CN115670445A - Human body posture detection and recognition system and method - Google Patents
Human body posture detection and recognition system and method Download PDFInfo
- Publication number
- CN115670445A CN115670445A CN202211396929.1A CN202211396929A CN115670445A CN 115670445 A CN115670445 A CN 115670445A CN 202211396929 A CN202211396929 A CN 202211396929A CN 115670445 A CN115670445 A CN 115670445A
- Authority
- CN
- China
- Prior art keywords
- human body
- acceleration
- attitude
- human
- value
- 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
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000001514 detection method Methods 0.000 title claims abstract description 23
- 239000011159 matrix material Substances 0.000 claims abstract description 36
- 238000012545 processing Methods 0.000 claims abstract description 12
- 230000005540 biological transmission Effects 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 230000001133 acceleration Effects 0.000 claims description 108
- 230000005484 gravity Effects 0.000 claims description 51
- 239000013598 vector Substances 0.000 claims description 34
- 238000004364 calculation method Methods 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 15
- 238000000605 extraction Methods 0.000 claims description 4
- 210000001015 abdomen Anatomy 0.000 claims description 2
- 239000000284 extract Substances 0.000 claims description 2
- 230000036544 posture Effects 0.000 abstract description 47
- 238000004458 analytical method Methods 0.000 abstract description 5
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 230000036541 health Effects 0.000 abstract description 2
- 230000008859 change Effects 0.000 description 6
- 238000011161 development Methods 0.000 description 6
- 230000018109 developmental process Effects 0.000 description 6
- 238000006243 chemical reaction Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 238000001914 filtration Methods 0.000 description 3
- 230000032683 aging Effects 0.000 description 2
- 230000000295 complement effect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 206010024453 Ligament sprain Diseases 0.000 description 1
- 208000010040 Sprains and Strains Diseases 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005802 health problem Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000004630 mental health Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a human body posture detection and recognition system and a method, wherein the system comprises: the data acquisition module is used for acquiring human motion data; the transmission processing module is used for preprocessing the human body motion data, acquiring attitude coordinate data and transmitting the attitude coordinate data; the attitude resolving module is used for resolving the attitude according to the attitude coordinate data to obtain the human body attitude; and the gesture recognition module is used for extracting the characteristics of the human motion data to acquire the accurate human gesture. The invention carries out characteristic analysis on human body postures through the Euler angle and the state matrix, further distinguishes the human body postures of standing, sitting, lying, walking and the like by utilizing characteristic values such as mean value, variance and the like, and has the function of identifying the human body posture analysis and identification based on the Euler angle and threshold distinguishing algorithm. Meanwhile, the function of remotely detecting and recognizing human body postures is realized, and the system has positive practical significance for monitoring the home health of the old.
Description
Technical Field
The invention belongs to the field of human body posture recognition, and particularly relates to a human body posture detection and recognition system and a human body posture detection and recognition method.
Background
With the further development of society, the aging phenomenon of the population is gradually intensified, and in a long period of time in the future, the aging phenomenon of the population will continuously exist, the population number of the elderly will continuously increase, and more people begin to pay attention to the health problems of the elderly. The old person falls down the phenomenon and often takes place, often causes serious consequence after the old person falls down, falls down and is one of the important reasons that cause the old person injury or even die, and the event of sprain, fracture or even die that causes because the old person falls down all has many cases every year, wherein because the discovery does not cause serious accident to account for a large part in time often, consequently, will not delay the development of the physical and mental health monitoring and medical guarantee system of old person.
In recent years, with the development of miniature electronic elements, sensors are greatly improved, the size of the sensors is gradually reduced, the precision of the sensors is gradually improved, the price of the sensors is gradually reduced, and the characteristics enable wearable equipment to obtain great development potential. Meanwhile, the rapid development of micro capacitors, novel materials and wireless transmission technology also provides more technical support for wearable equipment. In addition, the internet of things technology and the development thereof are one of the strongest communication technologies in the 21 st century. Nowadays, the internet of things technology is widely applied to daily life of people, such as mobile phone communication, household appliances, medical equipment and the like. The technology of the Internet of things also provides technical support for wearable human body posture detection equipment and mobile phone communication, and brings convenience to timely interaction with equipment.
Disclosure of Invention
The invention aims to provide a human body posture detection and recognition system and a human body posture detection and recognition method, which are used for solving the problems in the prior art.
On one hand, the invention provides a human body posture detection and recognition system, which comprises a data acquisition module, a transmission processing module, a posture resolving module and a posture recognition module which are connected in sequence;
the data acquisition module is used for acquiring human body motion data;
the transmission processing module is used for preprocessing the human body motion data, acquiring attitude coordinate data and transmitting the attitude coordinate data;
the attitude resolving module is used for resolving the attitude according to the attitude coordinate data to obtain the human body attitude;
the gesture recognition module is used for carrying out feature extraction on the human motion data to obtain an accurate human gesture.
Optionally, the human motion data includes an acceleration value and an original gyroscope value, the data acquisition module includes an acceleration sensor and a gyroscope, and the acceleration sensor and the gyroscope are both one in number and are arranged at the position of the abdomen of the human body; the acceleration sensor and the gyroscope are respectively used for acquiring the acceleration value and the original gyroscope value.
Optionally, the attitude coordinate data includes a gravitational acceleration and angular velocity values, where the angular velocity values include a pitch angle velocity, a yaw angle velocity, and a roll angle velocity; and the transmission processing module converts the acceleration value in the human motion data into the gravitational acceleration and converts the original gyroscope value in the human motion data into the angular velocity value.
Optionally, the human body posture comprises an upright state and a lying state; the attitude resolving module acquires a quaternion cosine matrix by constructing a cosine matrix and expressing the cosine matrix based on a quaternion; normalizing the gravity acceleration in the attitude coordinate data to obtain an actual gravity acceleration; acquiring a gravity unit vector under a machine position coordinate system according to the quaternion cosine matrix and the actual gravity acceleration; integrating the angular velocity values in the attitude coordinate data to obtain a calculation gravity vector; calculating the error between the calculated gravity vector and a gravity unit vector according to vector cross multiplication to obtain a gravity error, and compensating the angular velocity value according to the gravity error; according to the compensated angular velocity value, solving the quaternion cosine matrix by adopting a first-order Longgoku tower; and converting the solved quaternion cosine matrix into an attitude Euler angle, and acquiring the human body attitude according to the attitude Euler angle.
Optionally, the precise body posture comprises sitting, standing, walking; the gesture recognition module extracts characteristic values of the human motion data, wherein the characteristic values comprise maximum acceleration, minimum acceleration, extreme difference of acceleration, average acceleration and acceleration variance;
and distinguishing sitting and standing in the accurate human body posture according to the acceleration variance, and distinguishing standing and walking in the accurate human body posture according to the acceleration range.
On the other hand, in order to achieve the above object, the present invention provides a human body posture detection and recognition method, which comprises the following steps:
acquiring human body motion data;
preprocessing the human body motion data to obtain posture coordinate data;
performing attitude calculation based on the attitude coordinate data to obtain the human body attitude;
and extracting the characteristics of the human motion data to obtain the accurate human posture.
Optionally, the human motion data comprises acceleration values and raw gyroscope values;
the process of acquiring the human motion data comprises the following steps:
and acquiring the acceleration value and the original gyroscope value based on an acceleration sensor and a gyroscope respectively.
Optionally, the attitude coordinate data includes a gravitational acceleration and angular velocity values, where the angular velocity values include a pitch angle velocity, a yaw angle velocity, and a roll angle velocity; the process of preprocessing the human motion data comprises the following steps:
and converting the acceleration value in the human motion data into the gravitational acceleration, and converting the original gyroscope value in the human motion data into the angular velocity value.
Optionally, the human body posture comprises an upright state and a lying state;
the process of performing attitude calculation based on the attitude coordinate data includes:
constructing a cosine matrix and expressing the cosine matrix based on a quaternion to obtain a quaternion cosine matrix;
normalizing the gravity acceleration in the attitude coordinate data to obtain an actual gravity acceleration;
acquiring a gravity unit vector under a machine position coordinate system based on the quaternion cosine matrix and the actual gravity acceleration;
integrating the angular velocity values in the attitude coordinate data to obtain a calculation gravity vector;
calculating the error between the calculated gravity vector and a gravity unit vector based on vector cross multiplication to obtain a gravity error, and compensating the angular velocity value based on the gravity error;
on the basis of the compensated angular velocity value, solving the quaternion-cosine matrix by adopting a first-order Longgokuta;
and converting the solved quaternion cosine matrix into an attitude Euler angle, and acquiring the human body attitude based on the attitude Euler angle.
Optionally, the process of performing feature extraction on the human motion data by the accurate human posture including sitting, standing, and walking includes:
extracting characteristic values from the human motion data, wherein the characteristic values comprise maximum acceleration, minimum acceleration, extreme difference of acceleration, average acceleration and acceleration variance;
distinguishing sitting and standing in the accurate human body posture based on the acceleration variance, and distinguishing standing and walking in the accurate human body posture based on the acceleration range.
The invention has the technical effects that:
the invention carries out characteristic analysis on human body postures through the Euler angle and the state matrix, further distinguishes the human body postures of standing, sitting, lying, walking and the like by utilizing characteristic values such as a mean value, a variance and the like, and identifies the human body posture analysis and identification function based on the Euler angle and a threshold distinguishing algorithm. Meanwhile, the function of remotely detecting and recognizing human body postures is realized, and the method has positive practical significance for monitoring the home health of the old.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application. In the drawings:
FIG. 1 is a diagram of a human body posture detection and recognition system according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a coordinate transformation in an embodiment of the present invention;
FIG. 3 is a graph of stance angle data for a standing position in an embodiment of the invention;
FIG. 4 is a diagram of attitude angle data while seated in an embodiment of the present invention;
FIG. 5 is a diagram of the data of the lying-down attitude angles in the embodiment of the invention;
FIG. 6 is a graph of acceleration data while standing in an embodiment of the present invention;
FIG. 7 is a graph of acceleration data while sitting in an embodiment of the present invention;
FIG. 8 is a graph of acceleration data during walking in an embodiment of the present invention;
FIG. 9 is a flow chart of a complementary filtering algorithm in an embodiment of the present invention;
fig. 10 is a flowchart of a human body posture detection and recognition method in an embodiment of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowcharts, in some cases, the steps illustrated or described may be performed in an order different than here.
Example one
As shown in fig. 1, the embodiment provides a human body posture detection and recognition system, which includes a data acquisition module, a transmission processing module, a posture resolving module, and a posture recognition module, which are connected in sequence; specifically, the method comprises the following steps:
the MPU60X0 is a device integrated with a 6-axis motion processing component in the first example of the world. The MPU6050 integrates a 3-axis MEMS gyroscope, a 3-axis MEMS accelerometer, and an expandable Digital Motion Processor DMP (Digital Motion Processor). The DMP is internally provided with a Kalman filtering algorithm, can acquire data of a gyroscope and an acceleration sensor, and processes and outputs quaternions. Six data (3-axis acceleration, 3-axis angular velocity) and quaternion of the MPU6050 can be read by I2C. The MPU-6050 can also program the measuring range of the gyroscope and the accelerometer sensor according to the requirement of a user, the full-grid sensing range of the 3-axis angular velocity sensor (gyroscope) is +/-250, +/-500, +/-1000, +/-2000dps, and the programmable measuring range of the 3-axis accelerometer is +/-2, +/-4, +/-8 +/-18 g.
The number of the sensors can cause certain influence on the result of detecting the human body posture, the number of the sensors for detecting and identifying the human body posture has two modes, one mode is a single sensor, the other mode is a multi-sensor, the multi-sensor easily causes data redundancy, and the wearing equipment relative to the multi-sensor is complicated and the manufacturing cost can be improved. The purpose of this design is to detect and identify simple body gestures such as standing, sitting, lying, walking and falling, so this design uses a single sensor approach.
By initializing the MPU6050 sensor, the measurement range of the three-axis gyro sensor is set to ± 2000dps, the measurement range of the three-axis acceleration sensor is set to ± 2g, and the sampling rate is 50Hz. The digital low-pass filter 4Hz is set through internal programming, and the data acquisition module obtains an original gyroscope value and an acceleration value through I2C communication.
The transmission processing module can convert the original acceleration value into the gravity acceleration through conversion processing, and convert the original gyroscope value into an angle.
The result after conversion is that the unit of the result after acceleration processing is +/-g, the acceleration measuring range is +/-4 g/s, the conversion relation is 8192LSB/g, namely, the corresponding reading of 1g is 8192. The gyroscope processed results are in units of + -g, the gyroscope measurement range is + -500 deg./s, and the conversion relationship is 65.5 LSB/deg., i.e., 1 deg./s corresponds to a reading of 65.536. In practical application, the acceleration is multiplied by 0.0174533f, and the aim is to convert the acceleration into the gravity acceleration.
The attitude resolving module performs attitude resolving according to the attitude coordinate data to obtain human body attitude, specifically:
as shown in fig. 2, the Z coordinate is not changed after the coordinate transformation. If the euler angle in three-dimensional space is rotated three times, a direction cosine matrix representing the rotation can be obtained.
The differential equation needs to be applied mechanically when the attitude is solved through the Euler angle, but the calculation of the differential equation comprises a large amount of trigonometric function calculation and inverse trigonometric function calculation, the calculation process is very complex, and the calculation amount is large. And the quaternion rule only needs to solve a linear differential equation set of quaternion, so that the calculation process is simple and the calculation amount is small. The direction cosine matrix is thus represented by a quaternion.
As shown in fig. 9, a complementary filtering algorithm is used, i.e. weighting of the values obtained by the sensors and the predicted values.
The quaternion method firstly needs to normalize the values acquired by the acceleration sensor, namely dividing a vector by a module, and inputting parameters including gyroscope data gx, gy and gz and acceleration data ax, ay and az.
In the second step, three elements vx, vy and vz in the third row in the direction cosine matrix are expressed by quaternions, namely the quaternions are converted into gravity unit vectors in a machine position coordinate system.
And thirdly, performing vector cross product compensation under the coordinate system of the machine position to obtain compensation. Wherein, ax/ay/az is the gravity vector measured by the sensor under the machine position coordinate system, and VX/VY/VZ is the gravity vector estimated by the attitude after the gyro is integrated. The error between them is the error between the attitude after the gyro integration and the attitude measured by the sensor. The error between vectors can be expressed by vector cross multiplication, the magnitude of the vector cross multiplication is in direct proportion to the integral error, and the error can be subjected to PI calculation by the vector cross multiplication to compensate the angular velocity.
And fourthly, solving a quaternion differential equation, wherein T is a measurement period, the quaternion differential equation is solved by using a first-order Longkoku tower, a new quaternion is updated, and finally the Euler angle is reversely solved by using the quaternion according to the conversion relation between the quaternion matrix and the Euler angle to obtain the Euler angle representing the upright or flat lying posture of the human body.
And the Euler angle, the acceleration and the angular velocity converted into a machine position coordinate system can be obtained after data preprocessing. Simple human body gestures can be recognized through the analysis of the data. Through experiments, data of the posture angle under different human body posture conditions can be obtained, as shown in fig. 3-5.
It can be found from the attitude angle that the change of the attitude angle is very stable in the actions of standing, sitting, lying down, and the numerical value is kept in a specific range, and therefore the attitude angle can be used as an effective characteristic value for distinguishing the standing state and the lying state of the human body. The upright state is divided into standing, sitting and walking, so other characteristic values need to be extracted for judgment. In order to accurately identify the human body posture and improve the accuracy, the acceleration data needs to be extracted with characteristic values such as a maximum value, a minimum value, a range, an average value, a variance and the like.
The mean value mean can determine the intensity of the data change.
The variance var may reflect the degree of fluctuation of the data offset average. The larger the variance, the larger the magnitude of the motion of the user.
The range of maximum data change over a period of time may be reflected by the range of maximum M.
Acceleration data of a user in standing, sitting and walking states can be easily obtained through experiments.
As can be seen from the experimental data in fig. 6 to 8, the vertical acceleration changes steadily when standing, changes when sitting down, and changes when standing up. The two postures of standing and sitting can be distinguished by the extreme difference and variance of the acceleration data in the vertical direction.
During walking, the acceleration change in the forward direction is found to be frequent, and the acceleration change in the forward direction can be identified through the variance of the acceleration in the forward direction. Meanwhile, the lateral angular velocity also shows regular change when the user walks, and the lateral angular velocity can be used as an effective characteristic value for identifying the walking state.
Example two
As shown in fig. 10, the present embodiment provides a method for detecting and recognizing a human body gesture, including:
acquiring human body motion data;
preprocessing the human body motion data to acquire attitude coordinate data;
performing attitude calculation based on the attitude coordinate data to obtain the human body attitude;
and extracting the characteristics of the human motion data to obtain the accurate human posture.
Optionally, the human motion data comprises an acceleration value and a raw gyroscope value;
the process of acquiring the human motion data comprises the following steps:
and acquiring the acceleration value and the original gyroscope value based on an acceleration sensor and a gyroscope respectively.
As a preferred embodiment of the present application, the attitude coordinate data includes a gravitational acceleration and an angular velocity value, and the angular velocity value includes a pitch angle velocity, a yaw angle velocity, and a roll angle velocity; the process of preprocessing the human motion data comprises the following steps:
and converting the acceleration value in the human motion data into the gravitational acceleration, and converting the original gyroscope value in the human motion data into the angular velocity value.
As a preferred embodiment of the present application, the human body posture includes an upright state and a lying state;
the process of performing attitude calculation based on the attitude coordinate data includes:
constructing a cosine matrix and expressing the cosine matrix based on a quaternion to obtain a quaternion cosine matrix;
normalizing the gravity acceleration in the attitude coordinate data to obtain an actual gravity acceleration;
acquiring a gravity unit vector under a machine position coordinate system based on the quaternion cosine matrix and the actual gravity acceleration;
integrating the angular velocity values in the attitude coordinate data to obtain a calculation gravity vector;
calculating the error between the calculated gravity vector and a gravity unit vector based on vector cross multiplication to obtain a gravity error, and compensating the angular velocity value based on the gravity error;
based on the compensated angular velocity value, solving the quaternion cosine matrix by adopting a first-order Longgoku tower;
and converting the solved quaternion cosine matrix into an attitude Euler angle, and acquiring the human body attitude based on the attitude Euler angle.
As a preferred embodiment of the present application, the process of performing feature extraction on the human motion data by the accurate human posture including sitting, standing, and walking includes:
extracting characteristic values of the human motion data, wherein the characteristic values comprise maximum acceleration, minimum acceleration, extreme difference of acceleration, average acceleration and acceleration variance;
distinguishing sitting and standing in the precise human body posture based on the acceleration variance, and distinguishing standing and walking in the precise human body posture based on the acceleration range.
In the present specification, the embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The method disclosed by the embodiment corresponds to the system disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the system part for description.
The above description is only for the preferred embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A human body posture detection and recognition system is characterized by comprising a data acquisition module, a transmission processing module, a posture resolving module and a posture recognition module which are connected in sequence;
the data acquisition module is used for acquiring human body motion data;
the transmission processing module is used for preprocessing the human body motion data, acquiring attitude coordinate data and transmitting the attitude coordinate data;
the attitude resolving module is used for resolving the attitude according to the attitude coordinate data to obtain the human body attitude;
the gesture recognition module is used for extracting the characteristics of the human motion data to obtain the accurate human gesture.
2. The human body posture detection and recognition system according to claim 1, wherein the human body motion data includes an acceleration value and an original gyroscope value, the data acquisition module includes an acceleration sensor and a gyroscope, and the acceleration sensor and the gyroscope are one in number and are arranged at the position of the abdomen of the human body; the acceleration sensor and the gyroscope are respectively used for acquiring the acceleration value and the original gyroscope value.
3. The human body posture detection and recognition system of claim 1, wherein the posture coordinate data comprises a gravity acceleration and an angular velocity value, the angular velocity value comprises a pitch angle velocity, a yaw angle velocity and a roll angle velocity; and the transmission processing module converts the acceleration value in the human motion data into the gravitational acceleration and converts the original gyroscope value in the human motion data into the angular velocity value.
4. The human body posture detection and recognition system according to claim 3, wherein the human body posture comprises an upright state, a lying state; the attitude resolving module acquires a quaternion cosine matrix by constructing a cosine matrix and expressing the cosine matrix based on a quaternion; normalizing the gravity acceleration in the attitude coordinate data to obtain an actual gravity acceleration; acquiring a gravity unit vector under a machine position coordinate system according to the quaternion cosine matrix and the actual gravity acceleration; integrating the angular velocity values in the attitude coordinate data to obtain a calculation gravity vector; calculating the error between the calculated gravity vector and a gravity unit vector according to vector cross multiplication to obtain a gravity error, and compensating the angular velocity value according to the gravity error; according to the compensated angular velocity value, solving the quaternion cosine matrix by adopting a first-order Longgoku tower; and converting the solved quaternion cosine matrix into an attitude Euler angle, and acquiring the human body attitude according to the attitude Euler angle.
5. The human gesture detection and recognition system of claim 1, wherein the precise human gesture includes sitting, standing, walking; the gesture recognition module extracts characteristic values from the human motion data, wherein the characteristic values comprise maximum acceleration, minimum acceleration, acceleration extreme difference, average acceleration and acceleration variance;
and distinguishing sitting and standing in the accurate human body posture according to the acceleration variance, and distinguishing standing and walking in the accurate human body posture according to the acceleration range.
6. A human body posture detection and recognition method is characterized by comprising the following steps:
acquiring human motion data;
preprocessing the human body motion data to acquire attitude coordinate data;
performing attitude calculation based on the attitude coordinate data to obtain the human body attitude;
and extracting the characteristics of the human motion data to obtain the accurate human posture.
7. The human body posture detection and recognition method according to claim 6, wherein the human body motion data includes an acceleration value and a raw gyroscope value;
the process of acquiring the human motion data comprises the following steps:
and acquiring the acceleration value and the original gyroscope value based on an acceleration sensor and a gyroscope respectively.
8. The human body posture detection and identification method according to claim 6, wherein the posture coordinate data comprises a gravitational acceleration and an angular velocity value, and the angular velocity value comprises a pitch angle velocity, a yaw angle velocity and a roll angle velocity; the process of preprocessing the human motion data comprises the following steps:
and converting the acceleration value in the human motion data into the gravitational acceleration, and converting the original gyroscope value in the human motion data into the angular velocity value.
9. The human body posture detection and recognition method according to claim 6, wherein the human body posture comprises an upright state and a lying state;
the process of performing attitude calculation based on the attitude coordinate data includes:
constructing a cosine matrix and expressing the cosine matrix based on a quaternion to obtain a quaternion cosine matrix;
normalizing the gravity acceleration in the attitude coordinate data to obtain an actual gravity acceleration;
acquiring a gravity unit vector under a machine position coordinate system based on the quaternion cosine matrix and the actual gravity acceleration;
integrating the angular velocity values in the attitude coordinate data to obtain a calculation gravity vector;
calculating the error between the calculated gravity vector and a gravity unit vector based on vector cross multiplication to obtain a gravity error, and compensating the angular velocity value based on the gravity error;
based on the compensated angular velocity value, solving the quaternion cosine matrix by adopting a first-order Longgoku tower;
and converting the solved quaternion cosine matrix into an attitude Euler angle, and acquiring the human body attitude based on the attitude Euler angle.
10. The human body posture detection and recognition method according to claim 6, wherein the accurate human body posture comprises sitting, standing and walking, and the process of performing the feature extraction on the human body motion data comprises the following steps:
extracting characteristic values from the human motion data, wherein the characteristic values comprise maximum acceleration, minimum acceleration, extreme difference of acceleration, average acceleration and acceleration variance;
distinguishing sitting and standing in the accurate human body posture based on the acceleration variance, and distinguishing standing and walking in the accurate human body posture based on the acceleration range.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211396929.1A CN115670445A (en) | 2022-11-09 | 2022-11-09 | Human body posture detection and recognition system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211396929.1A CN115670445A (en) | 2022-11-09 | 2022-11-09 | Human body posture detection and recognition system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115670445A true CN115670445A (en) | 2023-02-03 |
Family
ID=85049597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211396929.1A Pending CN115670445A (en) | 2022-11-09 | 2022-11-09 | Human body posture detection and recognition system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115670445A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104063537A (en) * | 2014-05-30 | 2014-09-24 | 北京控制工程研究所 | Multi-body dynamics parameter determination system based on distributive time trigger and method thereof |
| US20180231374A1 (en) * | 2017-02-13 | 2018-08-16 | National Tsing Hua University | Object pose measurement system based on mems imu and method thereof |
| CN108703760A (en) * | 2018-06-15 | 2018-10-26 | 安徽中科智链信息科技有限公司 | Human motion gesture recognition system and method based on nine axle sensors |
| WO2020253854A1 (en) * | 2019-06-21 | 2020-12-24 | 台州知通科技有限公司 | Mobile robot posture angle calculation method |
| CN113350771A (en) * | 2021-06-23 | 2021-09-07 | 北京赛博星通科技有限公司 | Athlete dynamic posture recognition method, device, system and storage medium |
-
2022
- 2022-11-09 CN CN202211396929.1A patent/CN115670445A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104063537A (en) * | 2014-05-30 | 2014-09-24 | 北京控制工程研究所 | Multi-body dynamics parameter determination system based on distributive time trigger and method thereof |
| US20180231374A1 (en) * | 2017-02-13 | 2018-08-16 | National Tsing Hua University | Object pose measurement system based on mems imu and method thereof |
| CN108703760A (en) * | 2018-06-15 | 2018-10-26 | 安徽中科智链信息科技有限公司 | Human motion gesture recognition system and method based on nine axle sensors |
| WO2020253854A1 (en) * | 2019-06-21 | 2020-12-24 | 台州知通科技有限公司 | Mobile robot posture angle calculation method |
| CN113350771A (en) * | 2021-06-23 | 2021-09-07 | 北京赛博星通科技有限公司 | Athlete dynamic posture recognition method, device, system and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 吴聪;刘俊;: "四元数互补滤波的机械手手势姿态解算", 上海电机学院学报, no. 03, 25 June 2018 (2018-06-25) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hsu et al. | Human daily and sport activity recognition using a wearable inertial sensor network | |
| JP5674766B2 (en) | Sensing device for detecting wearing position | |
| Woodward et al. | Pervasive monitoring of motion and muscle activation: Inertial and mechanomyography fusion | |
| US9159213B1 (en) | Motion detection method and device | |
| CN109846487B (en) | Thigh movement posture measuring method and device based on MIMU/sEMG fusion | |
| Wade et al. | Design and implementation of an instrumented cane for gait recognition | |
| Fan et al. | Wearable motion attitude detection and data analysis based on Internet of Things | |
| Yang et al. | Smart wearable monitoring system based on multi-type sensors for motion recognition | |
| Yan et al. | Accurate fall detection by nine-axis IMU sensor | |
| CN114469073B (en) | Gait analysis and abnormality detection method based on wearable sensor | |
| CN113331829B (en) | Sole information monitoring method and intelligent insole device | |
| CN112762934B (en) | Lower limb movement direction prediction device and method | |
| CN113627500A (en) | Gait information acquisition system based on multi-sensor fusion and recognition algorithm | |
| CN115670445A (en) | Human body posture detection and recognition system and method | |
| Murgia et al. | A portable device for the clinical assessment of upper limb motion and muscle synergies | |
| Abyarjoo et al. | Monitoring human wrist rotation in three degrees of freedom | |
| Lee et al. | Development of wearable device and clustering based method for detecting falls in the elderly | |
| JP7172733B2 (en) | Human Posture Estimation Device | |
| Fahriddin et al. | Human activity recognition using new multi-sensor module in mobile environment | |
| CN109692004B (en) | Method and device for detecting standing posture of human body | |
| Ong et al. | BrainSmart: Ambient Assisted Living System Smartphone App Prototype for Parkinson's Disease Patients | |
| Tinh et al. | Image-based gramian angular field processing for pedestrian stride-length estimation using convolutional neural network | |
| CN109061231B (en) | Riding speed calculation method and device | |
| Liu et al. | A wearable sensor system for human motion analysis and humanoid robot control | |
| Xu et al. | A Wearable Micro-Electromechanical System Inertial Sensor System for Fall Behaviour Detection Based on a Multi-Level Threshold Algorithm |
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 |