CN109584521A - A kind of incorrect sitting-pose monitoring method based on Notch sensor - Google Patents
A kind of incorrect sitting-pose monitoring method based on Notch sensor Download PDFInfo
- Publication number
- CN109584521A CN109584521A CN201811194692.2A CN201811194692A CN109584521A CN 109584521 A CN109584521 A CN 109584521A CN 201811194692 A CN201811194692 A CN 201811194692A CN 109584521 A CN109584521 A CN 109584521A
- Authority
- CN
- China
- Prior art keywords
- sitting posture
- data
- sitting
- angle
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/24—Reminder alarms, e.g. anti-loss alarms
Landscapes
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a kind of incorrect sitting-pose monitoring methods based on Notch sensor.Comprising steps of acquiring sitting posture data based on Notch sensor;Collected data are pre-processed, dirty data is removed, improves the reliability and accuracy of data, data format is made to meet the requirements;Treated data are subjected to feature extraction and selection, the feature composition characteristic vector that will be extracted;By extracting feature, all kinds of sitting posture models are established, all kinds of sitting posture models are then based on, sitting posture monitoring is carried out to subjects in real time, finally obtains the result of Classification and Identification;It designs mobile application and sitting posture prompting is carried out to user, and propose to correct and suggest.The present invention has preferable accuracy and convenience, is conducive to protect cervical vertebra, lumbar vertebrae, myopia etc. because of disease caused by incorrect sitting-pose from source, improves the inferior health of people, reduce the medical burden of society.
Description
Technical field
The invention belongs to Activity recognition fields, and in particular to a kind of incorrect sitting-pose monitoring method based on Notch sensor.
Background technique
In routine work and study, sitting posture is longest posture of holding time.For other opposite postures, sitting posture consumption
Energy is small to be made one to be not easy tired out and facilitates office and study, but keeping improper sitting posture for a long time then can bring seriously to human health
Harm.It showing according to investigations, China's sitting bends over the desk in crowd that there are about 70%-80% in working and learning, and long-term sitting posture is improper,
Improperly sitting posture makes vertebra be in buckling position or certain special body positions for a long time, not only increases the pressure in spinal disc,
Also the muscle ligament in vertebra portion can be made to be chronically at non-coordinating stress and lateral bending occurs.Also result in pelvis expand outwardly,
Turn in thigh, knee joint also turns within, so that both feet, which interior may be skimmed, influences walking posture, backbone is also resulted in when serious
Lordosis influences the sense of equilibrium and the coordination ability of body entirety.The spinal curvature number of abnormal sitting posture generation has been more than according to statistics
50%, myopia of student rate is also with annual 6% speed increase.
With the development of sensor technology, so that sensor bulk is small and light-weight at this stage, it is easier to be made wearable
Equipment, it is easy to carry, the daily life of user will not more be impacted without limitation on resource-user cost method scene.Sensor
Development solve the problems, such as data acquisition i.e. sitting posture supervision, smart phone it is universal, solve the problems, such as prompting.Using Notch
Sensor acquires data, and cost is lower and low to environmental requirement, is easy to carry about with one and is used for a long time.It is real-time by Android phone
It reminds, user is helped to correct incorrect sitting-pose in time.Therefore, there is higher practicability and convenience, it is improper effectively to prevent
The inferior health of sitting posture bring and caused various diseases.
Summary of the invention
It is above-mentioned to solve the purpose of the present invention is to provide a kind of incorrect sitting-pose monitoring method based on Notch sensor
Technical problem.
To achieve the above object the invention adopts the following technical scheme:
A kind of incorrect sitting-pose monitoring method based on Notch sensor, comprising steps of
Step 1: sitting posture data are acquired based on Notch sensor;
Step 2: collected data being pre-processed, dirty data is removed, improves the reliability and accuracy of data, make
Data format meets the requirements;
Step 3: treated data being subjected to feature extraction and selection, the feature composition characteristic extracted is sweared
Amount;
Step 4: by extracting feature, establishing all kinds of sitting posture models, be then based on all kinds of sitting posture models, in real time to monitored
Person carries out sitting posture monitoring, finally obtains the result of Classification and Identification;
Step 5: design mobile application carries out sitting posture prompting to user, and proposes to correct and suggest.
As a further solution of the present invention, step 1 specifically includes following content: Notch sensor is fixed on human body
Lumbar vertebrae and thoracic vertebrae two at, then acquire master data information, Notch sensor built-in acceleration sensor, gyroscope and magnetic
Power meter, the data captured are stored with the data format of Z, X, Y of Eulerian angles, establish an independent movement in three-dimensional space
Coordinate system is to record the motion conditions in this coordinate system.
As a further solution of the present invention, step 2 specifically includes following content: collected by Notch sensor
User's sitting posture data, including time and Eulerian angles, and transmitted by bluetooth, the data received have been carried out at discretization
Continuous data, are divided into isometric discrete data section by reason, can be reduced calculating bring space resources after Data Discretization and be disappeared
Consumption, and can be improved the ability for resisting noise.
As a further solution of the present invention, 8 features of step 3 selection include:
Step 3.1: extracting the pitch angle X of Eulerian angles, pitch angle is also known as upward view angle, when being object rotation round a fixed point and X-axis
Between angle, the angle of representative's bench over and layback;
Step 3.2: extracting the yaw angle Y of Eulerian angles, yaw angle is also known as left and right corner, when being object rotation round a fixed point and Y-axis
Between angle, represent the angle of human body left-right rotation;
Step 3.3: the roll angle Z of extraction Eulerian angles, angle when tumbler angle is object rotation round a fixed point between Z axis,
Represent the downward shift amount of gravity center of human body;
Step 3.4: extracting the pitch angle X variance D (X) of sitting posture;Variance is the number that can measure a sample fluctuation size
Value, variance more great fluctuation process is bigger, and the data fluctuations situation of the smaller reaction of opposite variance is with regard to smaller, if data whithin a period of time
Fluctuating change is smaller to prove that user is just keeping a sitting posture in this time, and because it is by correct that majority, which keep sitting posture,
Sitting posture starts, and the pitch angle X value range of correct sitting posture is smaller, so can determine that user may keep at this time when variance is smaller
Correct sitting posture;
Step 3.5: extracting the yaw angle Y variance D (Y) of sitting posture, determine the sitting posture yaw angle Y of the user within this time
Amplitude of variation, i.e. left-right rotation range of the user when keeping this sitting posture;
Step 3.6: extracting the tumbler angle Z variance D (Z) of sitting posture, determine the sitting posture tumbler angle of the user in this time
The amplitude of variation of amplitude of variation, i.e. the user center of gravity when keeping this sitting posture;
Step 3.7: the sum of three axle clamp angle absolute values Sum (x, y, z) is extracted, due to Eulerian angles when using incorrect sitting-pose
Three axle clamp angles occur to change in various degree, so the value of Sum (x, y, z) can react the undesirable level of sitting posture at this time;
Sum (x, y, z)=| X |+| Y |+| Z | (4);
Step 3.8: extracting three axle clamp angle quadratic sum Sum (w), quadratic sum is used to the number of the sum of three axis absolute value of amplification characteristic
The deviation of feature under value and different sitting postures;
8 characteristic values are chosen, relationship between sitting posture type and data is embodied, wherein Eulerian angles are a kind of unique identifications
The value of moving object position in three-dimensional system of coordinate.
As a further solution of the present invention, step 4 specifically includes following content:
Step 4.1: determining 5 class sitting posture classifications, such as table 1, described according to 5 class sitting posture classifications, establish sitting posture model;
The classification of 1 sitting posture of table
Step 4.2: sitting posture recognition methods being carried out based on threshold value, the data that real-time monitoring uses are collected by sensor
The Eulerian angles of tester's sitting posture --- Z, X, Y, when human body keeps incorrect sitting-pose, chest and back can occur to turn using buttocks as the center of circle
Dynamic, Eulerian angles Z, X, Y can shift in various degree, can whether there is incorrect sitting-pose by the threshold decision user of Eulerian angles
Behavior, but can only substantially judge user's degrees of offset at this time, can not accurate judgement user specific sitting posture type;
Step 4.3: for cannot specific sitting posture classification, then by k nearest neighbor algorithm further identify that KNN is a kind of base
In the supervised learning of example, by calculating the distance between new data and training data characteristic value, then selection K (K >=1) is a
Classification judgement (ballot method) is carried out apart from nearest neighbours or is returned.
As a further solution of the present invention, it is identified according to sitting posture, does not give and remind at that time in user's long-time sitting posture, mention
The mode of waking up can use smart phone or tablet computer by vibration or jingle bell, mobile terminal, and support IOS or Android behaviour
Make system, meanwhile, by long-term sitting posture monitoring data, user health is instructed, and then it is improper that sitting posture is mitigated or eliminated
Bring disease problems.
The beneficial effects of the present invention are: a kind of incorrect sitting-pose monitoring method based on Notch sensor of the present invention, passes through base
Sitting posture data are acquired in the wearable device of Notch, mobile terminal is transmitted to by pretreatments such as discretizations, extracts and select
Feature relevant to sitting posture.The classifying identification method for being then based on threshold value carries out sitting posture classification, for fuzzy sitting posture row of classifying
For the method for further using k nearest neighbor is identified.This method has preferable accuracy and convenience, is conducive to from source
It protects cervical vertebra, lumbar vertebrae, myopia etc. because of disease caused by incorrect sitting-pose, improves the inferior health of people, reduce the medical burden of society.
Detailed description of the invention
Fig. 1 is monitoring method flow chart of the present invention.
Fig. 2 is the sitting posture identification process the present invention is based on Notch sensor.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Whole elaboration.
As shown in Figs. 1-2, a kind of incorrect sitting-pose monitoring method based on Notch sensor of the present invention, specifically according to
Following steps are implemented:
Step 1: the sitting posture data based on Notch sensor acquisition monitored personnel;
Step 1.1: wearing Notch wearable device, i.e., two sensors are respectively placed in monitored personnel's thoracic vertebrae and waist
Sitting posture data are acquired at vertebra;
Step 1.2: the pretreatment of data.Pass through the collected user's sitting posture data of Notch sensor, including time and Europe
Angle is drawn, and is transmitted by bluetooth.The collected data of sensor vulnerable to external interference, stability and data precision due to having
It is a little insufficient.Will cause the difference of calculated result by the collected dirty data of external interference, the present invention to the data received into
It has gone sliding-model control, continuous data is divided into isometric discrete data section.Calculating can be reduced after Data Discretization to bring
Space resources consumption, and can be improved the ability for resisting noise;
Step 2: the extraction and selection of feature;
This is the key that solve Activity recognition accuracy.Selection is suitably characterized in establishing action recognition and institute is collected
The important link contacted between data.With the accuracy of machine learning algorithm progress action recognition and computation complexity and feature
It chooses inseparable.By the feature of extraction convenient for the relationship between discovery data and Behavioral change.Feature mainly have temporal signatures,
The three classes such as frequency domain character and wavelet character.The characteristics of in order to keep sitting posture analysis more accurate, acquire data in conjunction with Notch sensor
And the characteristics of feature, select 8 features;
Step 2.1: extracting the pitch angle X of Eulerian angles, pitch angle is also known as upward view angle, when being object rotation round a fixed point and X-axis
Between angle, the angle of representative's bench over and layback;
Step 2.2: extracting the yaw angle Y of Eulerian angles, yaw angle is also known as left and right corner, when being object rotation round a fixed point and Y-axis
Between angle, represent the angle of human body left-right rotation;
Step 2.3: the roll angle Z of extraction Eulerian angles, angle when tumbler angle is object rotation round a fixed point between Z axis,
Represent the downward shift amount of gravity center of human body;
Step 2.4: extracting the pitch angle X variance D (X) of sitting posture;Variance is the number that can measure a sample fluctuation size
Value, variance more great fluctuation process is bigger, and the data fluctuations situation of the smaller reaction of opposite variance is with regard to smaller.If data whithin a period of time
Fluctuating change is smaller to prove that user is just keeping a sitting posture in this time.And because it is by correct that majority, which keep sitting posture,
Sitting posture starts, and the pitch angle X value range of correct sitting posture is smaller.So can determine that user may keep at this time when variance is smaller
Correct sitting posture;
Step 2.5: extracting the yaw angle Y variance D (Y) of sitting posture, determine the sitting posture yaw angle Y of the user within this time
Amplitude of variation.That is left-right rotation range of the user when keeping this sitting posture;
Step 2.6: extracting the tumbler angle Z variance D (Z) of sitting posture, determine the sitting posture tumbler angle of the user in this time
The amplitude of variation of amplitude of variation, i.e. the user center of gravity when keeping this sitting posture;
Step 2.7: the sum of three axle clamp angle absolute values Sum (x, y, z) is extracted, due to Eulerian angles when using incorrect sitting-pose
Three axle clamp angles occur to change in various degree, so the value of Sum (x, y, z) can react the undesirable level of sitting posture at this time;
Sum (x, y, z)=| X |+| Y |+| Z | (4);
Step 2.8: extracting three axle clamp angle quadratic sum Sum (w), quadratic sum is used to the number of the sum of three axis absolute value of amplification characteristic
The deviation of feature under value and different sitting postures;
The present invention chooses 8 characteristic values altogether, embodies relationship between sitting posture type and data.Wherein, Eulerian angles are a kind of
The value of moving object position in unique identification three-dimensional system of coordinate.
Step 3: by extracting feature, establishing all kinds of sitting posture models, be then based on all kinds of sitting posture models, in real time to monitored
Person carries out sitting posture monitoring;
Step 3.1: determining 5 class sitting posture classifications, described according to 5 class sitting posture classifications, establish sitting posture model;
The classification of 1 sitting posture of table
Step 3.2: sitting posture recognition methods being carried out based on threshold value, the data that real-time monitoring uses are collected by sensor
The Eulerian angles of tester's sitting posture --- Z, X, Y.When human body keeps incorrect sitting-pose, chest and back can occur to turn using buttocks as the center of circle
Dynamic, Eulerian angles Z, X, Y can shift in various degree.Incorrect sitting-pose can be whether there is by the threshold decision user of Eulerian angles
Behavior.But can only substantially judge user's degrees of offset at this time, can not accurate judgement user specific sitting posture type;
Sitting posture based on threshold value monitors part, and advantage is that calculating speed is fast, and calculating occupies little space, and clear logic is conducive to
Understand;Disadvantage is clearly distinguish sitting posture classification.
Step 3.3: for cannot specific sitting posture classification, then further identified by k nearest neighbor algorithm.KNN is a kind of base
In the supervised learning of example, by calculating the distance between new data and training data characteristic value, then selection K (K >=1) is a
Classification judgement (ballot method) is carried out apart from nearest neighbours or is returned.
Step 4: design mobile application carries out sitting posture prompting to user, and proposes to correct and suggest.
It is identified according to sitting posture, does not give and remind at that time in user's long-time sitting posture, alerting pattern passes through vibration or jingle bell.It moves
Dynamic terminal can use smart phone or tablet computer, and support IOS or Android operation system.Meanwhile passing through long-term seat
Appearance monitoring data, instruct user health, and then the improper bring disease problems of sitting posture are mitigated or eliminated.
The above is present pre-ferred embodiments, for the ordinary skill in the art, according to the present invention
Introduction, in the case where not departing from the principle of the present invention and spirit, changes, modifications, replacement and change that embodiment is carried out
Type is still fallen within protection scope of the present invention.
Claims (6)
1. a kind of incorrect sitting-pose monitoring method based on Notch sensor, which is characterized in that comprising steps of
Step 1: sitting posture data are acquired based on Notch sensor;
Step 2: collected data being pre-processed, dirty data is removed, improves the reliability and accuracy of data, make data
Format meets the requirements;
Step 3: treated data are subjected to feature extraction and selection, the feature composition characteristic vector that will be extracted;
Step 4: by extract feature, establish all kinds of sitting posture models, be then based on all kinds of sitting posture models, in real time to subjects into
The monitoring of row sitting posture, finally obtains the result of Classification and Identification;
Step 5: design mobile application carries out sitting posture prompting to user, and proposes to correct and suggest.
2. a kind of incorrect sitting-pose monitoring method based on Notch sensor as described in claim 1, which is characterized in that step 1
Specifically include following content: Notch sensor be fixed at the lumbar vertebrae and thoracic vertebrae two of human body, then acquire basic number it is believed that
Breath, Notch sensor built-in acceleration sensor, gyroscope and magnetometer, the data captured are with the number of Z, X, Y of Eulerian angles
It is stored according to format, establishes an independent kinetic coordinate system in three-dimensional space to record the motion conditions in this coordinate system.
3. a kind of incorrect sitting-pose monitoring method based on Notch sensor as described in claim 1, which is characterized in that step 2
It specifically includes following content: by the collected user's sitting posture data of Notch sensor, including time and Eulerian angles, and passing through
Bluetooth is transmitted, and has carried out sliding-model control to the data received, and continuous data are divided into isometric discrete data section,
It can reduce after Data Discretization and calculate the consumption of bring space resources, and can be improved the ability for resisting noise.
4. a kind of incorrect sitting-pose monitoring method based on Notch sensor as described in claim 1, which is characterized in that described
Step 3 select 8 features include:
Step 3.1: extracting the pitch angle X of Eulerian angles, pitch angle is also known as upward view angle, when being object rotation round a fixed point between X-axis
Angle, the angle of representative's bench over and layback;
Step 3.2: extracting the yaw angle Y of Eulerian angles, yaw angle is also known as left and right corner, when being object rotation round a fixed point between Y-axis
Angle, represent the angle of human body left-right rotation;
Step 3.3: extracting the roll angle Z of Eulerian angles, angle when tumbler angle is object rotation round a fixed point between Z axis represents
The downward shift amount of gravity center of human body;
Step 3.4: extracting the pitch angle X variance D (X) of sitting posture;Variance is the numerical value that can measure a sample fluctuation size, side
Poor more great fluctuation process is bigger, and the data fluctuations situation of the smaller reaction of opposite variance is with regard to smaller, if data fluctuations whithin a period of time
Changing smaller is to prove that user is just keeping a sitting posture in this time, and because it is by correct sitting posture that majority, which keep sitting posture,
Start, the pitch angle X value range of correct sitting posture is smaller, so can determine that user may keep correct at this time when variance is smaller
Sitting posture;
Step 3.5: extracting the yaw angle Y variance D (Y) of sitting posture, determine the change of the sitting posture yaw angle Y of user within this time
The left-right rotation range of change amplitude, i.e. user when keeping this sitting posture;
Step 3.6: extracting the tumbler angle Z variance D (Z) of sitting posture, determine the sitting posture tumbler angle variation of the user in this time
The amplitude of variation of amplitude, i.e. the user center of gravity when keeping this sitting posture;
Step 3.7: the sum of three axle clamp angle absolute values Sum (x, y, z) is extracted, due to three axis of Eulerian angles when using incorrect sitting-pose
Angle occurs to change in various degree, so the value of Sum (x, y, z) can react the undesirable level of sitting posture at this time;
Sum (x, y, z)=| X |+| Y |+| Z | (4);
Step 3.8: extract three axle clamp angle quadratic sum Sum (w), quadratic sum be used to the sum of three axis absolute value of amplification characteristic numerical value and
The deviation of feature under different sitting postures;
8 characteristic values are chosen, relationship between sitting posture type and data is embodied, wherein Eulerian angles are that a kind of unique identification is three-dimensional
The value of moving object position in coordinate system.
5. a kind of incorrect sitting-pose monitoring method based on Notch sensor as described in claim 1, which is characterized in that step 4
Specifically include following content:
Step 4.1: determining 5 class sitting posture classifications, such as table 1, described according to 5 class sitting posture classifications, establish sitting posture model;
The classification of 1 sitting posture of table
Step 4.2: sitting posture recognition methods being carried out based on threshold value, the data that real-time monitoring uses are that test is collected by sensor
The Eulerian angles of person's sitting posture --- Z, X, Y, when human body keeps incorrect sitting-pose, chest and back can rotate by the center of circle of buttocks, Europe
It draws angle Z, X, Y that can shift in various degree, incorrect sitting-pose behavior can be whether there is by the threshold decision user of Eulerian angles,
But can only substantially judge user's degrees of offset at this time, can not accurate judgement user specific sitting posture type;
Step 4.3: for cannot specific sitting posture classification, then further identified by k nearest neighbor algorithm, KNN is a kind of based on real
The supervised learning of example, by calculating the distance between new data and training data characteristic value, then a distance of selection K (K >=1)
Nearest neighbours carry out classification judgement (ballot method) or return.
6. a kind of incorrect sitting-pose monitoring method based on Notch sensor as described in claim 1, which is characterized in that according to
Sitting posture identification, does not give at that time in user's long-time sitting posture and reminds, and alerting pattern can be adopted by vibration or jingle bell, mobile terminal
With smart phone or tablet computer, and support IOS or Android operation system, meanwhile, by long-term sitting posture monitoring data,
User health is instructed, and then the improper bring disease problems of sitting posture are mitigated or eliminated.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811194692.2A CN109584521A (en) | 2018-10-15 | 2018-10-15 | A kind of incorrect sitting-pose monitoring method based on Notch sensor |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811194692.2A CN109584521A (en) | 2018-10-15 | 2018-10-15 | A kind of incorrect sitting-pose monitoring method based on Notch sensor |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109584521A true CN109584521A (en) | 2019-04-05 |
Family
ID=65920013
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811194692.2A Pending CN109584521A (en) | 2018-10-15 | 2018-10-15 | A kind of incorrect sitting-pose monitoring method based on Notch sensor |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109584521A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111657621A (en) * | 2020-06-04 | 2020-09-15 | 厉晨宇 | Sports shoes for detecting wearing time and sports strength and injury prevention method |
WO2021138964A1 (en) * | 2020-01-10 | 2021-07-15 | 鄢家厚 | Read/write distance identification method based on smart watch |
CN113288122A (en) * | 2021-05-21 | 2021-08-24 | 河南理工大学 | Wearable sitting posture monitoring device and sitting posture monitoring method |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010038707A (en) * | 2008-08-05 | 2010-02-18 | Shimadzu Corp | Motion tracker device |
CN104392583A (en) * | 2014-11-27 | 2015-03-04 | 北京工业大学 | Fall detection and alarm system and method based on KNN algorithm |
CN105769210A (en) * | 2016-05-09 | 2016-07-20 | 中科院合肥技术创新工程院 | Wearable home body posture detection Internet of Things terminal |
CN106643708A (en) * | 2016-09-21 | 2017-05-10 | 苏州坦特拉自动化科技有限公司 | IMU-based interactive sitting posture correction device, sitting posture correction appliance and monitoring software |
CN107478223A (en) * | 2016-06-08 | 2017-12-15 | 南京理工大学 | A kind of human body attitude calculation method based on quaternary number and Kalman filtering |
CN207055578U (en) * | 2017-02-28 | 2018-03-02 | 王泽宇 | A kind of sitting posture correction device for children based on 3-axis acceleration sensor |
CN108209929A (en) * | 2016-12-22 | 2018-06-29 | 欧普照明股份有限公司 | A kind of sitting posture identifying system and sitting posture recognition methods |
CN108509897A (en) * | 2018-03-29 | 2018-09-07 | 同济大学 | A kind of human posture recognition method and system |
-
2018
- 2018-10-15 CN CN201811194692.2A patent/CN109584521A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010038707A (en) * | 2008-08-05 | 2010-02-18 | Shimadzu Corp | Motion tracker device |
CN104392583A (en) * | 2014-11-27 | 2015-03-04 | 北京工业大学 | Fall detection and alarm system and method based on KNN algorithm |
CN105769210A (en) * | 2016-05-09 | 2016-07-20 | 中科院合肥技术创新工程院 | Wearable home body posture detection Internet of Things terminal |
CN107478223A (en) * | 2016-06-08 | 2017-12-15 | 南京理工大学 | A kind of human body attitude calculation method based on quaternary number and Kalman filtering |
CN106643708A (en) * | 2016-09-21 | 2017-05-10 | 苏州坦特拉自动化科技有限公司 | IMU-based interactive sitting posture correction device, sitting posture correction appliance and monitoring software |
CN108209929A (en) * | 2016-12-22 | 2018-06-29 | 欧普照明股份有限公司 | A kind of sitting posture identifying system and sitting posture recognition methods |
CN207055578U (en) * | 2017-02-28 | 2018-03-02 | 王泽宇 | A kind of sitting posture correction device for children based on 3-axis acceleration sensor |
CN108509897A (en) * | 2018-03-29 | 2018-09-07 | 同济大学 | A kind of human posture recognition method and system |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021138964A1 (en) * | 2020-01-10 | 2021-07-15 | 鄢家厚 | Read/write distance identification method based on smart watch |
CN111657621A (en) * | 2020-06-04 | 2020-09-15 | 厉晨宇 | Sports shoes for detecting wearing time and sports strength and injury prevention method |
CN113288122A (en) * | 2021-05-21 | 2021-08-24 | 河南理工大学 | Wearable sitting posture monitoring device and sitting posture monitoring method |
CN113288122B (en) * | 2021-05-21 | 2023-12-19 | 河南理工大学 | Wearable sitting posture monitoring device and sitting posture monitoring method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Aziz et al. | An analysis of the accuracy of wearable sensors for classifying the causes of falls in humans | |
CN109069066A (en) | Wearable and connection gait analysis system | |
US8246555B2 (en) | Method and system for monitoring sport related fitness by estimating muscle power and joint force of limbs | |
Van Laerhoven et al. | Spine versus porcupine: A study in distributed wearable activity recognition | |
JP5674766B2 (en) | Sensing device for detecting wearing position | |
Bonato et al. | Data mining of motor patterns recorded with wearable technology | |
CN105051799A (en) | Method for detecting falls and a fall detector. | |
CN109584521A (en) | A kind of incorrect sitting-pose monitoring method based on Notch sensor | |
Hamäläinen et al. | Jerk-based feature extraction for robust activity recognition from acceleration data | |
CN108514421A (en) | The method for promoting mixed reality and routine health monitoring | |
CN107506706A (en) | A kind of tumble detection method for human body based on three-dimensional camera | |
CN106210269A (en) | A kind of human action identification system and method based on smart mobile phone | |
Rodríguez-Martín et al. | Posture transition identification on PD patients through a SVM-based technique and a single waist-worn accelerometer | |
WO2014153665A1 (en) | System and method for monitoring a subject | |
CN109805935A (en) | A kind of intelligent waistband based on artificial intelligence hierarchical layered motion recognition method | |
CN103785157A (en) | Human body motion type identification accuracy improving method | |
CN110477924A (en) | Adaptive motion posture sensing System and method for | |
Bi et al. | Smartge: identifying pen-holding gesture with smartwatch | |
CN114818952A (en) | Intelligent falling posture classification and identification method based on mobile phone sensor | |
Ma et al. | Smart cushion-based activity recognition: Prompting users to maintain a healthy seated posture | |
EP4146068A1 (en) | Systems, devices, and methods for determining movement variability, illness and injury prediction and recovery readiness | |
CN114341947A (en) | System and method for exercise type recognition using wearable devices | |
Viana et al. | GymApp: A real time physical activity trainner on wearable devices | |
CN117109567A (en) | Riding gesture monitoring method and system for dynamic bicycle movement and wearable riding gesture monitoring equipment | |
Tlili et al. | A Review on posture monitoring systems |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190405 |