CN117556333A - Falling detection method and device, electronic equipment and storage medium - Google Patents
Falling detection method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117556333A CN117556333A CN202311741975.5A CN202311741975A CN117556333A CN 117556333 A CN117556333 A CN 117556333A CN 202311741975 A CN202311741975 A CN 202311741975A CN 117556333 A CN117556333 A CN 117556333A
- Authority
- CN
- China
- Prior art keywords
- acceleration
- determining
- target
- value
- detected
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 116
- 238000003860 storage Methods 0.000 title claims abstract description 18
- 230000001133 acceleration Effects 0.000 claims abstract description 348
- 230000008859 change Effects 0.000 claims abstract description 117
- 238000006073 displacement reaction Methods 0.000 claims abstract description 49
- 238000000034 method Methods 0.000 claims abstract description 49
- 230000033001 locomotion Effects 0.000 claims abstract description 46
- 238000009825 accumulation Methods 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims description 66
- 230000006870 function Effects 0.000 claims description 30
- 238000012706 support-vector machine Methods 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 16
- 230000001186 cumulative effect Effects 0.000 claims description 5
- 230000001681 protective effect Effects 0.000 claims description 2
- 239000000758 substrate Substances 0.000 claims 1
- 238000012545 processing Methods 0.000 abstract description 25
- 238000011897 real-time detection Methods 0.000 abstract description 4
- 238000012360 testing method Methods 0.000 description 26
- 238000004364 calculation method Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 9
- 238000004891 communication Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 230000010354 integration Effects 0.000 description 8
- 239000011159 matrix material Substances 0.000 description 7
- 238000010276 construction Methods 0.000 description 4
- 230000002829 reductive effect Effects 0.000 description 4
- 230000009466 transformation Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000037081 physical activity Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000008054 signal transmission Effects 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 229910052742 iron Inorganic materials 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000000284 resting effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000036962 time dependent Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a falling detection method, a falling detection device, electronic equipment and a storage medium; the method comprises the following steps: acquiring acceleration data of an object to be detected, and determining acceleration values corresponding to different direction axes according to the acceleration data; determining the amplitude change sum of the acceleration, the amplitude change rate of the acceleration, the total acceleration accumulation change rate and the total acceleration change displacement according to each acceleration value, and forming an acceleration characteristic data set; the acceleration characteristic data set is input into a pre-trained target detection model, whether the motion state of the target to be detected falls is determined according to the output result of the target detection model, the problem that the high-altitude target cannot be accurately and rapidly detected in a falling mode is solved, the acceleration characteristic data set capable of indicating the motion characteristic is obtained through processing of the acceleration value, the motion state is detected by combining the target detection model, whether the target to be detected falls is effectively detected, the detection result is accurate, the detection speed is high, and the real-time detection can be performed.
Description
Technical Field
The present invention relates to the field of information processing technologies, and in particular, to a method and apparatus for detecting a fall, an electronic device, and a storage medium.
Background
According to statistics, in the accident of three injuries (high falling, collapse and object striking) in the construction industry, the incidence rate of the high falling accident is highest and the danger is extremely high. Therefore, the method reduces and avoids the occurrence of falling accidents at high places, and is an effective way for guaranteeing safe production. However, when the iron tower is assembled and the ground lead frame is used for installing, the safety accident occurs when the high place falls due to the reasons of tension in construction period, scattered operation places, shortage of management staff, single monitoring means and the like, so that serious threat is brought to the life and property safety of people, and adverse effect is brought to the stable development of a harmonious society. Therefore, how to detect whether a target falls quickly and accurately becomes a problem to be solved.
Disclosure of Invention
The invention provides a falling detection method, a falling detection device, electronic equipment and a storage medium, which are used for solving the problem that high-altitude falling detection cannot be accurately and rapidly carried out on high-altitude operators.
According to an aspect of the present invention, there is provided a fall detection method including:
Acquiring acceleration data of an object to be detected, and determining acceleration values corresponding to different direction axes according to the acceleration data;
determining the amplitude change sum of the acceleration, the amplitude change rate of the acceleration, the total acceleration accumulation change rate and the total acceleration change displacement according to each acceleration value, and forming an acceleration characteristic data set;
and inputting the acceleration characteristic data set into a pre-trained target detection model, and determining whether the motion state of the target to be detected is falling or not according to the output result of the target detection model.
According to another aspect of the present invention, there is provided a fall detection device comprising:
the acceleration acquisition module is used for acquiring acceleration data of the target to be detected and determining acceleration values corresponding to different direction axes according to the acceleration data;
the characteristic data determining module is used for determining the amplitude change sum of the acceleration, the amplitude change rate of the acceleration, the total acceleration accumulation change rate and the total acceleration change displacement according to each acceleration value and forming an acceleration characteristic data set;
the detection module is used for inputting the acceleration characteristic data set into a pre-trained target detection model, and determining whether the motion state of the target to be detected is falling or not according to the output result of the target detection model.
According to another aspect of the present invention, there is provided an electronic apparatus including:
the at least one sensor is used for collecting acceleration data of an object to be detected;
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the fall detection method of any of the embodiments of the invention.
According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to execute the fall detection method according to any of the embodiments of the present invention.
According to the technical scheme, acceleration data of the object to be detected are obtained, and acceleration values corresponding to different direction axes are determined according to the acceleration data; determining the amplitude change sum of the acceleration, the amplitude change rate of the acceleration, the total acceleration accumulation change rate and the total acceleration change displacement according to each acceleration value, and forming an acceleration characteristic data set; the method comprises the steps of inputting an acceleration characteristic data set into a pre-trained target detection model, determining whether the motion state of a target to be detected falls according to the output result of the target detection model, solving the problem that the high-altitude target cannot be accurately and rapidly detected by falling, analyzing the acceleration data of the target to be detected, determining the acceleration value on each direction axis, calculating according to the acceleration value, obtaining the amplitude change and the amplitude change rate of the acceleration, the total acceleration accumulation change rate and the total acceleration change displacement, forming an acceleration characteristic data set, pre-training the target detection model, predicting the acceleration characteristic data set by the trained target detection model, judging whether the motion state of the target to be detected falls, obtaining the acceleration characteristic data set capable of indicating the motion characteristic by processing the acceleration value, detecting the motion state by combining the target detection model, effectively detecting whether the target to be detected occurs, and detecting the target to be detected is accurate in detection result and fast in falling.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for detecting a fall according to a first embodiment of the present invention;
FIG. 2 is a flow chart of a method for detecting a fall according to a second embodiment of the present invention;
FIG. 3 is a flow chart of the construction of an object detection model according to a second embodiment of the present invention;
fig. 4 is a schematic structural diagram of a transmitting end according to a second embodiment of the present invention;
fig. 5 is a schematic structural diagram of a receiving end according to a second embodiment of the present invention;
FIG. 6 is a schematic structural view of a fall detection device according to a third embodiment of the present invention;
Fig. 7 is a schematic structural diagram of an electronic device implementing a method for detecting a fall according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
Fig. 1 is a flowchart of a method for detecting a fall according to an embodiment of the present invention, where the method may be performed by a device for detecting a fall of a target, the device for detecting a fall may be implemented in hardware and/or software, and the device for detecting a fall may be configured in an electronic device. As shown in fig. 1, the method includes:
s101, acquiring acceleration data of an object to be detected, and determining acceleration values corresponding to different direction axes according to the acceleration data.
In this embodiment, the target to be detected may be specifically understood as a target with a requirement of falling detection, which may be an overhead operator, or an object in the high air, etc.; acceleration data may be specifically understood as data including information related to acceleration, and may be signal-type data; acceleration value is the specific value corresponding to acceleration.
The acceleration data can be acquired by a sensor, and the sensor can be arranged on the target to be detected to detect the acceleration data of the target to be detected in real time. The sensor can work according to a certain frequency, and acceleration data corresponding to a target to be detected are periodically detected; or triggering the sensor to detect acceleration data corresponding to the target to be detected after the triggering condition of the sensor is met; in order to ensure the detection of acceleration data of a target to be detected in real time and the accuracy of the detection, a periodic detection mode is preferably adopted.
Acceleration data of an object to be detected, which is acquired by a sensor, is acquired in real time, and when the object in space moves, the object generally moves in multiple directions, so that the acceleration data comprises related information of accelerations of different direction axes, such as direction axes in x, y and z directions. And analyzing and processing the acceleration data, removing the interferences such as impurities, noise and the like in the data, determining the related data of the acceleration corresponding to each direction axis, and extracting the data to obtain the acceleration value corresponding to each direction axis.
S102, determining the amplitude change sum of the acceleration, the amplitude change rate of the acceleration, the total acceleration accumulation change rate and the total acceleration change displacement according to each acceleration value, and forming an acceleration characteristic data set.
In this embodiment, the sum of the amplitude variation of the acceleration and the amplitude variation of the acceleration can be understood as a certain time; the amplitude change rate of the acceleration can be understood as the rate of change of the amplitude of the acceleration over a certain period of time; the total acceleration accumulation change rate can be specifically understood as the change rate of the total acceleration in a certain time; the resultant acceleration change displacement is understood to mean in particular a displacement which moves under the action of the resultant acceleration. And the sum of the amplitude changes of the accelerations, the amplitude change rate of the accelerations, the sum acceleration accumulation change rate and the sum acceleration change displacement are taken as characteristic data of the accelerations, so that the motion state is represented and described.
And according to the acceleration value on each direction axis, calculating according to a predefined calculation method of different characteristic values of the acceleration to respectively obtain the amplitude change sum of the acceleration, the amplitude change rate of the acceleration, the total acceleration accumulation change rate and the total acceleration change displacement. For example, calculating the amplitude variation of each acceleration in a period of time or at a time point, and carrying out comprehensive operations such as summation, weighted summation and the like on the amplitude variation to obtain the amplitude variation sum of the acceleration; correspondingly, the amplitude change rate of the acceleration is divided by a time interval or unit time of a period of time on the basis of the sum of the amplitude changes of the acceleration, so that the amplitude change rate is obtained. The cumulative rate of change of the combined acceleration may be calculated by calculating the combined acceleration and dividing by the time interval. The combined acceleration change displacement may be calculated by calculating the displacement of each direction axis, and then performing comprehensive operations such as summation, weighted summation, etc. on the displacement of each direction axis to obtain the combined acceleration change displacement, or calculating the combined acceleration according to each acceleration value, and calculating the displacement according to the combined acceleration to obtain the combined acceleration change displacement. And placing the obtained amplitude change sum of the acceleration, the amplitude change rate of the acceleration, the total acceleration accumulation change rate and the total acceleration change displacement into a data set to form an acceleration characteristic data set.
S103, inputting the acceleration characteristic data set into a pre-trained target detection model, and determining whether the motion state of the target to be detected is falling according to the output result of the target detection model.
In this embodiment, the target detection model may be specifically understood as a pre-trained neural network model for detecting a motion state of a target, and the type of the target detection model is not limited in this embodiment of the present application, for example, a support vector machine SVM, a CART decision tree, and the like. The present application detects a fall of a target, i.e., what motion the target is currently performing, e.g., falling, resting, rising, falling, etc., and thus the motion state may include only two, i.e., the motion state includes both falling and non-falling.
The method comprises the steps of pre-collecting data, processing the data to form a training sample, dividing the training sample into a training set and a testing set, training a model through the training set, testing the model through the testing set, and obtaining a model with accuracy meeting certain requirements, wherein the model can be used as a target detection model for falling detection. The method comprises the steps of inputting the amplitude change sum of acceleration in an acceleration characteristic data set, the amplitude change rate of the acceleration, the total acceleration accumulation change rate and the total acceleration change displacement into a trained target detection model, analyzing and processing input data by the target detection model according to experience learned in a training process and model parameters, outputting a predicted result, and determining whether the motion state of a target to be detected is falling according to an output result. For example, the output is a numerical value, 1 representing a fall; and 0 represents non-falling, and whether the object to be detected is in a falling state can be determined through the output value.
According to the method and the device, the acceleration data of the target to be detected are analyzed, so that the motion state of the target to be detected is detected, and when the motion state of the target to be detected is detected to fall, corresponding measures can be started to rescue the target to be detected.
The falling detection method solves the problem that a high-altitude target cannot be accurately and rapidly detected, the acceleration data of the target to be detected are analyzed, the acceleration value on each direction axis is determined, the acceleration value is calculated according to the acceleration value, the amplitude change and the amplitude change rate of acceleration, the total acceleration accumulation change rate and the total acceleration change displacement of the acceleration are obtained, the acceleration change trend of the target in the falling process at the high position is accurately reflected, an acceleration characteristic data set is formed, a target detection model is trained in advance, the trained target detection model predicts the acceleration characteristic data set, further whether the motion state of the target to be detected falls is judged, the acceleration characteristic data set capable of indicating the motion characteristic is obtained through the processing of the acceleration value, the detection of the motion state is further carried out by combining the target detection model, whether the target to be detected falls or not is effectively detected, the detection result is accurate, the detection speed is high, and the real-time detection can be carried out.
Example two
Fig. 2 is a flowchart of a method for detecting a fall according to a second embodiment of the present invention, where the method is refined based on the foregoing embodiment. As shown in fig. 2, the method includes:
s201, acquiring acceleration data of an object to be detected.
S202, determining acceleration signals corresponding to different direction axes according to the acceleration data.
The method comprises the steps that a sensor is arranged on an object to be detected to collect acceleration data, the acceleration data comprise accelerations of axes in different directions, and the three-axis acceleration is taken as an example in the detection process. And processing the acceleration data, separating acceleration signals of different direction axes, and obtaining acceleration signals corresponding to each direction axis.
S203, performing Hilbert transform on the acceleration signals corresponding to each direction axis, determining the envelope value of the acceleration signals, and determining the envelope value of the acceleration signals as the acceleration value.
Because noise and the like exist in the acceleration signals to influence the accuracy of data, hilbert transformation is carried out on the acceleration signals corresponding to each direction axis, the envelope value of the acceleration signals is obtained through calculation, and the envelope value of the acceleration signals is determined to be the acceleration value.
Exemplary, an embodiment of the present application provides a method for calculating an envelope value, as follows:
hilbert transform of signal x (t)The calculation flow of (1) is as follows:
wherein, the convolution symbol is represented by the signal x (t), and the analysis signal w (t) is:
from this, the envelope value a (t) of the signal x (t) can be obtained as:
according to the embodiment of the application, the Hilbert transform is carried out on the acceleration data, so that the filtering and noise reduction of the data are realized. The high frequency component in the signal is highlighted, so that the fast changing part in the signal is extracted, and the accuracy of the characteristic data is improved.
S204, for each acceleration value, performing integral operation according to the absolute value of the acceleration value and a preset acquisition period to obtain the amplitude change of the acceleration value.
In this embodiment, the acquisition period is a period in which the sensor acquires acceleration data, and the sensor periodically acquires acceleration data according to a time interval of the acquisition period, so as to realize real-time detection of the target to be detected. The acquisition period may be set according to the frequency of activity of the target. Since the acceleration has a direction, the direction thereof is related to the setting of the coordinate axis, but does not affect the movement speed, the displacement magnitude, and the like. For each acceleration value, determining the absolute value of the acceleration value, determining the upper and lower limits of integration according to the acquisition period, integrating the absolute value of the acceleration value over time, and determining the integration result as the amplitude change of the acceleration value.
S205, determining the sum of the amplitude changes of the acceleration values as the sum of the amplitude changes of the acceleration values.
The amplitude variation of each acceleration value is summed, and the sum obtained is determined as the sum of the amplitude variations of the acceleration.
In an exemplary embodiment, a calculation formula of a sum of amplitude variations of acceleration is provided, and all the related calculation formulas provided in the embodiment of the present application take three-axis acceleration values as examples.
Wherein RS is the sum of the amplitude changes of acceleration, T is the acquisition period, a x For acceleration values in the x-axis direction, a y For acceleration values in the y-axis direction, a z Is the acceleration value in the z-axis direction.
According to the method and the device, the sum of the amplitude changes of the acceleration of each direction axis is calculated, the amplitude changes of the acceleration are obtained and are used as characteristic data of the acceleration, and compared with the mode that the vector synthesis or the distance calculation is carried out in the prior art, the amplitude changes of the acceleration are directly added, so that the difference between the acceleration at the moment and the acceleration of normal motion can be effectively amplified.
In an exemplary embodiment, the present application provides a method for setting an acquisition period, where, during acceleration data acquisition, according to nyquist theorem, in order to be able to recover information contained in an acquired signal without distortion, the frequency during sampling must be twice the highest frequency of the acquired signal. The frequency of most human body activities is less than or equal to 20Hz, and the frequency of only a few human body activities exceeds 15Hz. When the human body is in a relatively static state, the frequency of the limb is further reduced to less than or equal to 10Hz, and only 2% of daily activities are more than 10Hz. When the object to be detected works at a high place, the MPU6050 sensor can be adopted to acquire triaxial acceleration data in consideration of a narrow moving space. Through the acceleration data of the three directions, the acceleration change condition of the limb in different directions can be observed. The sample rate of the sensor was 50Hz, i.e. the acceleration values of the x-axis, y-axis and z-axis were recorded every 0.02 seconds. Drawing a time-dependent acceleration change chart, wherein the horizontal axis is time, the vertical axis is an acceleration value image, and primarily judging the motion state of a human body in different time periods by observing the fluctuation condition of the image. For example, in normal walking or movement, a more regular waveform can be observed, and acceleration fluctuates over a range; while in running or other strenuous exercise the amplitude and frequency of acceleration fluctuations may be greater. While the acceleration fluctuations are relatively small and remain substantially at a low level when the body is at rest.
S206, dividing the sum of the amplitude changes of the acceleration by the acquisition period to obtain the amplitude change rate of the acceleration.
An exemplary embodiment of the present application provides a calculation formula of an amplitude change rate of acceleration, taking a triaxial acceleration value as an example:
wherein RST is the rate of change of the amplitude of the acceleration.
S207, calculating the square sum of each acceleration value, and squaring the square sum to obtain a square root.
And S208, performing integral operation according to the square root and a preset acquisition period, and dividing an integral result by the acquisition period to obtain the total acceleration accumulation change rate.
The square of the acceleration value is calculated, then the sum of squares is obtained by summation, the square sum is squared to obtain the square root, and the square root obtained at this time is the magnitude of the combined acceleration. And determining an upper limit and a lower limit of integration according to the acquisition period, integrating square root with time, and determining an integration result as an accumulated change rate of the combined acceleration value.
Exemplary, the embodiment of the application provides a calculation formula of a cumulative change rate of the total acceleration:
wherein ACC is the cumulative rate of change of the combined acceleration.
S209, for each acceleration value, performing integral operation according to the absolute value of the acceleration value and a preset acquisition period to obtain a speed value.
And calculating a corresponding speed value according to the acceleration value of each axial direction. For each acceleration value, the upper and lower limits of integration are determined according to the acquisition period, and the integration operation is carried out in time according to the absolute value of the acceleration value, so that the speed value is obtained.
Exemplary, the embodiment of the present application provides a calculation formula of a speed value:
wherein V is x Is a velocity value in the x-axis direction; v (V) y Is a velocity value in the y-axis direction; v (V) z Is the velocity value in the z-axis direction.
S210, performing integral operation according to the absolute value of the speed value and the acquisition period to obtain the change displacement.
And determining the upper limit and the lower limit of integration according to the acquisition period, and performing integration operation in time according to the absolute value of the speed value to obtain the change displacement.
S211, determining the sum of the various displacement changes as the combined acceleration displacement change.
And summing the various displacement changes to obtain a value which is the combined acceleration displacement.
Exemplary, the embodiment of the application provides a calculation formula of a variable displacement:
wherein S is the combined acceleration change displacement.
According to the method and the device, the sum of the displacements of the direction axes is calculated, the change displacement of the combined acceleration is obtained and is used as the characteristic data of the acceleration, and compared with the mode of vector synthesis or distance calculation in the prior art, the displacement of the direction axes is directly added, so that the difference between the acceleration at the moment and the acceleration of normal motion can be effectively enlarged.
It is to be understood that the determination of the amplitude change of the acceleration and the sum acceleration accumulation change rate and the sum acceleration change displacement is not strictly sequential in implementation, and the determination of the amplitude change rate of the acceleration is performed after the determination of the amplitude change of the acceleration. In the embodiment of the application, displacement is performed by the sum of the amplitude variation of the acceleration, the accumulated change rate of the combined acceleration and the combined acceleration variation in parallel.
S212, forming an acceleration characteristic data set.
S213, inputting the acceleration characteristic data set into a pre-trained target detection model, and determining whether the motion state of the target to be detected is falling according to the output result of the target detection model.
As an optional embodiment of the present embodiment, the determining step of the target detection model is further optimized according to the present optional embodiment, including:
a1, acquiring a training sample set comprising at least one training sample, wherein the training sample comprises a training characteristic data set and a label value.
In this embodiment, a training sample set may be specifically understood as a set of training samples, and in order to ensure model accuracy, the training sample set generally includes a large number of training samples. The training feature data set may be specifically understood as a set formed by acceleration features for training the model, where each training feature data set includes an amplitude variation sum of acceleration, an amplitude variation rate of acceleration, a total acceleration accumulation variation rate, and a total acceleration variation displacement, and generally, acceleration data of the target in different motion states are collected, and the acceleration data obtained at this time are acceleration data at different moments, and the acceleration data are processed by using the above manner to obtain a corresponding amplitude variation sum of acceleration, an amplitude variation rate of acceleration, a total acceleration accumulation variation rate, and a total acceleration variation displacement. The tag value is used to indicate the motion state corresponding to the training feature data set, e.g., 1 indicates a fall and 0 indicates a non-fall.
A2, training the initial detection model according to each training sample to obtain the target detection model with the recognition accuracy meeting the accuracy requirement.
The initial detection model is a support vector machine model, and a kernel function of the support vector machine model is a Gaussian kernel function.
In this embodiment, the initial detection model may be specifically understood as an initial model constructed during training; the accuracy requirement may be that the accuracy exceeds a certain threshold, for example, the accuracy exceeds 95%.
And constructing an initial detection model, wherein the initial detection model is a support vector machine model, and the kernel function of the initial detection model is a Gaussian kernel function. During training, the training sample can be divided into a training set and a testing set, the training set is used for training the initial detection model, the training feature data set is input into the initial detection model for prediction, a prediction result is obtained, a loss function is calculated according to the prediction result and the label value, the initial detection model is subjected to back propagation according to the loss function, model parameters are adjusted, and a new training feature data set is input for repeated prediction until a model meeting convergence conditions is obtained. And testing the obtained model through a test set, determining the prediction result of each group of training characteristic data set, comparing the prediction result with a label value, if the prediction result is consistent, determining that the prediction result is correct, otherwise, the prediction result is incorrect, and counting the correct proportion of the prediction result to obtain the identification precision. Judging whether the recognition precision of the model meets the precision requirement, if so, determining the model obtained at the moment as a target detection model, otherwise, continuing training the model until the target detection model meeting the precision requirement is obtained.
The Support Vector Machine (SVM) is used as a supervised learning algorithm to be applied to the construction of the intelligent detection model of the high falling. SVM is a classification machine learning method that classifies data by constructing a hyperplane. The feature data is classified using the SVM to determine if a high fall event has occurred. First, the acquired feature data needs to be divided into two parts, i.e., training data and test data. Training data is used to construct the classifier, while test data is used to evaluate the performance of the classifier. By separating the data into training data and test data, the accuracy and reliability of the results of evaluating the classifier are ensured. A gaussian kernel function is chosen as the kernel function of the SVM classifier. A gaussian kernel function is a commonly used kernel function that has a good nonlinear characteristic. By using the gaussian kernel function, complex relationships between feature data can be better processed, thereby improving the accuracy and generalization ability of the classifier. The classifier is trained using the training data. In the training process, the classifier learns the mode and rule of the characteristic data and forms a classification model, namely a target detection model, which is used for predicting the motion state of a certain target at a certain moment. Finally, the test data is used to evaluate the performance of the classifier. And calculating the accuracy of the classifier by comparing the difference between the predicted result and the actual result of the classifier on the test data.
Exemplary, the embodiment of the application provides a manner of determining a target detection model, which is specifically implemented as follows:
the training samples are divided into training and testing sets. The division can be completed by using the train_test_split function in the scikit-learn library, and a certain division proportion can be set according to practical situations, precision requirements and the like during the division, for example, data with the division proportion of 80% is used for training, and 20% is used for testing. Each feature data in the training feature data set can be represented in a matrix form, namely, a feature matrix X of one row and 4 columns, and the tag value is used as a target variable y. the train_test_split function divides the feature matrix X and the target variable y into four parts, x_train, x_test, y_train, and y_test, where x_train and y_train are the feature matrix and the target variable of the training set and x_test and y_test are the feature matrix and the target variable of the test set. Test_size=0.2 was used to specify a test set to total data set ratio of 20% and random_state=42 was used to set random seeds to ensure the same partitioning result was obtained each time the code was run.
An SVM model (i.e., an initial detection model) is created and a gaussian kernel function (Radialbasis function, RBF) is specified as the kernel function. In scikit-learn, the class of SVM model is SVC, and RBF kernel functions can be selectively used by setting kernel= 'RBF'. svm=svc (kernel= 'rbf'). The RBF kernel function can map the input space to a high-dimensional feature space, thereby enabling the SVM to handle nonlinear problems. It maps low-dimensional data into high-dimensional space by introducing nonlinear transformation, making data easier to separate in high-dimensional space. The RBF kernel function implements global optimization in the SVM because it takes into account the entire training dataset. The optimal decision boundary is determined by maximizing the classification interval, so that the generalization performance of the model is improved. The SVM model is then trained using the feature matrix x_train and the target variable y_train of the training set. Fjn (x_train, y_train) by calling the model's fit function, the model will train by maximizing the classification interval. After training is completed, the trained model may be used for prediction. The feature matrix x_test of the test set is taken as a parameter to be transmitted into a prediction function, and a prediction result of test data can be obtained by y_pred=svm. Finally, the accuracy of the prediction is calculated by using the accuracy_score function by taking the target variable y_test and the prediction result y_pred of the test set as parameters, and the prediction result is obtained by using the accuracy=accuracy_score (y_test, y_pred).
And obtaining a final algorithm program through experimental model parameter adjustment and actual data training. The accuracy of the model is up to 98%, wherein the false alarm rate is reduced to 0.3%, the false alarm rate is 2.2%, the prediction speed is lower than 1ms, and the method has the characteristics of high reliability and high speed.
By way of example, fig. 3 provides a flow chart for constructing the target detection model, and the flow chart for constructing the target detection model is described.
S1, collecting acceleration data.
The step can collect acceleration data under different motion states, and each acceleration speed is acceleration data at different moments.
And S2, extracting characteristics of the acceleration data, and forming a training sample.
The method comprises the steps of processing acceleration data and extracting features of the acceleration data respectively to obtain amplitude change and amplitude change rate of corresponding acceleration, total acceleration accumulation change rate and total acceleration change displacement of the corresponding acceleration, and forming a training feature data set. And marking each group of training characteristic data sets, determining the label value of each group of training characteristic data sets, and further forming training samples.
S3, constructing an initial detection model.
S4, selecting a kernel function of the initial detection model.
S5, training a model.
This step may train the initial detection model. And when the model precision does not meet the requirement, continuing to train the model.
S6, testing the model obtained after training.
S7, judging whether the identification precision of the obtained model meets the precision requirement, and if not, executing S5; if yes, S8 is executed.
S8, determining the target detection model.
As an optional embodiment of the present embodiment, the further optimization of the present optional embodiment includes generating alarm prompt information to perform alarm prompt when the motion state of the object to be detected is a fall.
In this embodiment, the alarm prompting information may be specifically understood as information for prompting the falling event, and may be an audio prompt, a light prompt, or a mixed use of multiple modes for prompting when the alarm is performed, for example, the alarm is performed by adding the voice of "noticing the falling event occurrence" to the alarm sound. The alarm prompt information can be used for indicating related alarm devices to alarm, such as an alarm for carrying out sound prompt, a sound box and the like, and a lamp for carrying out light prompt and the like.
The alarm prompting mode is preset and can be selected by a user independently. And determining alarm prompt information according to a preset alarm prompt mode when the motion state of the object to be detected is falling, determining a related alarm device according to the alarm prompt information, and controlling the alarm device to carry out alarm prompt.
As an optional embodiment of the present embodiment, the further optimization includes controlling activation of the guard according to the position of the object to be detected when the motion state of the object to be detected is a fall.
In this embodiment, the protection device is specifically understood as a facility for protecting a person or an object falling from a high altitude, for example, an air bed. The target to be detected can also be worn with the positioning device for positioning, or the position information of the target to be detected can be preset when the position of the target does not change greatly. When the motion state of the target to be detected is falling, the position of the target to be detected is determined according to the positioning device, or the position of the target to be detected is determined according to preset position information, the corresponding protection device is determined according to the position of the target to be detected, and the protection device is started to carry out safety protection on the target to be detected, so that the safety of personnel is guaranteed.
In this embodiment of the present application, detection may be further configured, alarm and protection may not be performed at one end, for example, a sending end and a receiving end are set, where the sending end is configured to perform fall detection, and send a detection result to the receiving end, where the receiving end parses the information after receiving the information, and performs alarm prompt and safety protection by adopting corresponding means when falling occurs.
Exemplary, fig. 4 provides a schematic structural diagram of a transmitting end, where the transmitting end includes: the system comprises an acceleration sensor module 11, an overhead falling intelligent detection module 12, a signal transmission module 13 and a power management module 14. The acceleration sensor module 11 is used for acquiring acceleration data; the high falling intelligent detection module 12 is used for judging whether the motion state of the target to be detected is falling according to the acceleration data; the signal sending module 13 is configured to send the detected result to the receiving end. The power management module 14 is used for managing power supply to the acceleration sensor module 11, the intelligent high-altitude falling detection module 12 and the signal transmission module 13. Fig. 5 provides a schematic structural diagram of a receiving end, where the receiving end includes: an information receiving module 21, a signal processing module 22, a fall processing module 23 and a power management module 24. The information receiving module 21 is configured to receive a detection result sent by the sending end; the signal processing module 22 is used for processing the received information; the fall processing module 23 is used to alert and activate the protective device in the event of a fall. The power management module 24 manages power supply to the information receiving module 21, the signal processing module 22, and the fall processing module 23.
According to the falling detection method provided by the embodiment of the invention, the collected triaxial acceleration sensor data is subjected to Hilbert transformation, so that the functions of filtering and noise reduction are realized, interference caused by the sensor itself, the environment and the like can be eliminated, and the accuracy and the reliability of the data are improved. By collecting and processing the acceleration data, the acceleration change condition of the human body in the falling process at a high place can be more accurately analyzed. The SVM-based machine learning algorithm has higher classification precision when processing nonlinear problems, and can extract more complex characteristic information, thereby accurately identifying falling events. By training and predicting the characteristic data of the acceleration, the high falling event is detected in time, and corresponding emergency measures are taken, so that the occurrence of the high falling event is effectively prevented and reduced, the personal safety is ensured, and the accident risk is reduced.
Example III
Fig. 6 is a schematic structural diagram of a falling detection device according to a third embodiment of the present invention. As shown in fig. 6, the apparatus includes: an acceleration acquisition module 31, a characteristic data determination module 32, and a detection module 33.
The acceleration acquisition module 31 is configured to acquire acceleration data of an object to be detected, and determine acceleration values corresponding to different direction axes according to the acceleration data;
A characteristic data determining module 32, configured to determine, according to each of the acceleration values, an amplitude change sum of acceleration, an amplitude change rate of acceleration, a total acceleration accumulation change rate, and a total acceleration change displacement, and form an acceleration characteristic data set;
the detection module 33 is configured to input the acceleration feature data set into a pre-trained target detection model, and determine whether the motion state of the target to be detected is falling according to the output result of the target detection model.
The falling detection device solves the problem that a high-altitude target cannot be accurately and rapidly detected, the acceleration value on each direction axis is determined by analyzing the acceleration data of the target to be detected, the acceleration value is calculated according to the acceleration value, the amplitude change and the amplitude change rate of acceleration, the total acceleration accumulation change rate and the total acceleration change displacement of the acceleration are obtained, the acceleration change trend of the target in the falling process at a high position is accurately reflected, an acceleration characteristic data set is formed, a target detection model is trained in advance, the trained target detection model predicts the acceleration characteristic data set, further whether the motion state of the target to be detected falls is judged, the acceleration characteristic data set capable of indicating the motion characteristic is obtained by processing the acceleration value, the detection of the motion state is further carried out by combining the target detection model, whether the target to be detected falls is effectively detected, the detection result is accurate, the detection speed is high, and the real-time detection can be carried out.
Optionally, the acceleration acquisition module 31 includes:
the signal determining unit is used for determining acceleration signals corresponding to different direction axes according to the acceleration data;
and the signal processing unit is used for carrying out Hilbert transformation on the acceleration signals corresponding to each direction axis, determining the envelope value of the acceleration signals and determining the envelope value of the acceleration signals as the acceleration value.
Optionally, the feature data determining module 32 includes:
the amplitude change determining unit is used for carrying out integral operation according to the absolute value of each acceleration value and a preset acquisition period to obtain the amplitude change of the acceleration value;
and the amplitude change and determination unit is used for determining the sum of the amplitude changes of the acceleration values as the sum of the amplitude changes of the acceleration.
Optionally, the feature data determining module 32 includes:
and the amplitude change rate determining unit is used for dividing the amplitude change sum of the acceleration by the acquisition period to obtain the amplitude change rate of the acceleration.
Optionally, the feature data determining module 32 includes:
the square operation unit is used for calculating the square sum of each acceleration value and squaring the square sum to obtain square root;
And the accumulated change rate determining unit is used for carrying out integral operation according to the square root and a preset acquisition period, and dividing an integral result by the acquisition period to obtain the accumulated change rate of the combined acceleration.
Optionally, the feature data determining module 32 includes:
the speed value determining unit is used for carrying out integral operation according to the absolute value of each acceleration value and a preset acquisition period to obtain a speed value;
the change displacement determining unit is used for carrying out integral operation according to the absolute value of the speed value and the acquisition period to obtain change displacement;
and a total variation displacement determining unit for determining the sum of the variation displacements as a total acceleration variation displacement.
Optionally, the apparatus further comprises:
the sample set acquisition module is used for acquiring a training sample set comprising at least one training sample, wherein the training sample comprises a training characteristic data set and a label value;
the model training module is used for training the initial detection model according to each training sample to obtain a target detection model with the recognition precision meeting the precision requirement;
the initial detection model is a support vector machine model, and a kernel function of the support vector machine model is a Gaussian kernel function.
Optionally, the apparatus further comprises:
and the alarm module is used for generating alarm prompt information to carry out alarm prompt when the motion state of the target to be detected is falling.
Optionally, the apparatus further comprises:
and the protection module is used for controlling the protection device to start according to the position of the target to be detected when the motion state of the target to be detected is falling.
The falling detection device provided by the embodiment of the invention can execute the falling detection method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example IV
Fig. 7 shows a schematic diagram of an electronic device that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 7, the electronic device includes at least one sensor 40, at least one processor 41, and a memory communicatively coupled to the at least one processor 41, such as a Read Only Memory (ROM) 42, a Random Access Memory (RAM) 43, etc., one sensor 40 and one processor 41 being illustrated. In which a memory stores a computer program executable by at least one processor, the processor 41 may perform various suitable actions and processes according to the computer program stored in a Read Only Memory (ROM) 42 or the computer program loaded from a storage unit 48 into a Random Access Memory (RAM) 43. In the RAM 43, various programs and data required for the operation of the electronic device 40 may also be stored. The processor 41, the ROM 42 and the RAM 43 are connected to each other via a bus 44. An input/output (I/O) interface 45 is also connected to bus 44.
Various components in electronic device 40 are connected to I/O interface 45, including: an input unit 46 such as a keyboard, a mouse, etc.; an output unit 47 such as various types of displays, speakers, and the like; a storage unit 48 such as a magnetic disk, an optical disk, or the like; and a communication unit 49 such as a network card, modem, wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The processor 41 may be various general and/or special purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 41 performs the various methods and processes described above, such as fall detection methods.
In some embodiments, the fall detection method may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as the storage unit 48. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 40 via the ROM 42 and/or the communication unit 49. When a computer program is loaded into RAM 43 and executed by processor 41, one or more steps of the fall detection method described above may be performed. Alternatively, in other embodiments, processor 41 may be configured to perform the fall detection method in any other suitable manner (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. A method of fall detection comprising:
acquiring acceleration data of an object to be detected, and determining acceleration values corresponding to different direction axes according to the acceleration data;
determining the amplitude change sum of the acceleration, the amplitude change rate of the acceleration, the total acceleration accumulation change rate and the total acceleration change displacement according to each acceleration value, and forming an acceleration characteristic data set;
and inputting the acceleration characteristic data set into a pre-trained target detection model, and determining whether the motion state of the target to be detected is falling or not according to the output result of the target detection model.
2. The method according to claim 1, wherein determining acceleration values corresponding to different directional axes from the acceleration data comprises:
determining acceleration signals corresponding to different direction axes according to the acceleration data;
and performing Hilbert transform on the acceleration signals corresponding to each direction axis, determining the envelope value of the acceleration signals, and determining the envelope value of the acceleration signals as the acceleration value.
3. The method of claim 1, wherein the step of determining the position of the substrate comprises,
the determining the amplitude variation sum of the acceleration according to each acceleration value comprises the following steps:
for each acceleration value, carrying out integral operation according to the absolute value of the acceleration value and a preset acquisition period to obtain the amplitude change of the acceleration value;
determining the sum of the amplitude changes of the acceleration values as the sum of the amplitude changes of the acceleration values;
determining the amplitude change rate of the acceleration according to each acceleration value, including:
dividing the amplitude change sum of the acceleration by the acquisition period to obtain the amplitude change rate of the acceleration.
4. The method of claim 1, wherein determining a cumulative rate of change of the combined acceleration from each of the acceleration values comprises:
Calculating the square sum of each acceleration value, and squaring the square sum to obtain a square root;
and carrying out integral operation according to the square root and a preset acquisition period, and dividing an integral result by the acquisition period to obtain the cumulative change rate of the combined acceleration.
5. The method of claim 1, wherein determining a resultant acceleration-varying displacement from each of the acceleration values comprises:
for each acceleration value, carrying out integral operation according to the absolute value of the acceleration value and a preset acquisition period to obtain a speed value;
performing integral operation according to the absolute value of the speed value and the acquisition period to obtain a change displacement;
and determining the sum of the variable displacements as the combined acceleration variable displacement.
6. The method of claim 1, wherein the step of determining the object detection model comprises:
acquiring a training sample set comprising at least one training sample, wherein the training sample comprises a training characteristic data set and a label value;
training the initial detection model according to each training sample to obtain a target detection model with identification precision meeting the precision requirement;
the initial detection model is a support vector machine model, and a kernel function of the support vector machine model is a Gaussian kernel function.
7. The method of any one of claims 1-6, further comprising:
when the motion state of the target to be detected is falling, generating alarm prompt information to carry out alarm prompt;
the method further comprises the steps of:
and when the motion state of the target to be detected is falling, controlling the starting of the protective device according to the position of the target to be detected.
8. A fall detection device, comprising:
the acceleration acquisition module is used for acquiring acceleration data of the target to be detected and determining acceleration values corresponding to different direction axes according to the acceleration data;
the characteristic data determining module is used for determining the amplitude change sum of the acceleration, the amplitude change rate of the acceleration, the total acceleration accumulation change rate and the total acceleration change displacement according to each acceleration value and forming an acceleration characteristic data set;
the detection module is used for inputting the acceleration characteristic data set into a pre-trained target detection model, and determining whether the motion state of the target to be detected is falling or not according to the output result of the target detection model.
9. An electronic device, the electronic device comprising:
the at least one sensor is used for collecting acceleration data of an object to be detected;
At least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the fall detection method of any of claims 1-7.
10. A computer readable storage medium storing computer instructions for causing a processor to perform the fall detection method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311741975.5A CN117556333A (en) | 2023-12-18 | 2023-12-18 | Falling detection method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311741975.5A CN117556333A (en) | 2023-12-18 | 2023-12-18 | Falling detection method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117556333A true CN117556333A (en) | 2024-02-13 |
Family
ID=89814702
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311741975.5A Pending CN117556333A (en) | 2023-12-18 | 2023-12-18 | Falling detection method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117556333A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108009572A (en) * | 2017-11-22 | 2018-05-08 | 中国地质大学（武汉） | Mobile device fall detection method and its model forming method and mobile equipment |
| CN108259670A (en) * | 2018-01-22 | 2018-07-06 | 广东欧珀移动通信有限公司 | Electronic device falls processing method and Related product |
| CN108447225A (en) * | 2018-02-07 | 2018-08-24 | 广东中科慈航信息科技有限公司 | A kind of tumble detection method for human body and device |
| CN111466918A (en) * | 2020-04-16 | 2020-07-31 | 汤佳梅 | Abnormal behavior detection method and device, computer equipment and storage medium |
| CN112346050A (en) * | 2020-10-23 | 2021-02-09 | 清华大学 | Fall detection method and system based on Wi-Fi equipment |
| WO2021074295A1 (en) * | 2019-10-18 | 2021-04-22 | Dietmar Basta | Individualized fall prevention |
| US11170295B1 (en) * | 2016-09-19 | 2021-11-09 | Tidyware, LLC | Systems and methods for training a personalized machine learning model for fall detection |
| CN116259156A (en) * | 2021-12-09 | 2023-06-13 | 深圳迈瑞生物医疗电子股份有限公司 | Fall detection method and fall detection system |
| US20230215261A1 (en) * | 2020-07-06 | 2023-07-06 | Epic Safety Inc. | Method and apparatus for detecting fall events |
-
2023
- 2023-12-18 CN CN202311741975.5A patent/CN117556333A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11170295B1 (en) * | 2016-09-19 | 2021-11-09 | Tidyware, LLC | Systems and methods for training a personalized machine learning model for fall detection |
| CN108009572A (en) * | 2017-11-22 | 2018-05-08 | 中国地质大学（武汉） | Mobile device fall detection method and its model forming method and mobile equipment |
| CN108259670A (en) * | 2018-01-22 | 2018-07-06 | 广东欧珀移动通信有限公司 | Electronic device falls processing method and Related product |
| CN108447225A (en) * | 2018-02-07 | 2018-08-24 | 广东中科慈航信息科技有限公司 | A kind of tumble detection method for human body and device |
| WO2021074295A1 (en) * | 2019-10-18 | 2021-04-22 | Dietmar Basta | Individualized fall prevention |
| CN111466918A (en) * | 2020-04-16 | 2020-07-31 | 汤佳梅 | Abnormal behavior detection method and device, computer equipment and storage medium |
| US20230215261A1 (en) * | 2020-07-06 | 2023-07-06 | Epic Safety Inc. | Method and apparatus for detecting fall events |
| CN112346050A (en) * | 2020-10-23 | 2021-02-09 | 清华大学 | Fall detection method and system based on Wi-Fi equipment |
| CN116259156A (en) * | 2021-12-09 | 2023-06-13 | 深圳迈瑞生物医疗电子股份有限公司 | Fall detection method and fall detection system |
Non-Patent Citations (2)
| Title |
|---|
| DHINESH R. ET AL.: "Fall Detection Using Kinematic Features from a Wrist-Worn Inertial Sensor", 《IEEE INSTRUMENTATION AND MEASUREMENT SOCIETY》, 15 August 2019 (2019-08-15) * |
| 刘德阳 等: "自动扶梯乘客摔倒智能识别方法研究", 《中国特种设备安全》, 30 June 2023 (2023-06-30) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10204501B2 (en) | Providing predictive alerts for workplace safety | |
| EP3289573B1 (en) | System for integrating multiple sensor data to predict a fall risk | |
| US10228994B2 (en) | Information processing system, information processing method, and program | |
| US20220163348A1 (en) | Positioning method and electronic device | |
| US11875246B2 (en) | Intelligent FODAS system and method based on AI chip | |
| CN108109336A (en) | A kind of human body tumble recognition methods based on acceleration transducer | |
| CN104297519A (en) | Human motion attitude identification method and mobile terminal | |
| CN115565101A (en) | Production safety abnormity identification method and device, electronic equipment and storage medium | |
| Aiello et al. | Machine Learning approach towards real time assessment of hand-arm vibration risk | |
| CN114357567A (en) | BIM-based wind vibration monitoring system, BIM-based wind vibration monitoring storage medium and computer equipment | |
| CN117556333A (en) | Falling detection method and device, electronic equipment and storage medium | |
| CN109620241B (en) | Wearable device and motion monitoring method based on same | |
| CN116842366A (en) | Abnormality warning method, device, equipment and medium for hydroelectric generating set | |
| Wu et al. | Fall detection monitoring system based on mems sensor | |
| CN112386249B (en) | Fall detection method and device, equipment and storage medium | |
| CN114120252B (en) | Automatic driving vehicle state identification method and device, electronic equipment and vehicle | |
| CN114743332A (en) | Perception early warning method and device for intelligent fire fighting, storage medium and terminal | |
| Luca et al. | Anomaly detection using the Poisson process limit for extremes | |
| CN112450901A (en) | Intelligent live-wire work wearing equipment and analysis method for safety of working personnel | |
| Hsieh et al. | An intelligent fall detection design for mobile health-care applications | |
| CN110852260B (en) | Drowning behavior recognition method based on accelerometer | |
| CN116363751A (en) | Climbing action recognition method, device and equipment for electric power tower climbing operation and storage medium | |
| CN115127514B (en) | Height measurement method, device, electronic equipment and storage medium | |
| CN116738296B (en) | Comprehensive intelligent monitoring system for machine room conditions | |
| AU2021105041A4 (en) | Method and equipment for judging safety of operating personnel |
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 |