CN112180359B - FMCW-based human body tumbling detection method - Google Patents
FMCW-based human body tumbling detection method Download PDFInfo
- Publication number
- CN112180359B CN112180359B CN202011210306.1A CN202011210306A CN112180359B CN 112180359 B CN112180359 B CN 112180359B CN 202011210306 A CN202011210306 A CN 202011210306A CN 112180359 B CN112180359 B CN 112180359B
- Authority
- CN
- China
- Prior art keywords
- information
- human body
- centroid
- centroid information
- fmcw
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 18
- 230000006399 behavior Effects 0.000 claims abstract description 51
- 238000000034 method Methods 0.000 claims abstract description 27
- 238000013528 artificial neural network Methods 0.000 claims abstract description 19
- 206010000117 Abnormal behaviour Diseases 0.000 claims abstract description 14
- 238000012549 training Methods 0.000 claims abstract description 14
- 239000013598 vector Substances 0.000 claims abstract description 13
- 230000008447 perception Effects 0.000 claims abstract description 5
- 230000009471 action Effects 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 230000009466 transformation Effects 0.000 claims description 2
- 238000000605 extraction Methods 0.000 claims 2
- 238000002372 labelling Methods 0.000 abstract description 2
- 238000004364 calculation method Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 4
- 230000003247 decreasing effect Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000032683 aging Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013524 data verification Methods 0.000 description 1
- 238000013079 data visualisation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000010287 polarization Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
- G01S13/50—Systems of measurement based on relative movement of target
- G01S13/58—Velocity or trajectory determination systems; Sense-of-movement determination systems
- G01S13/583—Velocity or trajectory determination systems; Sense-of-movement determination systems using transmission of continuous unmodulated waves, amplitude-, frequency-, or phase-modulated waves and based upon the Doppler effect resulting from movement of targets
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/415—Identification of targets based on measurements of movement associated with the target
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/417—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section involving the use of neural networks
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Social Psychology (AREA)
- Biophysics (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Psychiatry (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (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 human body tumbling detection method based on FMCW, which comprises the steps of firstly obtaining various human body behavior data, performing labeling treatment, and simultaneously calculating corresponding distance parameters, doppler speed parameters, horizontal angle parameters and pitch angle parameters according to the obtained intermediate frequency signals and the arrangement of receiving antennas; then, abnormal behaviors are extracted and identified by utilizing a neural network, and two kinds of centroid information including heart centroid information and human trunk centroid information are extracted according to the input point cloud information; secondly, the extracted feature vectors are input into a multi-layer perception neural network for training, whether abnormal behaviors occur in all acquired behavior data is judged, the directions of the two mass center information are judged by combining a threshold method, whether falling behaviors occur is comprehensively judged, and the falling behaviors can be comprehensively monitored.
Description
Technical Field
The invention relates to the technical field of human body fall detection, in particular to a human body fall detection method based on FMCW.
Background
At the same time of rapid development of technology, population aging has become a trend. The problem of aging population is more and more, and more countries pay attention to the health life problem of the old.
According to medical surveys, many elderly people may experience a single fall over the year, and such falls are often likely to occur in toilets, bathrooms. At present, traditional wearable fall detection equipment is likely to be removed when old people wash one's face and rinse one's mouth, and these products need to leave the body and charge, and the old people probably forget factors such as wearing, can not carry out comprehensive monitoring to old people's fall action.
Disclosure of Invention
The invention aims to provide a human body tumbling detection method based on FMCW, which can be used for comprehensively monitoring tumbling behaviors.
In order to achieve the above purpose, the invention provides a human body fall detection method based on FMCW, comprising the following steps:
extracting information according to the acquired behavior data, and performing space coordinate conversion;
extracting and identifying abnormal behaviors by using a neural network, and extracting two kinds of centroid information according to input point cloud information;
and judging the directions of the two centroid information by using a threshold method, and judging whether a tumbling action occurs or not by combining the training results of the feature vectors.
The method comprises the steps of extracting information according to the acquired behavior data and converting space coordinates, and comprises the following steps:
and acquiring various human behavior data, performing labeling, sequentially passing the acquired transmitting signals and echo signals through a mixer and a low-pass filter, and calculating corresponding distance parameters and Doppler speed parameters according to the obtained intermediate frequency signals.
The method comprises the steps of extracting information according to the acquired behavior data, converting space coordinates, and further comprising:
and obtaining corresponding horizontal angle parameters and pitch angle parameters through Fourier transformation phase changes of different channels according to the horizontal arrangement and the vertical arrangement of the receiving antennas.
The method for extracting and identifying abnormal behaviors by utilizing the neural network, and extracting two kinds of centroid information according to input point cloud information comprises the following steps:
and extracting corresponding heart centroid information according to the acquired point cloud information, and extracting human body trunk centroid information according to the space information set of all points.
The method for judging whether the tumbling action occurs or not by judging the directions of the two centroid information by using a threshold method and combining the training result of the feature vector comprises the following steps:
and inputting the extracted feature vectors into a multi-layer perception neural network for training, and judging whether abnormal behaviors occur in all the acquired behavior data.
The method for judging whether the tumbling action occurs or not by judging the directions of the two centroid information by using a threshold method and combining the training result of the feature vector comprises the following steps:
and judging whether the Z directions of the two kinds of centroid information are downward and larger than a set first threshold according to the extracted heart centroid information and the human body trunk centroid information, and judging whether the difference value of the set values of the two kinds of centroid information in the Z directions is within a set threshold range.
According to the FMCW-based human body tumbling detection method, various human body behavior data are firstly obtained, marking processing is carried out, and corresponding distance parameters, doppler speed parameters, horizontal angle parameters and pitch angle parameters are calculated according to the obtained intermediate frequency signals and the arrangement of receiving antennae; then, abnormal behaviors are extracted and identified by utilizing a neural network, and two kinds of centroid information including heart centroid information and human trunk centroid information are extracted according to the input point cloud information; secondly, the extracted feature vectors are input into a multi-layer perception neural network for training, whether abnormal behaviors occur in all acquired behavior data is judged, the directions of the two mass center information are judged by combining a threshold method, whether falling behaviors occur is comprehensively judged, and the falling behaviors can be comprehensively monitored.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic step diagram of a human body fall detection method based on FMCW according to the present invention.
Fig. 2 is a schematic flow chart of a human body fall detection method based on FMCW.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative and intended to explain the present invention and should not be construed as limiting the invention.
In the description of the present invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like indicate orientations or positional relationships merely to facilitate describing the present invention and simplify the description, and do not indicate or imply that the devices or elements being referred to must have a particular orientation, be configured and operated in a particular orientation, and thus should not be construed as limiting the present invention. Furthermore, in the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
Referring to fig. 1 and 2, the invention provides a human body fall detection method based on FMCW, comprising the following steps:
s101, extracting information according to the acquired behavior data, and performing space coordinate conversion.
Specifically, the frequency of the signal transmitted by the FMCW (Frequency Modulated Continuous Waves, modulated continuous wave) radar system increases linearly with time, and is transmitted again after a time interval Tc, and N chirp signals are transmitted in total; the synthesizer generates a chirp; then the antenna Tx is transmitted out; encountering an object in space, producing a reflected signal to be captured by the receiving antenna Rx; the mixer combines the Rx signals and the Tx signals together to generate an intermediate frequency signal IF; relevant parameters of the FMCW radar system are as follows:
table 1 relevant parameters of FMCW radar system
In order to distinguish the falling behaviors, the patent needs to collect other behavior data to distinguish the real falling behaviors, and the current scenes of the patent are mainly applied to the falling detection of the old, so that the behaviors at a slow speed and a normal speed can be distinguished in the behaviors, after the two types of data are respectively processed and the characteristics are extracted, the common characteristics are extracted, and the real-time detection of the falling behaviors is realized by the relative characteristic expression. The specific human behavior acquisition data list is as follows:
table 2 human behavior acquisition data list
| Human behavior | Specific behavior patterns (two major classes of slow and normal speed) |
| Tumble behavior | Forward, backward, leftward, rightward, and buttress |
| Squatting | Semi-squatting; squatting pan; squatting; |
| sit down | Sitting on a chair; is seated on a small stool, and the small stool is provided with a small seat, |
| get up | Standing up; from the chair; standing up from a small stool |
| Walk around | Randomly walking; swinging and walking; |
| running on | Randomly jogging; running while walking; |
in one aspect, FMCW data is acquired on a ROS platform; on the other hand, the data is labeled by using Kinect, the time stamps extracted from the two platforms ensure that the data are simultaneous, and label obtained from Kinect is transplanted to a label column of FMCW data.
Transmitting signal S t And echo signal S r Is input into a mixer to obtain a mixed signal S m Then the intermediate frequency signal S is obtained through a low-pass filter if And obtains the distance parameter R, the horizontal angle parameter theta, the vertical angle (azimuth angle) from the intermediate frequency signalDoppler velocity parameter D v 。
Distance R parameter
Distance R of radar signal is based on flight time delay t of signal back and forth d The method comprises the following steps:
and then according to the similar triangle similarity principle, the method can be used for obtaining:
this can be achieved by:
where c is the speed of light (constant), t s Is half of the period of the frequency modulated wave generated by the frequency generator; f (f) dev Sweep frequency bandwidth for the frequency modulation wave; f (f) b Is the difference between the transmitted and reflected frequencies.
After FFT conversion is carried out on the corresponding intermediate frequency signal data, an R peak value of a related distance can be obtained;
doppler velocity calculation D v
After FFT conversion is carried out on the intermediate frequency signal data, doppler FFT conversion is carried out on N equally spaced linear frequency modulation pulses, so that the speeds of different objects can be distinguished;
calculation of horizontal angle parameter θ
The phase change formula for the two receiving antennas is as follows:
on the XY plane, Δd=lsin (θ), where l is the distance between antennas. The calculation method of the angle is thus as follows:
pitch angle parameterCalculation of
Pitch angle parameterThe calculation mode of the receiving antennas is the same as the calculation principle of the horizontal angle parameter theta, and the arrangement modes of the receiving antennas are arranged in the vertical direction.
The FMCW used is a 4T4R antenna arrangement and can extract data in the Z direction. So according to the distance parameter R, the horizontal angle parameter θ, the vertical angle (azimuth angle)Doppler velocity parameter D v The method comprises the steps of carrying out a first treatment on the surface of the Can be converted into three-bit space point information, the conversion formula is as follows,
z＝Rconθ
thus, each valid data point is a four-dimensional point (x, y, z, D v )。
S102, utilizing a neural network to extract and identify abnormal behaviors, and extracting two kinds of centroid information according to input point cloud information.
Specifically, the behavior data captured by the present patent is dynamic behavior data, in order to extract the features of the data stream more abundantly, on one hand, a 3D reel neural network is used to extract the features, the hardware hard connection layer is in the form of (5×3×3), and the first convolution kernel is in the form of 16 layers (3*3); the second core is 32-layered (5*5), the third core is 64-layered (3*3); according to the fact that the falling process of old people in real life is probably slower, the hard connection layer is roughly set to data of 3 to 9 frames, and the data flow is roughly 100ms to 500ms long;
after the point cloud information of the human body is processed, the point cloud of the heart part of the human body is extracted, and the point cloud set at the position can be effectively mentioned because the reflection at the heart is more sensitive, and the centroid information (x heart ,y heart ,z heart ,D heart ) The method comprises the steps of carrying out a first treatment on the surface of the At the same time, according to the space information set of all points, the centroid information (x torse ,y torse ,z torse ,D torse ) The method comprises the steps of carrying out a first treatment on the surface of the And the representation of the mass centers of the two parts in the Z direction is emphasized and observed as an important judgment basis.
S103, judging the directions of the two centroid information by using a threshold method, and judging whether a tumbling action occurs or not by combining the training result of the feature vector.
Specifically, in model training, after the neural network extracts the feature values, the expanded feature vectors are put into the multi-layer perceptual neural network for training, and all the behaviors are non-tumbling behavior data. The overall model architecture is shown in fig. 2, and in actual monitoring, when a tumbling action occurs, the neural network above judges that the action is abnormal.
In all behavior data observations, the Doppler speed above the trunk of the effective point in most tumbling behaviors is accelerated to be increased to be decreased, and other daily behaviors are not obvious, so that the characteristic value is extracted from the 3DCNN, the information which is used for the Doppler speed as the fourth dimension is used as a color characteristic, and a colored 3DCNN model is formed.
In addition, after the effective heart centroid information and the trunk centroid information are mentioned in the data of 10 to 15 frames, judging whether the two centroid Z directions are downward or not respectively and are larger than a corresponding first threshold value, and finally judging that the last position of the Z direction is downward and the difference is within a threshold value range at the back, wherein the specific judging method is as follows:
wherein,a valid value is obtained through a large amount of data verification and testing.
The data visualization of this patent verifies and tests on ROS. In the actual test process, various behaviors are tested, more and richer data can be collected along with model iteration to improve the accuracy of neural network model judgment, and parameters are effectively adjusted. At present, under a real scene, the falling accuracy is 99%, and the false alarm rate is 2.5%.
The beneficial effect of this patent is:
1. establishing 4-dimensional information of point locations
The patent uses FMCW device containing 4T4R of longitudinal receiving antenna to mention four information of distance, polarization angle, inclination angle and Doppler speed of human body point in space, and can obtain the information of human body (x, y, z, D) in Cartesian coordinate system v ) The four-dimensional point information can describe the behavior of the human body in the space more effectively.
2. Effectively utilize the behavior characteristics of falling
The falling behavior detected in the patent refers to the behavior that a human body falls on the ground and does not climb up within a certain time. According to the human body posture expression of the tumbling behaviors, the Doppler speed appears when the point position above the trunk is excavated, the Doppler speed appears when the point position is gradually increased and then is rapidly decreased, and the characteristic that the Doppler speed is firstly increased and then rapidly decreased to zero in acceleration is realized. And the Doppler speed is used as a color channel in the 3DCNN to effectively extract the characteristic expression of the whole process.
In addition, after effective human point position information is extracted, a point cloud set of a heart part is found to obtain centroid information of the heart, centroid information of a human body is extracted, and according to the characteristic that the human body lies on the ground after falling is defined, the most direct Z-direction data representation is mined to comprehensively judge whether the abnormal behavior is falling behavior or not;
3. less calculation amount
Because the data performance characteristics of the tumbling behaviors and the point position performance characteristics in the Z direction are fully utilized in the whole model structure, the neural network structure is not complicated, and 20%,30% and 30% of Dropout are respectively in three convolution processes. And only one multi-layer sensing neural network is used for judging. The overall model has reasonable structural parameters and less calculated amount.
According to the FMCW-based human body tumbling detection method, various human body behavior data are firstly obtained, marking processing is carried out, and corresponding distance parameters, doppler speed parameters, horizontal angle parameters and pitch angle parameters are calculated according to the obtained intermediate frequency signals and the arrangement of receiving antennae; then, abnormal behaviors are extracted and identified by utilizing a neural network, and two kinds of centroid information including heart centroid information and human trunk centroid information are extracted according to the input point cloud information; secondly, the extracted feature vectors are input into a multi-layer perception neural network for training, whether abnormal behaviors occur in all acquired behavior data is judged, the directions of the two mass center information are judged by combining a threshold method, whether falling behaviors occur is comprehensively judged, and the falling behaviors can be comprehensively monitored.
The above disclosure is only a preferred embodiment of the present invention, and it should be understood that the scope of the invention is not limited thereto, and those skilled in the art will appreciate that all or part of the procedures described above can be performed according to the equivalent changes of the claims, and still fall within the scope of the present invention.
Claims (4)
1. The detection method of human body tumbling based on FMCW is characterized by comprising the following steps:
extracting information according to the acquired behavior data, and performing space coordinate conversion;
extracting and identifying abnormal behaviors by using a neural network, and extracting two kinds of centroid information according to input point cloud information;
judging the directions of the two centroid information by using a threshold method, and judging whether a tumbling action occurs or not by combining the training results of the feature vectors;
information extraction is carried out according to the acquired behavior data, and space coordinate conversion is carried out, and the method comprises the following steps:
acquiring various human behavior data, performing marking, and simultaneously sequentially passing the acquired transmitting signals and echo signals through a mixer and a low-pass filter, and calculating corresponding distance parameters and Doppler speed parameters according to the obtained intermediate frequency signals; and obtaining corresponding horizontal angle parameters and pitch angle parameters through Fourier transformation phase changes of different channels according to the horizontal arrangement and the vertical arrangement of the receiving antennas.
2. The FMCW-based human body fall detection method according to claim 1, wherein the abnormal behavior is identified by using neural network extraction, and the two kinds of centroid information are extracted according to the input point cloud information, comprising:
and extracting corresponding heart centroid information according to the acquired point cloud information, and extracting human body trunk centroid information according to the space information set of all points.
3. The FMCW-based human body fall detection method according to claim 2, wherein determining the directions of the two centroid information by using a thresholding method, and determining whether a fall behavior occurs in combination with a training result of a feature vector, comprises:
and inputting the extracted feature vectors into a multi-layer perception neural network for training, and judging whether abnormal behaviors occur in all the acquired behavior data.
4. The FMCW-based human body fall detection method according to claim 3, wherein the determining of the directions of the two kinds of centroid information by using a thresholding method, and determining whether a fall behavior occurs by combining training results of feature vectors, further comprises:
and judging whether the Z directions of the two kinds of centroid information are downward and larger than a set first threshold according to the extracted heart centroid information and the human body trunk centroid information, and judging whether the difference value of the set values of the two kinds of centroid information in the Z directions is within a set threshold range.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011210306.1A CN112180359B (en) | 2020-11-03 | 2020-11-03 | FMCW-based human body tumbling detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011210306.1A CN112180359B (en) | 2020-11-03 | 2020-11-03 | FMCW-based human body tumbling detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112180359A CN112180359A (en) | 2021-01-05 |
| CN112180359B true CN112180359B (en) | 2024-04-05 |
Family
ID=73916916
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011210306.1A Active CN112180359B (en) | 2020-11-03 | 2020-11-03 | FMCW-based human body tumbling detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112180359B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112578726A (en) * | 2021-01-06 | 2021-03-30 | 常州百芝龙智慧科技有限公司 | Human body abnormity monitoring equipment for independently learning human body behavior habits |
| CN117746510B (en) * | 2024-02-19 | 2024-05-28 | 河海大学 | Real-time three-dimensional behavior recognition method based on point cloud feature map |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104473648A (en) * | 2014-09-24 | 2015-04-01 | 上海大学 | Physiological parameter monitoring-combined human body tumble warning and detecting method |
| CN109726672A (en) * | 2018-12-27 | 2019-05-07 | 哈尔滨工业大学 | It is a kind of to fall down detection method based on human skeleton sequence and convolutional neural networks |
| CN110738154A (en) * | 2019-10-08 | 2020-01-31 | 南京熊猫电子股份有限公司 | pedestrian falling detection method based on human body posture estimation |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SE0203483D0 (en) * | 2002-11-21 | 2002-11-21 | Wespot Ab | Method and device for fall detection |
| US9582072B2 (en) * | 2013-09-17 | 2017-02-28 | Medibotics Llc | Motion recognition clothing [TM] with flexible electromagnetic, light, or sonic energy pathways |
| US10321873B2 (en) * | 2013-09-17 | 2019-06-18 | Medibotics Llc | Smart clothing for ambulatory human motion capture |
| CN111134685B (en) * | 2018-11-02 | 2022-08-09 | 富士通株式会社 | Fall detection method and device |
-
2020
- 2020-11-03 CN CN202011210306.1A patent/CN112180359B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104473648A (en) * | 2014-09-24 | 2015-04-01 | 上海大学 | Physiological parameter monitoring-combined human body tumble warning and detecting method |
| CN109726672A (en) * | 2018-12-27 | 2019-05-07 | 哈尔滨工业大学 | It is a kind of to fall down detection method based on human skeleton sequence and convolutional neural networks |
| CN110738154A (en) * | 2019-10-08 | 2020-01-31 | 南京熊猫电子股份有限公司 | pedestrian falling detection method based on human body posture estimation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112180359A (en) | 2021-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108226892B (en) | Deep learning-based radar signal recovery method in complex noise environment | |
| US11906616B2 (en) | Method of target feature extraction based on millimeter-wave radar echo | |
| CN111134685B (en) | Fall detection method and device | |
| CN112180359B (en) | FMCW-based human body tumbling detection method | |
| CN112346050B (en) | Fall detection method and system based on Wi-Fi equipment | |
| CN112998668B (en) | Millimeter wave-based non-contact far-field multi-human-body respiration heart rate monitoring method | |
| CN111738060A (en) | Human gait recognition system based on millimeter wave radar | |
| CN109711251B (en) | Commercial Wi-Fi-based independent direction gait recognition method | |
| CN108872977B (en) | Life body double-station cooperative detection method based on single-channel ultra-wideband radar | |
| CN105913074A (en) | Combined SAR image moving target clustering method based on amplitude and radial speed | |
| CN114814832A (en) | Millimeter wave radar-based real-time monitoring system and method for human body falling behavior | |
| Janakaraj et al. | STAR: Simultaneous tracking and recognition through millimeter waves and deep learning | |
| CN116148801B (en) | Millimeter wave radar-based target detection method and system | |
| CN110200607A (en) | Method for eliminating body motion influence in vital sign detection based on optical flow method and LMS algorithm | |
| CN112200146A (en) | Gesture recognition detection method based on FMCW | |
| Yang et al. | Spatial-temporal-circulated GLCM and physiological features for in-vehicle people sensing based on IR-UWB radar | |
| CN116338684A (en) | Human body falling detection method and system based on millimeter wave radar and deep learning | |
| CN114847931A (en) | Human motion tracking method, device and computer-readable storage medium | |
| DE102019128729A1 (en) | Driving support device, vehicle and driving support method | |
| Kittiyanpunya et al. | Millimeter-Wave Radar-Based Elderly Fall Detection Fed by One-Dimensional Point Cloud and Doppler | |
| CN112741617A (en) | CSI-based omnidirectional gait detection algorithm | |
| Sona et al. | A measurement approach based on micro-Doppler maps for human motion analysis and detection | |
| US20210223389A1 (en) | Synthetic aperture radar signal analysis device, synthetic aperture radar signal analysis method, and synthetic aperture radar signal analysis program | |
| JP2021179323A (en) | Attribute identification device and identification method | |
| Tahmoush | Micro-range micro-Doppler for classification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |