling individual 2 and the sampling individual 25 show that both people have habits of regular body-building and exercise, so that the exercise state and the normal state of the sampling individual are researched in a related manner so as to know the data rule of the young people with the exercise habits.
The sampling individual 2 keeps regular body-building exercise for a long time, and the experimental record shows that the sampling individual normally carries out body-building exercise for 2 hours every day, mainly strength training. After the experiment is finished, data time periods of the sampling individual before, during and after the body-building exercise are respectively intercepted for analysis and research, the related data index changes are shown in table 5, the results are expressed by M +/-SD, the non-parameter test result of the matched sample is that the P is more than 0.01 and less than 0.05, and the P is more significant difference and less than 0.01.
TABLE 5 number of steps, calories, heart rate changes before, during, and after exercise
Status of state Number of steps Calories of Heart rate (times/min)
Before exercise 19±75.36 13±2.99 77±11.98
In motion 189±168.67 18±2.81 91±16.40
After exercise 32±102.05 14±2.00 82±12.30
As can be seen from Table 5: the step number, the calorie consumption and the heart rate are obviously changed in the exercise process, and compared with the step number, the calorie and the heart rate which are quickly increased before and after the exercise, the step number, the calorie and the heart rate are greatly changed. The non-parametric test result of the matched sample shows that: the number of steps in the exercise is compared with the number of steps before the exercise, and P is less than 0.01; the number of steps after exercise is compared with the number of steps in exercise, and P is less than 0.01; calorie in exercise compared to calorie before exercise P < 0.01; calorie after exercise compared with calorie in exercise P < 0.01; the heart rate during exercise is compared with the heart rate before exercise by P < 0.01; the heart rate after exercise is compared with the exercise center rate P <0.01, namely the steps, calories and heart rate before and after exercise are very different from those in exercise.
Under normal conditions, the average adult walks between 60-100 steps per minute, so that the step count is divided into 3 categories of 0 step, 0-100 steps and more than 100 steps for step count statistics, as shown in fig. 5; as can be seen from the overall measurement data, the quartiles of the heart rate values of the sampled individuals are 64, 73 and 86, so that the heart rate changes before, during and after exercise are classified into 5 categories of 64 or less, 64 to 73, 73 to 86, 86 to 100 and 100 or more for heart rate statistics, as shown in FIG. 6.
As can be seen from fig. 5, the number of steps before and after exercise: mainly takes 0 step as the main, which accounts for about 85 percent, and also takes 0 to 100 steps and more than 100 steps, but the proportion is very small; number of steps in exercise: almost none is obtained in step 0, and the number of steps 0-100 and 100 is at most, and each of the steps accounts for about 50%. The statistics are consistent with the daily rules of fitness exercise, consistent with the description of the course of the exercise (walk to gym-sport-go-dorm) for the sampled individual.
As can be seen from fig. 6, the pre-and post-exercise heart rate values: most of the numerical values are below 100 and account for more than 90 percent; heart rate in motion value: the value is at most 100% or more, and about 35% or less. It can be seen that the heart rate variation of the body-building exercise is characterized by the maximum value, the minimum value, the obvious variation and the stable intermediate value.
The sampling individual 25 is a badminton fan and always has a good habit of playing badminton every week, and experimental records show that the sampling individual normally plays badminton every week for 2 hours. After the experiment is finished, data time periods of the object before, during and after the body-building exercise are respectively intercepted for analysis and research, the related data index changes are shown in table 6, the results are expressed by M +/-SD, the matched sample nonparametric test result is that the difference between 0.01 and P <0.05 is a significant difference, and the difference between P <0.01 is an extremely significant difference.
TABLE 6 variation of step number, calorie, and heart rate before, during, and after exercise
Status of state Number of steps Calories of Heart rate (times/min)
Before exercise 2.17±10.62 9.92±2.36 71.83±9.56
In motion 494.42±92.37 22.88±4.37 113.92±19.36
After exercise 103.13±172.46 8.83±2.93 92.04±10.77
As can be seen from Table 6: the step number, the calorie consumption and the heart rate are obviously changed in the exercise process, and compared with the step number, the calorie and the heart rate values which are rapidly increased before and after the exercise, the numerical value is changed greatly. The non-parametric test result of the matched sample shows that: the number of steps in the exercise is compared with the number of steps before the exercise, and P is less than 0.01; the number of steps after exercise is compared with the number of steps in exercise, and P is less than 0.01; calorie in exercise compared to calorie before exercise P < 0.01; calorie after exercise compared with calorie in exercise P < 0.01; the heart rate during exercise is compared with the heart rate before exercise by P < 0.01; the heart rate after exercise is compared with the exercise center rate P <0.01, namely the steps, calories and heart rate before and after exercise are very different from those in exercise.
Under normal conditions, the average adult walks between 60-100 steps per minute, so that the step count is divided into 3 categories of 0 step, 0-100 steps and more than 100 steps for step count statistics, as shown in fig. 7; as can be seen from the overall measurement data, the quartiles of the heart rate values of the sampled individuals are 62, 70 and 85, so that the heart rate changes before, during and after exercise are divided into 5 categories of 62 or less, 62 to 70, 70 to 85, 85 to 100 and 100 or more, and the heart rate statistics is carried out, as shown in FIG. 8.
As shown in fig. 7, the number of steps before and after exercise: mainly comprising 0 step and 0-100 steps, accounting for more than 70 percent, almost not more than 100 steps; number of steps in exercise: the ratio of more than 100 steps is the largest, which is close to 100%, and 0 step and 0-100 steps are almost none. The performance completely conforms to the daily rules of badminton, consistent with the description of the course of the individual taking the sample (walking to the field-sports-relaxing and going back to dormitory).
As shown in fig. 8, the pre-and post-exercise heart rate values: most of the numerical values are below 100, and the proportion is close to 90%; heart rate in motion value: the value is the largest at a ratio of 100 or more, and is close to 80%. It can be seen that the variation of the heart rate of the badminton is characterized in that the heart rate before, during and after sports is very unstable.
In summary, the following steps: body-building and sports two sets of data overall trend: the step number, the calorie and the heart rate value before and after the exercise are relatively small, the step number, the calorie and the heart rate value are large in the exercise process, and the change range of different exercise conditions is different. Therefore, when the heart rate value is above 100 and the percentage is maximum and the number of steps is above 100 and the percentage is maximum, the state of exercise or fitness is considered. That is, in practice, by analyzing the variation trend of the watch data of the user, especially when the step number and the heart rate are simultaneously increased to a larger extent, the user can be considered in the process of exercise.
The above embodiments are only used for illustrating the present invention, and the structure, connection mode, manufacturing process, etc. of the components may be changed, and all equivalent changes and modifications performed on the basis of the technical solution of the present invention should not be excluded from the protection scope of the present invention.

Claims (5)

1. A human motion characteristic analysis method based on smart watch data is characterized by comprising the following steps:
1) extracting and analyzing basic data and measurement data of each sampling individual collected by the intelligent watch, wherein the measurement data comprises step number, calorie and heart rate;
the method for extracting and analyzing the basic data and the step number, the heart rate and the calorie measurement data collected by the intelligent watch comprises the following steps:
1.1) adopting an intelligent watch to collect basic data of each sampling individual, and simultaneously periodically collecting measurement data of each sampling individual;
1.2) periodically uploading basic data and measurement data acquired by each intelligent watch to a mobile terminal for storage;
1.3) extracting the basic data and the measurement data stored in each mobile terminal to a computer terminal so as to process data;
2) dividing the measurement data in any measurement time period according to the step number change and the heart rate change to obtain the heart rate and calorie data of the sampling individual in the exercise state and normal state time periods in the measurement time period;
the method for dividing the measurement data to obtain the heart rate and calorie data of the sampling individual in the exercise state and the normal state time period respectively comprises the following steps:
2.1) carrying out sum operation on the step numbers corresponding to all adjacent sampling nodes according to the time sequence to obtain a time period corresponding to the sampling node of which the sum operation result meets a preset step number change value;
2.2) carrying out sum operation on the heart rate data corresponding to all adjacent sampling nodes according to the time sequence to obtain a time period corresponding to the sampling node of which the sum operation result meets a preset heart rate change value;
2.3) obtaining the time period of each sampling individual in the motion state according to the time period corresponding to the sampling node which meets the preset step number change value and the preset heart rate change value through the obtained sum operation and the step number and the heart rate change corresponding to different motion types, wherein other time periods are the time periods in the normal state;
3) analyzing the human motion characteristic state based on the heart rate and calorie data of the motion state and the normal state time period to obtain the corresponding relation between the human motion state and the normal state and the change of the heart rate and the calorie;
the method for obtaining the corresponding relation between the human body motion state and the normal state and the heart rate and the calorie change comprises the following steps of:
3.1) carrying out statistical analysis on the measurement data of all the sampling individuals to obtain the activity state analysis results of all the sampling individuals, including the average step number, the calorie value and the heart rate value of all the sampling individuals;
3.2) randomly extracting a plurality of sampling individuals, and analyzing the step number, the calorie and the heart rate change corresponding to the exercise state and the normal state before and after the exercise state in any measurement time period by adopting a nonparametric inspection method to obtain the difference characteristics of the step number, the calorie and the heart rate in the exercise state and the normal state;
and 3.3) guiding and judging whether each sampling individual is in the exercise state or the normal state according to the difference characteristics of the step number, the calorie and the heart rate in the exercise state and the normal state.
2. The human motion feature analysis method based on smart watch data as claimed in claim 1, wherein: in the step 1.1), the method for collecting basic data and measurement data by using the smart watch comprises the following steps:
1.1.1) determining the type of the intelligent watch and a sampling individual according to the experiment requirement;
1.1.2) determining the type of basic data and measurement data to be collected according to the determined type of the intelligent watch and the sampling individual;
1.1.3) the intelligent watch periodically collects the relevant measurement data of each sampling individual in normal daily activities and uploads the data to the mobile terminal.
3. The human motion feature analysis method based on smart watch data as claimed in claim 1, wherein: in the step 2.1), the obtained sample point data of the sampling node satisfies the following formula:
Figure FDA0002191083040000021
wherein s isiRepresenting the number of steps, s, corresponding to the sampling node ii-1Represents the number of steps, S, corresponding to the sampling node i-1iRepresenting the sum of the steps between the sampling node i and the sampling node i-1; s0Indicating a motion condition preset value.
4. The human motion feature analysis method based on smart watch data as claimed in claim 1, wherein: in the step 2.2), the obtained sample point data of the sampling node satisfies the following formula:
Figure FDA0002191083040000022
wherein h isiRepresenting the heart rate, h, corresponding to the sampling node ii-1Representing the heart rate, S, corresponding to the sampling node i-1iRepresenting the sum of the heart rates, H, between sampling node i and sampling node i-10Is a preset heart rate variation value.
5. The human motion feature analysis method based on smart watch data as claimed in claim 1, wherein: in the step 3.2), the method for analyzing the data of each sampling individual by adopting a non-inspection method comprises the following steps:
3.2.1) assume that the measurement data of any sampling individual, i.e. the overall distribution of the sample records, is symmetric about the point θ, and at the same time assume that the overall distribution of the sample records is symmetric, thereby obtaining the test hypothesis H0：θ＝θ0Wherein, theta0To check the overall median;
3.2.2) calculating the difference z between the sample point corresponding to each sampling node and the median of the total inspection under the assumption of the inspection in the step 3.2.1)iThe calculation formula is as follows:
zi＝xi-θ0，i＝1,2,…,n，
wherein i is the sampling node ordinal number of the sampling unit, xiFor the sample point value corresponding to the sampling node, ziIs a sample point to theta0The distance of (d);
3.2.3) taking the absolute value of the difference obtained in the step 3.2.2), sequencing the sample points corresponding to each sampling node according to the absolute value of the difference, and calculating to obtain the rank R corresponding to each sample pointi；
3.2.4) comparing the difference z between each sample point and the median of the test population obtained in step 3.2.2)iAnd the rank R of each sample point obtained in step 3.2.3)iCalculating to obtain initial test statistic W+And W-；
Figure FDA0002191083040000031
Wherein
Figure FDA0002191083040000032
Figure FDA0002191083040000033
Wherein
Figure FDA0002191083040000034
3.2.5) judging whether the assumption is established in the step 3.2.1) according to the obtained initial test statistic, and further obtaining the total actual test statistic of the sample record of the sampling individual;
3.2.6) obtaining the difference result of the sample data of the sampling individual according to the obtained actual inspection statistic value.
CN201711015722.4A 2017-10-25 2017-10-25 Human motion characteristic analysis method based on smart watch data Active CN107773967B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711015722.4A CN107773967B (en) 2017-10-25 2017-10-25 Human motion characteristic analysis method based on smart watch data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711015722.4A CN107773967B (en) 2017-10-25 2017-10-25 Human motion characteristic analysis method based on smart watch data

Publications (2)

Publication Number Publication Date
CN107773967A CN107773967A (en) 2018-03-09
CN107773967B true CN107773967B (en) 2020-04-10

Family

ID=61435332

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711015722.4A Active CN107773967B (en) 2017-10-25 2017-10-25 Human motion characteristic analysis method based on smart watch data

Country Status (1)

Country Link
CN (1) CN107773967B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109059957B (en) * 2018-05-23 2020-12-04 歌尔科技有限公司 Step counting method, device and system
CN108897845A (en) * 2018-06-28 2018-11-27 浙江工业大学之江学院 A kind of student's Exercise after class exercise achievement appraisal procedure based on Intelligent bracelet
CN109509125B (en) * 2018-10-29 2023-11-17 广东精天科技有限公司 Intelligent physical education management method and system based on big data cloud platform

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020151810A1 (en) * 2001-04-16 2002-10-17 Acumen, Inc. Wrist-based fitness monitoring devices
JP3801163B2 (en) * 2003-03-07 2006-07-26 セイコーエプソン株式会社 Body motion detection device, pitch meter, pedometer, wristwatch type information processing device, control method, and control program
CN104574807B (en) * 2014-12-11 2018-05-11 惠州Tcl移动通信有限公司 User is reminded to adjust the method and device of amount of exercise
CN105467826B (en) * 2016-01-06 2018-09-18 上海应用技术学院 A kind of sport intellect wrist-watch and the fast acquisition methods of best race
CN106994238B (en) * 2017-03-30 2020-07-24 联想(北京)有限公司 Data processing method and electronic device for executing the same

Also Published As

Publication number Publication date
CN107773967A (en) 2018-03-09

Similar Documents

Publication Publication Date Title
CN107773967B (en) Human motion characteristic analysis method based on smart watch data
Padmaja et al. A machine learning approach for stress detection using a wireless physical activity tracker
Patel et al. Analysis of feature space for monitoring persons with Parkinson's disease with application to a wireless wearable sensor system
CN105496416B (en) A kind of recognition methods of human motion state and device
Dong et al. Detecting periods of eating during free-living by tracking wrist motion
Husted et al. The accuracy of pedometers in measuring walking steps on a treadmill in college students
US20160223580A1 (en) Calculating Pace and Energy Expenditure from Athletic Movement Attributes
US20110137836A1 (en) Method and system for generating history of behavior
CN103717125A (en) System and method for determining sleep and sleep stages of a person
EP3677171A1 (en) A method and apparatus for determining sleep need and sleep pressure based on physiological data
WO2018077020A1 (en) Wristband-based method for determining human emotions
WO2019056606A1 (en) Method and system for intelligently generating professional running training plan
CN110025321B (en) Psychological stress assessment method and related equipment
CN107590358A (en) Exercise guidance method and apparatus
CN106344034A (en) Sleep quality evaluation system and method
CN110911018A (en) Human health data acquisition system and health monitoring method based on cloud computing
CN109199336A (en) A kind of sleep quality quantization method, device and equipment based on machine learning
CN111530037A (en) Method and apparatus for assessing the state of physical function response during running exercise
CN112270970A (en) Intelligent motion management system and execution method
CN105574471B (en) The recognition methods of the method for uploading, user behavior of user behavior data and device
CN106446543A (en) Method and device for processing health data, and server cluster
CN112735563A (en) Recommendation information generation method and device and processor
CN108992040A (en) A kind of sleep quality state monitoring method and device
US20210353999A1 (en) Method and Device for Assessing Exercise Fatigue
JP2003319921A (en) Method, apparatus and system for evaluating condition of mind and body in state of group, and computer- readable recording medium having program recorded thereon

Legal Events

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