CN115778378A - Fall identification method and device - Google Patents
Fall identification method and device Download PDFInfo
- Publication number
- CN115778378A CN115778378A CN202211607593.9A CN202211607593A CN115778378A CN 115778378 A CN115778378 A CN 115778378A CN 202211607593 A CN202211607593 A CN 202211607593A CN 115778378 A CN115778378 A CN 115778378A
- Authority
- CN
- China
- Prior art keywords
- acceleration
- human body
- human
- data
- falling
- 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
Images
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention provides a falling identification method and device, and relates to the technical field of falling detection. The method comprises the steps of acquiring human motion data including acceleration and heart rate data; then, calculating to obtain a resultant acceleration according to the acceleration in the human motion data; then, determining the motion state of the human body according to the resultant acceleration and the heart rate data; resolving the human motion data by adopting a quaternion attitude resolving method according to the human motion state to obtain human attitude parameters; and finally, inputting the human body posture parameters and the human body motion data into a preset falling recognition model, wherein the preset falling recognition model comprises a feature converter constructed based on a DBN (digital broadcast network) deep belief network algorithm and a classifier constructed based on an RF (radio frequency) random forest algorithm, the feature converter performs feature conversion on the human body posture parameters to obtain a feature matrix, and the feature matrix is input into the classifier to obtain a falling recognition result. The fall recognition method and the fall recognition device can obtain more satisfactory accuracy and ensure the accuracy of fall recognition.
Description
Technical Field
The invention relates to the technical field of fall detection, in particular to a fall identification method and device.
Background
With the trend of more serious aging of population, the solution of the problem of old people is more urgent, although the welfare system is continuously improved, the sudden events of death of the old people caused by falling down still frequently occur, and therefore the prevention of the falling down of the old people becomes more important in the actual life.
At present, with the continuous aggravation of the aging of China, the health monitoring of the old people becomes a key problem concerned by the whole society, a falling detection device is also one of the research hotspots, and numerous scientific researchers and funds at home and abroad are dedicated to the research and development of the falling detection device or the algorithm. According to the data acquisition mode and the use condition of the fall detection device, fall detection can be divided into an environmental signal fall detection technology, a video image fall detection technology and a wearable equipment fall detection technology.
However, all of these fall detection devices need to design an algorithm for fall identification, and the existing algorithm has high identification difficulty, low algorithm accuracy and sometimes false alarm.
Disclosure of Invention
The invention aims to provide a fall identification method and a fall identification device, which are used for solving the problem that in the prior art, the difference of evaluation indexes in different areas is not considered, so that the evaluation is not accurate enough.
In a first aspect, an embodiment of the present application provides a fall identification method, including the following steps:
acquiring human body motion data, wherein the human body motion data at least comprises acceleration data and heart rate data;
calculating to obtain a resultant acceleration according to the acceleration in the human motion data;
determining the motion state of the human body according to the resultant acceleration and the heart rate data;
resolving the human motion data by adopting a quaternion attitude resolving method according to the human motion state to obtain human attitude parameters;
inputting the human body posture parameters and the human body motion data into a preset falling identification model, wherein the preset falling identification model comprises a feature converter constructed based on a DBN (digital base network) depth belief network algorithm and a classifier constructed based on an RF (radio frequency) random forest algorithm, the feature converter performs feature conversion on the human body posture parameters to obtain a feature matrix, and the classifier performs classification according to the feature matrix to obtain a falling identification result.
In some embodiments of the invention according to the first aspect, the acceleration comprises an x-axis acceleration a x Y-axis acceleration a y And z-axis acceleration a z ；
The step of calculating the resultant acceleration according to the acceleration in the human motion data comprises the following steps:
the x-axis acceleration a x Y-axis acceleration a y And z-axis acceleration a z Substituting into a resultant acceleration calculation formula to obtain a resultant acceleration, wherein the resultant acceleration calculation formula is as follows:
in the formula: a is the resultant acceleration, A 1 Acceleration in the x-axis, A 2 Acceleration in the y-axis, A 3 Is the z-axis acceleration.
Based on the first aspect, in some embodiments of the present invention, the step of determining the motion state of the human body according to the resultant acceleration and the heart rate data comprises the steps of:
comparing the resultant acceleration with a preset acceleration threshold value to obtain a first comparison result;
comparing the heart rate data with a preset heart rate threshold value to obtain a second comparison result;
and obtaining the motion state of the human body according to the first comparison result and the second comparison result.
Based on the first aspect, in some embodiments of the present invention, the calculating the human motion data by using a quaternion attitude calculation method according to the human motion state to obtain human attitude parameters includes the following steps:
judging whether the human motion state is violent motion, if so, resolving the human motion data by adopting a quaternion attitude resolving method to obtain human attitude parameters; if not, generating normal activity reminding information.
Based on the first aspect, in some embodiments of the present invention, the method further comprises the following steps:
obtaining a sample data set, wherein each sample in the sample data set comprises a sample characteristic matrix and a sample category;
and according to the sample data set, taking the sample characteristic matrix as input and the sample category as output, training an initial classifier constructed by the RF random forest algorithm, and obtaining the classifier constructed based on the RF random forest algorithm.
Based on the first aspect, in some embodiments of the present invention, the method further comprises the following steps:
generating a control command according to the falling identification result;
and sending the control command to a falling prevention device to realize falling protection.
Based on the first aspect, in some embodiments of the present invention, the method further comprises the following steps:
and generating falling reminding information according to the falling identification result, and sending the falling reminding information to the user side.
In a second aspect, an embodiment of the present application provides a fall identification apparatus, including:
the data acquisition module is used for acquiring human motion data, and the human motion data at least comprises acceleration data and heart rate data;
the resultant acceleration calculation module is used for calculating to obtain resultant acceleration according to the acceleration in the human motion data;
the motion state determining module is used for determining the motion state of the human body according to the combined acceleration and the heart rate data;
the posture resolving module is used for resolving the human body motion data by adopting a quaternion resolving posture method according to the human body motion state to obtain human body posture parameters;
and the falling identification model module is used for inputting the human body posture parameters and the human body motion data into a preset falling identification model, the preset falling identification model comprises a feature converter constructed based on a DBN (direct bus) depth belief network algorithm and a classifier constructed based on an RF (radio frequency) random forest algorithm, the feature converter performs feature conversion on the human body posture parameters to obtain a feature matrix, and the classifier performs classification according to the feature matrix to obtain a falling identification result.
In a third aspect, an embodiment of the present application provides an electronic device, which includes a memory for storing one or more programs; a processor. The one or more programs, when executed by the processor, implement the method of any of the first aspects as described above.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the method according to any one of the above first aspects.
The embodiment of the invention at least has the following advantages or beneficial effects:
the embodiment of the invention provides a fall identification method and a fall identification device, wherein human motion data is obtained and at least comprises acceleration data and heart rate data; then, calculating to obtain a resultant acceleration according to the acceleration in the human motion data; then determining the motion state of the human body according to the resultant acceleration and the heart rate data; resolving the human motion data by adopting a quaternion attitude resolving method according to the human motion state to obtain human attitude parameters; and finally, inputting the human body posture parameters and the human body motion data into a preset falling identification model, wherein the preset falling identification model comprises a feature converter constructed based on a DBN (direct bus network) depth belief network algorithm and a classifier constructed based on an RF (radio frequency) random forest algorithm, the feature converter performs feature conversion on the human body posture parameters to obtain a feature matrix, and the classifier performs classification according to the feature matrix to obtain a falling identification result. The method comprises the steps of determining the motion state of a human body, accurately judging the state of the human body, performing suspected falling judgment, further resolving human body attitude parameters according to the motion state of the human body, combining the human body attitude parameters and human body motion data to serve as input of a falling identification model, and judging whether falling occurs, wherein the falling identification model adopts a feature converter constructed based on a DBN (deep belief network) algorithm and a classifier constructed based on an RF (radio frequency) random forest algorithm, and utilizes the DBN to perform nonlinear feature conversion, so that better features are introduced, more satisfactory accuracy can be obtained, and the accuracy of falling identification is ensured.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a flowchart of a fall recognition method according to an embodiment of the present invention;
fig. 2 is a block diagram of a fall identification apparatus according to an embodiment of the present invention;
fig. 3 is a block diagram of an electronic device according to an embodiment of the present invention.
Icon: 110-a data acquisition module; 120-resultant acceleration calculation module; 130-a motion state determination module; 140-attitude resolving module; 150-a fall identification model module; 101-a memory; 102-a processor; 103-a communication interface.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of additional like elements in a process, method, article, or apparatus that comprises the element.
In the description of the present application, it should be noted that the terms "upper", "lower", "inner", "outer", and the like indicate orientations or positional relationships based on orientations or positional relationships shown in the drawings or orientations or positional relationships conventionally found in use of products of the application, and are used only for convenience in describing the present application and for simplification of description, but do not indicate or imply that the referred devices or elements must have a specific orientation, be constructed in a specific orientation, and be operated, and thus should not be construed as limiting the present application.
Examples
Some embodiments of the present application will be described in detail below with reference to the accompanying drawings. The embodiments and features of the embodiments described below can be combined with one another without conflict.
Referring to fig. 1, fig. 1 is a flowchart of a fall identification method according to an embodiment of the invention. The fall identification method comprises the following steps:
step S110: acquiring human body motion data, wherein the human body motion data at least comprises acceleration data and heart rate data; in this embodiment, human motion data still includes angular velocity, and above-mentioned angular velocity and acceleration can be obtained through wearing the equipment of preventing tumbleing on one's body in the human body, and the above-mentioned equipment of preventing tumbleing can be wearable undershirt, and inside is provided with the sensor, can real-time data collection, and above-mentioned sensor can be the nine motion sensor that adopts the model to be MPU9250, the inside integration of nine motion sensor has triaxial accelerometer, triaxial gyroscope and triaxial magnetometer, can solve for subsequent human gesture and provide original nine sensor data. Above-mentioned rhythm of heart data can be obtained through wearable bracelet real-time supervision, also can be through the aforesaid prevent rhythm of the heart check out test set real-time supervision among the equipment of tumbleing and obtain. It should be noted that the human motion data may be real-time data or acquired at certain intervals, and if the human motion data is acquired at certain intervals, the human motion data further includes time.
Step S120: calculating to obtain a resultant acceleration according to the acceleration in the human motion data;
wherein, for the acceleration collected by the three-axis accelerometer, the acceleration includes the acceleration a of the x axis x Y-axis acceleration a y And z-axis acceleration a z (ii) a The resultant acceleration can be calculated by a resultant acceleration calculation formula. The method specifically comprises the following steps:
the x-axis acceleration a is measured x Y-axis acceleration a y And z-axis acceleration a z Substituting into a resultant acceleration calculation formula to obtain a resultant acceleration, wherein the resultant acceleration calculation formula is as follows:
in the formula: a is the resultant acceleration, A 1 Acceleration in the x-axis, A 2 Acceleration in the y-axis, A 3 Is the z-axis acceleration.
The resultant acceleration is obtained according to the above calculation formula
Step S130: determining the motion state of the human body according to the resultant acceleration and the heart rate data; in this embodiment, the resultant acceleration can be used to distinguish the motion state of the human body, and the larger the resultant acceleration, the more intense the motion is, and conversely, the smoother the motion of the human body is. In order to further improve the accuracy rate of determining the motion state of the human body, heart rate data can be added for determination. The method specifically comprises the following steps:
firstly, comparing the resultant acceleration with a preset acceleration threshold to obtain a first comparison result; the acceleration threshold can be set according to actual conditions.
Then, comparing the heart rate data with a preset heart rate threshold value to obtain a second comparison result; the preset heart rate threshold value can be set by the user according to the body condition of the user, so that the judgment rate is improved, a preset default value can be adopted, and the default value can be set based on experience.
And finally, obtaining the motion state of the human body according to the first comparison result and the second comparison result. In this embodiment, if the first comparison result is greater than the acceleration threshold and the second comparison result is greater than the heart rate threshold, it indicates that a violent movement may occur, and it is possible to obtain that the movement state of the human body is a violent movement; in other cases, the user does not exercise violently, and the exercise state of the human body can be obtained as normal activity.
The combined acceleration and the heart rate data are combined to determine the human motion state, so that the accuracy of determining the human motion state can be improved, and the human motion state determination deviation caused by data acquisition errors is avoided.
Step S140: resolving the human motion data by adopting a quaternion attitude resolving method according to the human motion state to obtain human attitude parameters; under the condition of violent movement, the probability of falling is higher, so after the movement state of the human body is determined, suspected falling judgment can be carried out according to the movement state of the human body, and the method specifically comprises the following steps:
judging whether the human motion state is violent motion, if so, resolving the human motion data by adopting a quaternion attitude resolving method to obtain human attitude parameters; if not, generating normal activity reminding information. If the motion state of the human body is not violent motion and belongs to normal activity, normal activity reminding information can be generated and sent to the user, so that the user can check the motion state conveniently. The user can be a remote user, and the user accesses the server to obtain the reminding information by uploading the reminding information of the normal activities to the server, so that the user can know the motion state of related personnel in time.
In the present embodiment, when the euler angle represents the posture change, three rotations around three coordinate axes i, j, and k are represented. According to the Euler's theorem, these three rotations are equivalent to one rotation around a certain axis. The method for resolving the attitude by adopting the quaternion is to set the quaternion to represent the rotation of a three-dimensional space, and the basic representation form of the quaternion is as follows:
Q(q 0 ,q 1 ,q 2 ,q 3 )＝q 0 +q 1 i+q 2 j+q 3 k, wherein i rotation represents the rotation of the X axis in the X axis and Y axis intersection plane to the positive direction of the Y axis, j rotation represents the rotation of the Z axis in the Z axis and X axis intersection plane to the positive direction of the X axis, k rotation represents the rotation of the Y axis in the Y axis and Z axis intersection plane to the positive direction of the Z axis, and the rotation is used for q axis positive direction rotation in posture representation 0 ＝cos(θ/2)，q n ＝u n *sin(θ/2)，n＝1，2，3，u n Represents a certain axis, and theta is a rotation angle, and can be obtained according to the acceleration in the human motion data. Then, resolving the quaternion to obtain a human body posture parameter, wherein the human body posture parameter comprises an angle parameter roll, an angle parameter pitch and an angle parameter yaw, and can be expressed as:
step S150: inputting the human body posture parameters and the human body motion data into a preset falling recognition model, wherein the preset falling recognition model comprises a feature converter constructed based on a DBN (digital broadcast network) deep belief network algorithm and a classifier constructed based on an RF (radio frequency) random forest algorithm, the feature converter performs feature conversion on the human body posture parameters to obtain a feature matrix, and the classifier performs classification according to the feature matrix to obtain a falling recognition result.
In this embodiment, the human body posture parameters and the human body motion data may be constructed into an initial feature matrix, and the initial feature matrix includes an angle parameter roll, an angle parameter pitch, an angle parameter yaw in the human body posture parameters, and an acceleration, heart rate data, and an angular velocity in the human body motion data.
The feature converter constructed based on the DBN deep belief network algorithm may be that K limited boltzmann machines are set in the DBN network, feature transformation is performed on a feature matrix in an original feature domain to obtain a corresponding feature matrix in the K regions, and the corresponding feature matrix in the K regions is used as a new feature matrix for training or testing. Wherein, the output characteristic matrix of the mth restricted boltzmann machine is expressed as: f m ＝F m-1 W m +B m M =1,2, \8230;, K, wherein W m Weight matrix for the mth restricted Boltzmann machine, B m Is the deviation matrix of the mth restricted boltzmann machine. The specific value of K can be confirmed according to the recognition accuracy, and when the DBN is used for feature conversion, the performance of the DBN can be optimized by adjusting the iteration period, the momentum coefficient, the learning rate and the network structure of the DBN.
After feature transformation is carried out through a feature transformer based on a DBN deep belief network, classification and identification are carried out through a classifier based on an RF random forest algorithm. The classifier can be obtained by sample data training, and specifically comprises the following steps:
firstly, acquiring a sample data set, wherein each sample in the sample data set comprises a sample characteristic matrix and a sample category; the sample data set may be a historical data set, including data sets in which a fall has occurred and data sets in which no fall has occurred. The sample feature matrix is a feature matrix converted by a feature converter constructed by a DBN deep belief network algorithm, and the sample category can be fall or normal.
And then, according to a sample data set, taking the sample characteristic matrix as input and the sample category as output, training an initial classifier constructed by the RF random forest algorithm, and obtaining the classifier constructed based on the RF random forest algorithm.
In this embodiment, the classifier includes a plurality of decision trees and a voting selection, the decision trees adopt classification regression trees based on a CART algorithm, the optimal number of the decision trees is set to 500, the decision results are defined as two categories, i.e., fall and normal, the voting selection performs voting according to 500 decision results, cross validation can be performed on training samples by using a leave-one-out method, meanwhile, on the basis of the LOO algorithm, statistics can be performed on sample test results by using a leave-one-subject-out method, and if most of the decision trees are determined to fall, the identification result is selected to be considered as fall; otherwise, the recognition result is normal.
In the implementation process, the human body movement data is acquired, and at least comprises acceleration data and heart rate data; then, calculating to obtain a resultant acceleration according to the acceleration in the human motion data; then, determining the motion state of the human body according to the combined acceleration and the heart rate data; resolving the human motion data by adopting a quaternion attitude resolving method according to the human motion state to obtain human attitude parameters; and finally, inputting the human body posture parameters and the human body motion data into a preset falling recognition model, wherein the preset falling recognition model comprises a feature converter constructed based on a DBN (digital broadcast network) deep belief network algorithm and a classifier constructed based on an RF (radio frequency) random forest algorithm, the feature converter performs feature conversion on the human body posture parameters to obtain a feature matrix, and the classifier performs classification according to the feature matrix to obtain a falling recognition result. The method comprises the steps of determining the motion state of a human body, accurately judging the state of the human body, performing suspected falling judgment, further resolving human body attitude parameters according to the motion state of the human body, combining the human body attitude parameters and human body motion data to serve as input of a falling identification model, and judging whether falling occurs, wherein the falling identification model adopts a feature converter constructed based on a DBN (deep belief network) algorithm and a classifier constructed based on an RF (radio frequency) random forest algorithm, and utilizes the DBN to perform nonlinear feature conversion, so that better features are introduced, more satisfactory accuracy can be obtained, and the accuracy of falling identification is ensured.
After obtaining the fall identification result, the fall protection operation can be performed, and the method specifically comprises the following steps:
firstly, generating a control command according to the falling identification result; in the present embodiment, when the fall recognition result is a fall, a control command to turn on the fall prevention device is generated.
And then, sending the control command to a falling prevention device to realize falling protection. Prevent falling down the device and after receiving control command, open and prevent falling down the device, fall down the protection, above-mentioned prevent falling down the device and belong to prior art, for example, prevent falling down the device and install high-pressure gas cylinder and pass through first trachea and a plurality of air bag intercommunication, at implementation in-process, prevent falling down the device and still be provided with the inlayer from inside to outside, flexible buffer layer and skin, the gasbag is installed in the skin, air bag can open automatically fast, flexible buffer layer is used for buffering the produced pressure of first air bag aerifing in the twinkling of an eye, in order to prevent that the gasbag from leading to the fact accidental injury to the human body at the instantaneous inflation in-process. After receiving the control command, the falling prevention device opens the high-pressure gas cylinder to inflate the first safety airbag, so that falling protection is achieved.
Wherein, still include the following step:
and generating falling reminding information according to the falling identification result, and sending the falling reminding information to the user side. In this embodiment, can be through the application of radio communication function, when the discernment result of tumbleing is for tumbleing, generate the warning information of tumbleing to send to the user terminal, above-mentioned user terminal can be urgent contact person, also can be medical institution etc. can in time know the accident of tumbleing through the warning information of tumbleing, and then can in time take measures, in order to reduce the follow-up injury that tumbles and lead to.
It should be noted that, if a false alarm occurs, the information for reminding the fall can be manually cancelled within a certain time, so as to reduce unnecessary panic caused by the false alarm.
Based on the same inventive concept, the invention further provides a fall recognition device, please refer to fig. 2, and fig. 2 is a block diagram of a fall recognition device according to an embodiment of the invention. The fall recognition device includes:
a data acquiring module 110, configured to acquire human motion data, where the human motion data at least includes acceleration data and heart rate data;
a resultant acceleration calculation module 120, configured to calculate a resultant acceleration according to an acceleration in the human motion data;
a motion state determining module 130, configured to determine a motion state of the human body according to the combined acceleration and the heart rate data;
the posture resolving module 140 is used for resolving the human body motion data by adopting a quaternion resolving posture method according to the human body motion state to obtain human body posture parameters;
the fall recognition model module 150 is configured to input the human body posture parameters into a preset fall recognition model, where the preset fall recognition model includes a feature converter constructed based on a DBN deep belief network algorithm and a classifier constructed based on an RF random forest algorithm, the feature converter performs feature conversion on the human body posture parameters to obtain a feature matrix, and the classifier performs classification according to the feature matrix to obtain a fall recognition result.
In the implementation process, the data acquisition module 110 is used for acquiring human body motion data, wherein the human body motion data at least comprises acceleration data and heart rate data; the resultant acceleration calculation module 120 calculates to obtain a resultant acceleration according to the acceleration in the human motion data; the motion state determination module 130 determines the motion state of the human body according to the combined acceleration and the heart rate data; the posture resolving module 140 resolves the human body motion data by adopting a quaternion resolving posture method according to the human body motion state to obtain human body posture parameters; the fall recognition model module 150 inputs the human body posture parameters and the human body motion data into a preset fall recognition model, the preset fall recognition model comprises a feature converter constructed based on a DBN deep belief network algorithm and a classifier constructed based on an RF random forest algorithm, the feature converter performs feature conversion on the human body posture parameters to obtain a feature matrix, and the classifier performs classification according to the feature matrix to obtain a fall recognition result. The method comprises the steps of determining the motion state of a human body, accurately judging the state of the human body, judging suspected falling, further calculating the posture parameters of the human body according to the motion state of the human body, combining the posture parameters of the human body and the motion data of the human body as the input of a falling recognition model to judge whether falling occurs, and performing nonlinear feature transformation by using a DBN (database network node) by the falling recognition model by using a feature converter constructed based on a DBN (deep belief network) algorithm and a classifier constructed based on an RF (radio frequency) random forest algorithm, so that better features are introduced, more satisfactory accuracy can be obtained, and the accuracy of falling recognition is ensured.
Referring to fig. 3, fig. 3 is a schematic structural block diagram of an electronic device according to an embodiment of the present disclosure. The electronic device comprises a memory 101, a processor 102 and a communication interface 103, wherein the memory 101, the processor 102 and the communication interface 103 are electrically connected to each other directly or indirectly to realize data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The memory 101 may be used to store software programs and modules, such as program instructions/modules corresponding to the fall recognition device 100 provided in the embodiments of the present application, and the processor 102 executes various functional applications and data processing by executing the software programs and modules stored in the memory 101. The communication interface 103 may be used for communicating signaling or data with other node devices.
The Memory 101 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like.
The processor 102 may be an integrated circuit chip having signal processing capabilities. The processor 102 may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.
It will be appreciated that the configuration shown in fig. 3 is merely illustrative and that the electronic device may include more or fewer components than shown in fig. 3 or have a different configuration than shown in fig. 3. The components shown in fig. 3 may be implemented in hardware, software, or a combination thereof.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made to the present application by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Claims (10)
1. A fall identification method, comprising the steps of:
acquiring human body motion data, wherein the human body motion data at least comprises acceleration data and heart rate data;
calculating to obtain a resultant acceleration according to the acceleration in the human motion data;
determining the motion state of the human body according to the resultant acceleration and the heart rate data;
resolving the human motion data by adopting a quaternion attitude resolving method according to the human motion state to obtain human attitude parameters;
inputting the human body posture parameters and the human body motion data into a preset falling identification model, wherein the preset falling identification model comprises a feature converter constructed based on a DBN (digital base network) depth belief network algorithm and a classifier constructed based on an RF (radio frequency) random forest algorithm, the feature converter performs feature conversion on the human body posture parameters to obtain a feature matrix, and the classifier performs classification according to the feature matrix to obtain a falling identification result.
2. Fall identification method according to claim 1, characterized in that the acceleration comprises an x-axis acceleration a x Y-axis acceleration a y And z-axis acceleration a z ；
The step of calculating the resultant acceleration according to the acceleration in the human motion data comprises the following steps:
the x-axis acceleration a x Y-axis acceleration a y And z-axis acceleration a z Substituting into a resultant acceleration calculation formula to obtain a resultant acceleration, wherein the resultant acceleration calculation formula is as follows:
in the formula: a is the resultant acceleration, A 1 Acceleration of the x-axis, A 2 Acceleration of the y-axis, A 3 Is the z-axis acceleration.
3. A fall recognition method as claimed in claim 1, wherein the step of determining the state of motion of the person from the resultant acceleration and heart rate data comprises the steps of:
comparing the resultant acceleration with a preset acceleration threshold value to obtain a first comparison result;
comparing the heart rate data with a preset heart rate threshold value to obtain a second comparison result;
and obtaining the motion state of the human body according to the first comparison result and the second comparison result.
4. The fall recognition method according to claim 1, wherein the step of solving the human motion data by a quaternion attitude solving method according to the human motion state to obtain human attitude parameters comprises the steps of:
judging whether the human motion state is violent motion, if so, resolving the human motion data by adopting a quaternion attitude resolving method to obtain human attitude parameters; if not, generating normal activity reminding information.
5. A fall identification method as claimed in claim 1, further comprising the steps of:
acquiring a sample data set, wherein each sample in the sample data set comprises a sample characteristic matrix and a sample category;
and according to a sample data set, taking the sample characteristic matrix as input and the sample category as output, training an initial classifier constructed by the RF random forest algorithm, and obtaining the classifier constructed based on the RF random forest algorithm.
6. A fall identification method according to claim 1, further comprising the steps of:
generating a control command according to the falling identification result;
and sending the control command to a falling prevention device to realize falling protection.
7. Fall identification method according to claim 6, characterized in that it further comprises the steps of:
and generating falling reminding information according to the falling identification result, and sending the falling reminding information to the user side.
8. A fall recognition device, comprising:
the data acquisition module is used for acquiring human motion data, and the human motion data at least comprises acceleration data and heart rate data;
the resultant acceleration calculation module is used for calculating to obtain resultant acceleration according to the acceleration in the human motion data;
the motion state determining module is used for determining the motion state of the human body according to the combined acceleration and the heart rate data;
the posture resolving module is used for resolving the human body motion data by adopting a quaternion resolving posture method according to the human body motion state to obtain human body posture parameters;
and the falling identification model module is used for inputting the human body posture parameters and the human body motion data into a preset falling identification model, the preset falling identification model comprises a feature converter constructed based on a DBN (digital broadcast network) deep belief network algorithm and a classifier constructed based on an RF (radio frequency) random forest algorithm, the feature converter performs feature transformation on the human body posture parameters to obtain a feature matrix, and the classifier performs classification according to the feature matrix to obtain a falling identification result.
9. An electronic device, comprising:
a memory for storing one or more programs;
a processor;
fall identification method according to any of claims 1-7, when the one or more programs are executed by the processor.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out a fall identification method as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211607593.9A CN115778378A (en) | 2022-12-14 | 2022-12-14 | Fall identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211607593.9A CN115778378A (en) | 2022-12-14 | 2022-12-14 | Fall identification method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115778378A true CN115778378A (en) | 2023-03-14 |
Family
ID=85419329
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211607593.9A Pending CN115778378A (en) | 2022-12-14 | 2022-12-14 | Fall identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115778378A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116439694A (en) * | 2023-06-14 | 2023-07-18 | 深圳市魔样科技有限公司 | Intelligent watch dynamic data monitoring method based on motion model training |
-
2022
- 2022-12-14 CN CN202211607593.9A patent/CN115778378A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116439694A (en) * | 2023-06-14 | 2023-07-18 | 深圳市魔样科技有限公司 | Intelligent watch dynamic data monitoring method based on motion model training |
| CN116439694B (en) * | 2023-06-14 | 2023-08-15 | 深圳市魔样科技有限公司 | Intelligent watch dynamic data monitoring method based on motion model training |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bi et al. | Safewatch: A wearable hand motion tracking system for improving driving safety | |
| Lee et al. | Development of an enhanced threshold-based fall detection system using smartphones with built-in accelerometers | |
| Warwick et al. | Detecting driver drowsiness using wireless wearables | |
| US8786448B2 (en) | State detecting device, state detecting method, and non-transitory computer-readable medium | |
| CN111166357A (en) | Fatigue monitoring device system with multi-sensor fusion and monitoring method thereof | |
| JP6064280B2 (en) | System and method for recognizing gestures | |
| CN105708470B (en) | Based on doppler detector and the united fall detection system of sensor and method | |
| US9801569B2 (en) | Method of detecting activity with a motion sensor, and corresponding device and computer program | |
| US20170372582A1 (en) | Device for monitoring and preventing slump and processing method thereof | |
| Li et al. | A novel learning model of driver fatigue features representation for steering wheel angle | |
| He et al. | A low power fall sensing technology based on FD-CNN | |
| US20150227837A1 (en) | System monitoring | |
| Kansiz et al. | Selection of time-domain features for fall detection based on supervised learning | |
| CN105528859A (en) | Nursing intelligent alarm system based on human body falling down detection technology | |
| CN107169512B (en) | HMM-SVM falling model construction method and falling detection method based on model | |
| Shi et al. | Fall Detection Algorithm Based on Triaxial Accelerometer and Magnetometer. | |
| CN115778378A (en) | Fall identification method and device | |
| CN111743545B (en) | Deep learning-based old man fall detection method, detection bracelet and storage medium | |
| Cao et al. | A fall detection method based on acceleration data and hidden Markov model | |
| Gjoreski et al. | Context-based fall detection and activity recognition using inertial and location sensors | |
| Zeng et al. | In-vehicle sensing for smart cars | |
| Parmar et al. | A Comprehensive Survey of Various Approaches on Human Fall Detection for Elderly People | |
| Kambhampati et al. | Unified framework for triaxial accelerometer‐based fall event detection and classification using cumulants and hierarchical decision tree classifier | |
| Postawka et al. | Lifelogging system based on Averaged Hidden Markov Models: dangerous activities recognition for caregivers support | |
| Wu et al. | Fall detection monitoring system based on mems sensor |
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 |