CN114994663A - Fall detection method and system based on millimeter wave radar - Google Patents
Fall detection method and system based on millimeter wave radar Download PDFInfo
- Publication number
- CN114994663A CN114994663A CN202210701234.3A CN202210701234A CN114994663A CN 114994663 A CN114994663 A CN 114994663A CN 202210701234 A CN202210701234 A CN 202210701234A CN 114994663 A CN114994663 A CN 114994663A
- Authority
- CN
- China
- Prior art keywords
- target
- human body
- millimeter wave
- wave radar
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
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
- G01S13/88—Radar or analogous systems specially adapted for specific applications
-
- 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
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/886—Radar or analogous systems specially adapted for specific applications for alarm systems
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
- G08B21/043—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis detecting an emergency event, e.g. a fall
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- General Health & Medical Sciences (AREA)
- Gerontology & Geriatric Medicine (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention relates to the technical field of fall detection, and particularly discloses a fall detection method based on a millimeter wave radar, which comprises the following steps: acquiring intermediate frequency ADC data through a millimeter wave radar; performing signal preprocessing operation on the intermediate frequency ADC data to obtain moving target point cloud data; identifying the current posture of a target human body according to the moving target point cloud data; and judging the falling state of the target human body according to the current posture recognition result of the target human body. The invention also discloses a fall detection system based on the millimeter wave radar. The method for detecting the falling of the millimeter wave radar based on the invention can select the millimeter wave radar with better environmental adaptability, does not relate to the privacy problem, not only judges the falling state through the information of the radar, but also adds means such as behavior logic judgment, vital sign detection and the like, and improves the falling accuracy of the millimeter wave radar.
Description
Technical Field
The invention relates to the technical field of fall detection, in particular to a fall detection method based on a millimeter wave radar and a fall detection system based on the millimeter wave radar.
Background
The problem of old people nursing has become a focus of government and social attention. Under the narrow and slippery environment of a bathroom or a kitchen, the old and even adults are easy to fall accidentally, and the timely rescue of the old after falling is very important. Therefore, how to detect the falling behavior in real time and output correct alarm information in time is an important research topic.
Sensors used in the current fall detection technology mainly comprise an acceleration sensor, a camera and a radar. The acceleration sensor detects displacement information through a plurality of axial accelerations so as to judge whether a person falls down, but the wearable sensor is not suitable for occasions such as a bathroom and causes discomfort after being worn for a long time; the camera judges the posture of a person through an image recognition technology, does not need to be worn on the body, but relates to the privacy problem, and the detection performance can be greatly influenced in a dark place of the environment.
The falling detection methods based on the radar are all to extract human body posture information acquired by the radar and then identify falling postures by using different networks. However, due to the need to collect a large number of data samples, covering the data feature differences caused by various falling postures and various environments, the accuracy and the adaptability thereof still need to be improved.
Disclosure of Invention
In order to solve the defects in the prior art, the invention provides a method and a system for detecting falling based on a millimeter wave radar, the millimeter wave radar with better environmental adaptability is selected, the privacy problem is not involved, the falling state is judged through the information of the radar, and measures such as behavior logic judgment, vital sign detection and the like are added, so that the falling accuracy of the millimeter wave radar is improved.
As a first aspect of the present invention, there is provided a fall detection method based on a millimeter wave radar, including the steps of:
acquiring intermediate frequency ADC data through a millimeter wave radar;
performing signal preprocessing operation on the intermediate frequency ADC data to obtain moving target point cloud data;
identifying the current posture of the target human body according to the moving target point cloud data;
and judging the falling state of the target human body according to the current posture recognition result of the target human body.
Further, the method also comprises the following steps:
step S1: transmitting electromagnetic wave signals to a detected area through a millimeter wave radar, receiving the electromagnetic wave signals reflected by each target in the detected area, and processing the electromagnetic wave signals reflected by each target to obtain intermediate frequency ADC data;
step S2: performing signal preprocessing operation on the intermediate frequency ADC data to obtain moving target point cloud data with noise points and multipath removed;
step S3: performing sliding accumulation on distance dimensional data in the signal preprocessing process to generate and store a target distance-time map;
step S4: judging whether the number of the moving target point cloud data output in the step S2 exceeds a threshold value N, if so, entering a step S5; if not, go to step S13;
step S5: processing the moving target point cloud data to obtain a target tracking list;
step S6: inputting the feature set in the target tracking list into a posture recognition model to judge the current human body posture information;
step S7: sliding and accumulating the current human body posture information judged and obtained in the step S6 according to a time sequence;
step S8: judging whether the target human body is in a falling state according to the accumulated current human body posture information, if so, entering the step S9; if not, go to step S10;
step S9: outputting a primary alarm signal, wherein the state is that a living body with a target falls down;
step S10: judging whether the target human body posture information in the previous action time window falls down, if so, entering the step S11; if not, go to step S12;
step S11: outputting a primary alarm release signal in a state that a moving target living body exists;
step S12: outputting no alarm signal and judging that a moving target living body exists;
step S13: recognizing the current human body posture information as an unknown state and accumulating;
step S14: performing target position locking and phase extraction operations according to the target distance-time map stored in the step S3 to obtain a target body motion signal;
step S15: judging whether the human body respiration value, the heartbeat value and the waveforms of the human body respiration value and the heartbeat value can be extracted from the target body motion signal or not; if it is judged that the extraction is possible, the process proceeds to step S17; if it is determined that the extraction cannot be performed, the process proceeds to step S16;
step S16: outputting a non-alarm signal, wherein the state is that no living body exists;
step S17: performing sliding accumulation on the time dimension on the extracted human body respiration value and the extracted heartbeat value;
step S18: judging whether the target human body posture information in the previous action time window falls down or not according to the human body posture information accumulated in the step S13, if so, entering a step S20; if not, go to step S19;
step S19: outputting a non-alarm signal, wherein the state is that a static target living body exists;
step S20: and outputting final alarm signals and the human respiration value and the heartbeat value accumulated in a fixed time period, wherein the state is that a static target living body is dangerous.
Further, in step S3, the distance dimension data is data obtained by performing distance dimension FFT on the intermediate frequency ADC data.
Further, the step S6 further includes:
and accumulating the distance, speed and angle information in the target tracking list on a time dimension to obtain a distance-time, speed-time and angle-time feature set, and inputting the distance-time, speed-time and angle-time feature set into the gesture recognition model to judge the current human body gesture information.
As a second aspect of the present invention, there is provided a fall detection system based on millimeter wave radar, configured to implement the fall detection method based on millimeter wave radar described in any one of the foregoing, including:
the acquisition module is used for acquiring intermediate frequency ADC data through a millimeter wave radar;
the processing module is used for carrying out signal preprocessing operation on the intermediate frequency ADC data to obtain moving target point cloud data;
the identification module is used for identifying the current posture of the target human body according to the moving target point cloud data;
and the judging module is used for judging the falling state of the target human body according to the current posture recognition result of the target human body.
The fall detection method and system based on the millimeter wave radar provided by the invention have the following advantages:
1. the millimeter wave radar technology is used, privacy is protected, and the method is widely suitable for various environments;
2. the radar outputs richer personnel attitude information through signal processing;
3. behavior logic judgment is added on the basis of radar falling judgment, so that the detection accuracy is improved;
4. a radar breath and heartbeat detection function is added to improve rich information for alarming;
5. the millimeter wave radar is used for storing effective data, various cloud platforms and peripheral equipment do not need to be carried, and personal information is protected.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention.
Fig. 1 is a flow chart of a fall detection method based on millimeter wave radar according to the present invention.
Fig. 2 is a flowchart of a fall detection method based on millimeter wave radar according to an embodiment of the present invention.
Detailed Description
To further illustrate the technical means and effects of the present invention adopted to achieve the predetermined object, the following detailed description will be given to the specific implementation, structure, features and effects of the fall detection method and system based on millimeter wave radar according to the present invention with reference to the accompanying drawings and preferred embodiments. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
In this embodiment, a fall detection method based on a millimeter wave radar is provided, as shown in fig. 1, the fall detection method based on the millimeter wave radar includes:
acquiring intermediate frequency ADC data through a millimeter wave radar;
performing signal preprocessing operation on the intermediate frequency ADC data to obtain moving target point cloud data;
identifying the current posture of a target human body according to the moving target point cloud data;
and judging the falling state of the target human body according to the current posture recognition result of the target human body.
Preferably, as shown in fig. 2, the method further comprises the following steps:
step S1: transmitting electromagnetic wave signals to a detected area through a millimeter wave radar, receiving the electromagnetic wave signals reflected by each target in the detected area, and processing the electromagnetic wave signals reflected by each target to obtain intermediate frequency ADC data;
it should be understood that the radar hardware system includes signal transmission, signal reception, difference frequency, gain amplification, analog-to-digital conversion, storage and other preprocessing, and intermediate frequency ADC data is obtained after preprocessing.
Step S2: performing signal preprocessing operation on the intermediate frequency ADC data to obtain moving target point cloud data with noise points and multipath removed;
step S3: performing sliding accumulation on distance dimensional data in the signal preprocessing process to generate and store a target distance-time map;
it should be noted that, the sliding accumulation here means that, if M frames are set in the time dimension, after the accumulated distance dimension data reaches M frames, the distance dimension data of the oldest frame is replaced with the distance dimension data of the newest frame; it is understood that the data is first in first out, and the oldest frame of data is automatically discarded, and the newest frame of data is placed at the nearest position.
It should be noted that the distance dimension data of the M accumulated frames is a two-dimensional matrix of M × N (N is the distance dimension FFT point number), and a target distance-time map is generated and stored, where the vertical axis of the target distance-time map is the target distance, and the horizontal axis is the time series.
Step S4: judging whether the number of the moving target point cloud data output in the step S2 exceeds a threshold value N, if so, entering a step S5; if not, go to step S13;
step S5: processing the moving target point cloud data to obtain a target tracking list;
it should be noted that the data processing method includes algorithms such as clustering and tracking.
Step S6: inputting the feature set in the target tracking list into a posture recognition model to judge the current human body posture information;
step S7: sliding and accumulating the current human body posture information judged and obtained in the step S6 according to a time sequence; wherein, the accumulated duration can be set, and the data is first in first out;
step S8: judging whether the target human body is in a falling state according to the accumulated current human body posture information, if so, entering the step S9; if not, go to step S10;
step S9: outputting a primary alarm signal, wherein the state is that a living body with a target falls down;
step S10: judging whether the target human body posture information in the previous action time window falls down, if so, entering the step S11; if not, go to step S12;
it should be noted that, here, an action time window may be understood as a time period T that it takes for a human body action to occur, that is, whether the target posture in the previous time period T is a fall is determined.
Step S11: outputting a primary alarm release signal in a state that a moving target living body exists;
step S12: outputting a non-alarm signal, wherein the state is that a moving target living body exists;
step S13: at the moment, the moving target point cloud data is not enough for gesture recognition and judgment, and at the moment, the current human body gesture information is recognized as an unknown state and accumulated;
step S14: performing target position locking and phase extraction operations according to the target distance-time map stored in the step S3 to obtain a target body motion signal;
step S15: judging whether the human body respiration value, the heartbeat value and the waveforms of the human body respiration value and the heartbeat value can be extracted from the target body motion signal or not; if it is judged that the extraction is possible, the process proceeds to step S17; if it is determined that the extraction cannot be performed, the process proceeds to step S16;
step S16: outputting no alarm signal and judging whether a living body exists;
step S17: performing sliding accumulation on the time dimension on the extracted human body respiration value and the extracted heartbeat value; wherein, the accumulated duration can be set, and the data is first in first out;
step S18: judging whether the target human body posture information in the previous action time window falls down or not according to the human body posture information accumulated in the step S13, if so, entering a step S20; if not, go to step S19;
here, the operation time window definition coincides with step S10.
Step S19: outputting a non-alarm signal, wherein the state is that a static target living body exists;
step S20: and outputting final alarm signals and the human respiration value and the heartbeat value accumulated in a fixed time period, wherein the state is that a static target living body is dangerous.
Preferably, in step S3, the distance dimension data is data obtained by performing distance dimension FFT on the intermediate frequency ADC data.
Preferably, in step S6, the method further includes:
and accumulating the distance, speed and angle information in the target tracking list on a time dimension to obtain a distance-time, speed-time and angle-time feature set, and inputting the distance-time, speed-time and angle-time feature set into the gesture recognition model to judge the current human body gesture information.
As another embodiment of the present invention, there is provided a fall detection system based on millimeter wave radar, including:
the acquisition module is used for acquiring intermediate frequency ADC data through a millimeter wave radar;
the processing module is used for carrying out signal preprocessing operation on the intermediate frequency ADC data to obtain moving target point cloud data;
the identification module is used for identifying the current posture of the target human body according to the moving target point cloud data;
and the judging module is used for judging the falling state of the target human body according to the current posture recognition result of the target human body.
The invention provides a falling detection method based on a millimeter wave radar, which can detect dynamic target points in an environment through the millimeter wave radar, recognize postures according to the characteristics of the dynamic target points, add behavior logic judgment and analysis of vital sign signals of human breath and heartbeat, and integrate all information to judge falling states.
Although the present invention has been described with reference to a preferred embodiment, it should be understood that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (5)
1. A fall detection method based on a millimeter wave radar is characterized by comprising the following steps:
acquiring intermediate frequency ADC data through a millimeter wave radar;
performing signal preprocessing operation on the intermediate frequency ADC data to obtain moving target point cloud data;
identifying the current posture of a target human body according to the moving target point cloud data;
and judging the falling state of the target human body according to the current posture recognition result of the target human body.
2. The millimeter wave radar-based fall detection method according to claim 1, further comprising the steps of:
step S1: transmitting electromagnetic wave signals to a detected area through a millimeter wave radar, receiving the electromagnetic wave signals reflected by each target in the detected area, and processing the electromagnetic wave signals reflected by each target to obtain intermediate frequency ADC data;
step S2: performing signal preprocessing operation on the intermediate frequency ADC data to obtain moving target point cloud data with noise points and multipath removed;
step S3: performing sliding accumulation on distance dimensional data in the signal preprocessing process to generate and store a target distance-time map;
step S4: judging whether the number of the moving target point cloud data output in the step S2 exceeds a threshold value N, if so, entering a step S5; if not, go to step S13;
step S5: processing the moving target point cloud data to obtain a target tracking list;
step S6: inputting the feature set in the target tracking list into a posture recognition model to judge the current human body posture information;
step S7: sliding and accumulating the current human body posture information judged and obtained in the step S6 according to a time sequence;
step S8: judging whether the target human body is in a falling state according to the accumulated current human body posture information, if so, entering the step S9; if not, go to step S10;
step S9: outputting a primary alarm signal, wherein the state is that a living body with a target falls down;
step S10: judging whether the target human body posture information in the previous action time window falls down, if so, entering the step S11; if not, go to step S12;
step S11: outputting a primary alarm release signal in a state that a moving target living body exists;
step S12: outputting no alarm signal and judging that a moving target living body exists;
step S13: recognizing the current human body posture information as an unknown state and accumulating;
step S14: performing target position locking and phase extraction operations according to the target distance-time map stored in the step S3 to obtain a target body motion signal;
step S15: judging whether the human body respiration value, the heartbeat value and the waveforms of the human body respiration value and the heartbeat value can be extracted from the target body movement signal or not; if it is judged that the extraction is possible, the process proceeds to step S17; if it is determined that the extraction cannot be performed, the process proceeds to step S16;
step S16: outputting no alarm signal and judging whether a living body exists;
step S17: performing sliding accumulation on the time dimension on the extracted human body respiration value and the extracted heartbeat value;
step S18: judging whether the target human body posture information in the previous action time window falls down or not according to the human body posture information accumulated in the step S13, if so, entering a step S20; if not, go to step S19;
step S19: outputting no alarm signal and the state is that a static target living body exists;
step S20: and outputting final alarm signals and the human respiration value and the heartbeat value accumulated in a fixed time period, wherein the state is that a static target living body is dangerous.
3. The fall detection method based on millimeter wave radar according to claim 2, wherein in the step S3, the distance dimension data is obtained by performing distance dimension FFT on the intermediate frequency ADC data.
4. The millimeter wave radar-based fall detection method according to claim 2, wherein in step S6, the method further comprises:
and accumulating the distance, speed and angle information in the target tracking list on a time dimension to obtain a distance-time, speed-time and angle-time feature set, and inputting the distance-time, speed-time and angle-time feature set into the gesture recognition model to judge the current human body gesture information.
5. A fall detection system based on millimeter wave radar for implementing the fall detection method based on millimeter wave radar according to any one of claims 1 to 4, comprising:
the acquisition module is used for acquiring intermediate frequency ADC data through a millimeter wave radar;
the processing module is used for carrying out signal preprocessing operation on the intermediate frequency ADC data to obtain moving target point cloud data;
the identification module is used for identifying the current posture of the target human body according to the moving target point cloud data;
and the judging module is used for judging the falling state of the target human body according to the current posture recognition result of the target human body.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210701234.3A CN114994663A (en) | 2022-06-21 | 2022-06-21 | Fall detection method and system based on millimeter wave radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210701234.3A CN114994663A (en) | 2022-06-21 | 2022-06-21 | Fall detection method and system based on millimeter wave radar |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114994663A true CN114994663A (en) | 2022-09-02 |
Family
ID=83036724
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210701234.3A Pending CN114994663A (en) | 2022-06-21 | 2022-06-21 | Fall detection method and system based on millimeter wave radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114994663A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116106855A (en) * | 2023-04-13 | 2023-05-12 | 中国科学技术大学 | Tumble detection method and tumble detection device |
| CN117831224A (en) * | 2024-02-29 | 2024-04-05 | 深圳市迈远科技有限公司 | Fall alarm method, device, equipment and medium based on millimeter radar |
-
2022
- 2022-06-21 CN CN202210701234.3A patent/CN114994663A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116106855A (en) * | 2023-04-13 | 2023-05-12 | 中国科学技术大学 | Tumble detection method and tumble detection device |
| CN117831224A (en) * | 2024-02-29 | 2024-04-05 | 深圳市迈远科技有限公司 | Fall alarm method, device, equipment and medium based on millimeter radar |
| CN117831224B (en) * | 2024-02-29 | 2024-05-24 | 深圳市迈远科技有限公司 | Fall alarm method, device, equipment and medium based on millimeter radar |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114994663A (en) | Fall detection method and system based on millimeter wave radar | |
| CN107749143B (en) | WiFi signal-based system and method for detecting falling of personnel in through-wall room | |
| Planinc et al. | Introducing the use of depth data for fall detection | |
| CN108734055B (en) | Abnormal person detection method, device and system | |
| EP2464991B1 (en) | A method for human only activity detection based on radar signals | |
| Liu et al. | Fall detection using doppler radar and classifier fusion | |
| CN111505632A (en) | Ultra-wideband radar action attitude identification method based on power spectrum and Doppler characteristics | |
| CN114469076B (en) | Identity-feature-fused fall identification method and system for solitary old people | |
| CN112327288B (en) | Radar human body action recognition method, radar human body action recognition device, electronic equipment and storage medium | |
| Ulrich et al. | Person recognition based on micro-Doppler and thermal infrared camera fusion for firefighting | |
| CN114814832A (en) | Millimeter wave radar-based real-time monitoring system and method for human body falling behavior | |
| Wu et al. | A novel high performance human detection, tracking and alarm system based on millimeter-wave radar | |
| Mohamed et al. | Fall detection systems for elderly care: A survey | |
| CN113945915A (en) | Method and device for detecting falling by millimeter wave radar | |
| Ding et al. | A fall detection method based on K-nearest neighbor algorithm with MIMO millimeter-wave radar | |
| CN110800053A (en) | Method and apparatus for obtaining event indications based on audio data | |
| Merrouche et al. | Fall detection using head tracking and centroid movement based on a depth camera | |
| Qu et al. | Convolutional neural network for human behavior recognition based on smart bracelet | |
| CN115422962A (en) | Gesture and gesture recognition method and device based on millimeter wave radar and deep learning algorithm | |
| CN115291184B (en) | Attitude monitoring method combining millimeter wave radar and deep learning | |
| Saleh et al. | An efficient machine learning-based fall detection algorithm using local binary features | |
| Bansal et al. | Elderly people fall detection system using skeleton tracking and recognition | |
| CN116243303A (en) | Indoor personnel target feature extraction method based on millimeter wave radar | |
| CN114580473A (en) | Radar-based human behavior identification method and system | |
| CN112597961B (en) | Interest target extraction method and system based on big data |
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 |