CN108257352B - Fall detection and early warning method based on intelligent wearable equipment - Google Patents
Fall detection and early warning method based on intelligent wearable equipment Download PDFInfo
- Publication number
- CN108257352B CN108257352B CN201711489128.9A CN201711489128A CN108257352B CN 108257352 B CN108257352 B CN 108257352B CN 201711489128 A CN201711489128 A CN 201711489128A CN 108257352 B CN108257352 B CN 108257352B
- Authority
- CN
- China
- Prior art keywords
- acceleration
- value
- data
- early warning
- warning method
- 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
Classifications
-
- 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
- G08B21/0438—Sensor means for detecting
- G08B21/0446—Sensor means for detecting worn on the body to detect changes of posture, e.g. a fall, inclination, acceleration, gait
-
- 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
- G08B21/0438—Sensor means for detecting
- G08B21/0453—Sensor means for detecting worn on the body to detect health condition by physiological monitoring, e.g. electrocardiogram, temperature, breathing
Abstract
The invention discloses a falling detection early warning method based on intelligent wearing equipment, which comprises the following steps: reading in training data; extracting characteristic parameters; training an SVM classifier; collecting sensing data; testing the acquired data by using an SVM classifier, judging whether a falling risk exists or not and prompting; and detecting whether the person falls down and alarming. The invention trains the SVM classifier according to the characteristic parameters extracted from the gait data of the healthy person and the abnormal gait data, and accurately distinguishes and identifies whether the gait is in an abnormal state, thereby realizing the pre-prompt of the possible falling condition; and then, under the condition of possible falling, the user is subjected to real-time falling detection, so that the safety of the user is protected. The falling detection early warning method based on the intelligent wearing equipment can be widely applied to the field of data processing.
Description
Technical Field
The invention relates to the field of data processing, in particular to a falling detection and early warning method based on intelligent wearing equipment.
Background
With the aging of population, the population over 60 years old in China exceeds 2.3 hundred million, the health and safety problems of the old have attracted the wide attention of society, and the old has decreased daily life capacity and has the problems of falling down, tendency, sudden diseases, forgetting to take medicine and the like. The existing intelligent wearable equipment can realize the detection of static parameters of blood pressure and heart rate and the detection of dynamic parameters such as falling, and then more feedback is obtained in the use process of the product, and the detection after falling can only be realized through the prior setting. For the elderly, once the elderly fall down, the joints may be damaged or even fractured, and the current fall-down detection technology can only realize the detection after the fall down and is difficult to meet other functional requirements.
In gait recognition, gait research on Parkinson patients is provided in the prior art, and the technical scheme adopts data based on the plantar pressure of the Parkinson patients. Firstly, a pressure sensor of a sole is adopted for data acquisition; secondly, the gait characteristics of the Parkinson patient are greatly different from the gait characteristics of the old people in the easy-to-fall state, and the falling state, the dangerous state and the normal state which are possibly fallen are difficult to distinguish by adopting the prior art scheme.
Disclosure of Invention
In order to solve the technical problems, the invention aims to: the method for realizing early warning of possible falling states and falling detection based on the intelligent wearable device is provided.
The technical scheme adopted by the invention is as follows: a fall detection early warning method based on intelligent wearing equipment comprises the following steps:
A. reading training data, wherein the training data comprises a plurality of groups of gait data of healthy people and a plurality of groups of abnormal gait data, and the training data is the acceleration of hand swing in walking;
B. for each training data set, searching the gait cycle in the training data set, and calculating the added speed wave trough value a in each gait cycle1Peak value a of acceleration2The valley value a of the acceleration1Peak value a of acceleration2Time t in between1Peak value a of acceleration2And the wave valley value a 'of the next acceleration'1Time t in between2Then calculating the mean of the acceleration wave-trough values in all gait cycles in the set of training dataAverage value of wave peak value of accelerationValley value a of acceleration1Peak value a of acceleration2Average value of time therebetweenPeak value a of acceleration2And the wave valley value a 'of the next acceleration'1Average value of time therebetween
C. By mean of valley values of accelerationAverage value of wave peak value of accelerationValley value a of acceleration1Peak value a of acceleration2Average value of time therebetweenPeak value a of acceleration2And the wave valley value a 'of the next acceleration'1Average value of time therebetweenConstructing an SVM classifier as a characteristic parameter for training;
D. acquiring sensing data through a sensor, wherein the sensor comprises an acceleration sensor, a gyroscope, a geomagnetic instrument and a gas pressure sensor;
E. testing the acceleration sensor data by adopting a trained SVM classifier, if the result is healthy human gait, not prompting, otherwise, sending a prompting signal and executing the step F;
F. and detecting whether the device is in a falling state according to the acquired sensing data, and if the device is in the falling state, sending an alarm signal.
Further, the step F specifically includes the following substeps:
f1, determining the vertical position variation of the user according to the atmospheric pressure of the air pressure sensor, taking the movement direction detected by the magnetometer as the reference direction, and determining the inclination angle in the movement direction according to the data detected by the gyroscope;
f2, judging whether the acceleration value and the movement direction detected by the acceleration sensor are both the reference direction and the vertical position variation of the user exceeds a set threshold value, if so, executing a step F3, otherwise, returning to the step D;
f3, collecting positioning data through a positioning module;
and F4, sending the positioning data and the alarm signal through the communication module.
Further, the step F3 is followed by the following steps:
f5, collecting heart rate data collected by a heart rate sensor, judging whether the heart rate of the user in preset time exceeds a set threshold value, if so, executing a step F6, otherwise, returning to the step D;
f6, judging whether the user presses a help-seeking key, if so, executing a step F7, otherwise, returning to the step D;
f7, controlling the voice communication module to start a speed dialing function;
f8, judging whether the call is successful, if so, entering a voice communication mode, otherwise, returning to the step F7.
Further, step length l of each gait cycle is calculated in the step B1And step size averageThe characteristic parameters in the step C also comprise step length average values
Further, the step B also calculates a trough value a of acceleration in each gait cycle1Peak value a of acceleration2Time t in between1Peak value a of acceleration2And the wave valley value a 'of the next acceleration'1Time t in between2Ratio t of1/t2The characteristic parameters in the step C also include a valley a of the acceleration1Peak value a of acceleration2Time t in between1Peak value a of acceleration2And the wave valley value a 'of the next acceleration'1Time t in between2Average value of the ratio of (1)
Further, the step B also calculates the velocity v corresponding to the first zero point of each gait cycle1Velocity v corresponding to the second zero point2Average value of rate corresponding to first zero pointAverage value of rate corresponding to second zero pointThe characteristic parameter in the step C further includes an average value of the rates corresponding to the first zero pointAverage value of rate corresponding to second zero point
The invention has the beneficial effects that: training an SVM classifier according to the gait data of the healthy person and the characteristic parameters extracted from the abnormal gait data, and accurately distinguishing and identifying whether the gait is in an abnormal state, so that the condition of possible falling is prompted in advance; and then, under the condition of possible falling, the user is subjected to real-time falling detection, so that the safety of the user is protected.
Drawings
FIG. 1 is a flow chart of the steps of the method of the present invention;
FIG. 2 is an ideal model of gait data of a healthy person according to the invention;
fig. 3 is an ideal model of abnormal gait data in the invention.
Detailed Description
The following further describes embodiments of the present invention with reference to the accompanying drawings:
referring to fig. 1, a fall detection and early warning method based on intelligent wearable equipment includes the following steps:
A. reading training data, wherein the training data comprises a plurality of groups of gait data of healthy people and a plurality of groups of abnormal gait data, and the training data is the acceleration of hand swing in walking;
firstly, accurately reflecting the hand swing state in a uniform motion state by adopting hand swing acceleration data, referring to a healthy person gait data ideal model shown in FIG. 2, when the time is 0, the acceleration is the maximum, the arm swings forwards from the last, the arm swing speed is the maximum when the first zero point is reached, then the speed is reduced until the arm swings to the forefront position, and thus the first half period of a gait is completed; the latter half cycle of a gait is then completed in the opposite direction. Generally, the old people fall down due to imbalance of muscle forces of two feet, and it is found in research that in the state of imbalance of muscle forces of two feet, in order to maintain balance, the forward and backward swinging speeds of the arm are different, and the abnormal gait data ideal model described with reference to fig. 3 is the biggest difference from fig. 2: a gait cycle is divided into two time intervals by the wave crest and the wave trough of the image, the lengths of the two time intervals are obviously different, and the speed corresponding to the zero point in the two time intervals is also obviously different. For the elderly, the degree of muscle imbalance is gradually increased, and a gradual process exists between normal gait and fall, so that the fall early warning prompt opportunity is provided in the process.
B. For each training data set, searching the gait cycle in the training data set, and calculating the added speed wave trough value a in each gait cycle1Peak value a of acceleration2The valley value a of the acceleration1Peak value a of acceleration2Time t in between1Peak value a of acceleration2And the wave valley value a 'of the next acceleration'1Time t in between2Then calculating the mean of the acceleration wave-trough values in all gait cycles in the set of training dataAverage value of wave peak value of accelerationValley value a of acceleration1Peak value a of acceleration2In betweenAverage value of timePeak value a of acceleration2And the wave valley value a 'of the next acceleration'1Average value of time therebetween
C. By mean of valley values of accelerationAverage value of wave peak value of accelerationValley value a of acceleration1Peak value a of acceleration2Average value of time therebetweenPeak value a of acceleration2And the wave valley value a 'of the next acceleration'1Average value of time therebetweenConstructing an SVM classifier as a characteristic parameter for training;
D. acquiring sensing data through a sensor, wherein the sensor comprises an acceleration sensor, a gyroscope, a geomagnetic instrument and a gas pressure sensor;
E. testing the acceleration sensor data by adopting a trained SVM classifier, if the result is healthy human gait, not prompting, otherwise, sending a prompting signal and executing the step F;
F. and detecting whether the device is in a falling state according to the acquired sensing data, and if the device is in the falling state, sending an alarm signal.
Since the actual raw data is different from the ideal model, for example, the starting point of the waveform is not necessarily the starting point of the gait cycle, and there may be a plurality of zero points or extreme points in a short time. Therefore, as a further preferred embodiment, when the gait cycle in the set of training data is searched in step B, the gait cycle is searched by searching for a zero point or searching for a peak or a trough.
Determining the gait cycle in a zero point searching manner is taken as an example:
searching zero points from left to right, continuously searching the zero points to the right by taking the first zero point as a starting point, and if a plurality of zero points exist in a time interval, taking the rightmost zero point as the starting point of a gait cycle;
continuously searching a zero point from left to right, continuously searching the zero point from right by taking the first zero point as a starting point, and if a plurality of zero points exist in a certain time interval, taking the rightmost zero point as the middle point of a gait cycle;
in order to avoid the difference between the lengths of the front half period and the rear half period in abnormal gait, continuously searching for a zero point from left to right, continuously searching for the zero point from right by taking the first zero point as a starting point, and if a plurality of zero points exist in a period of time, taking the rightmost zero point as the terminal point of the gait period; the gait cycle length is determined from the start and end points of the gait cycle.
The length of the time interval may be determined by searching for peaks and valleys, for example, by searching for a peak between the start of the gait cycle and the midpoint of the gait cycle, and then determining the start, midpoint and end of the gait cycle by using half of the time between the start of the gait cycle and the peak (about 1/4 gait cycles) as the length of the time interval.
Further as a preferred embodiment, the step F specifically includes the following sub-steps:
f1, determining the vertical position variation of the user according to the atmospheric pressure of the air pressure sensor, taking the movement direction detected by the magnetometer as the reference direction, and determining the inclination angle in the movement direction according to the data detected by the gyroscope;
f2, judging whether the acceleration value and the movement direction detected by the acceleration sensor are both the reference direction and the vertical position variation of the user exceeds a set threshold value, if so, executing a step F3, otherwise, returning to the step D;
f3, collecting positioning data through a positioning module;
and F4, sending the positioning data and the alarm signal through the communication module.
As a further preferred embodiment, the step F3 further includes the following steps:
f5, collecting heart rate data collected by a heart rate sensor, judging whether the heart rate of the user in preset time exceeds a set threshold value, if so, executing a step F6, otherwise, returning to the step D;
f6, judging whether the user presses a help-seeking key, if so, executing a step F7, otherwise, returning to the step D;
f7, controlling the voice communication module to start a speed dialing function;
f8, judging whether the call is successful, if so, entering a voice communication mode, otherwise, returning to the step F7.
Further preferably, the step length l of each gait cycle is calculated in step B1And step size averageThe characteristic parameters in the step C also comprise step length average values
Further preferably, in the step B, a trough value a of acceleration in each gait cycle is calculated1Peak value a of acceleration2Time t in between1Peak value a of acceleration2And the wave valley value a 'of the next acceleration'1Time t in between2Ratio t of1/t2The characteristic parameters in the step C also include a valley a of the acceleration1Peak value a of acceleration2Time t in between1Peak value a of acceleration2And the wave valley value a 'of the next acceleration'1Time t in between2Average value of the ratio of (1)
For healthy people, the ratio t in the gait data1/t2Ideally, the gait is close to 1, and the larger the deviation is, the more possible the gait is abnormal is, so that the healthy person gait and the abnormal gait can be effectively distinguished by the ratio.
Further preferably, in the step B, a velocity v corresponding to the first zero point of each gait cycle is calculated1Velocity v corresponding to the second zero point2Average value of rate corresponding to first zero pointAverage value of rate corresponding to second zero pointThe characteristic parameter in the step C further includes an average value of the rates corresponding to the first zero pointAverage value of rate corresponding to second zero point
Velocity v corresponding to the first zero point1Velocity v corresponding to the second zero point2The maximum speed values in the process of the forward swing and the backward swing of the arm are shown, so that the gait of a healthy person and the abnormal gait can be effectively distinguished through the two maximum speed values.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (6)
1. A fall detection early warning method based on intelligent wearing equipment is characterized by comprising the following steps:
A. reading training data, wherein the training data comprises a plurality of groups of gait data of healthy people and a plurality of groups of abnormal gait data, and the training data is the acceleration of hand swing in walking;
B. for each training data set, searching the gait cycle in the training data set, and calculating the added speed wave trough value a in each gait cycle1Peak value a of acceleration2The valley value a of the acceleration1Peak value a of acceleration2Time t in between1Peak value a of acceleration2And the wave valley value a 'of the next acceleration'1Time t in between2Then calculating the mean of the acceleration wave-trough values in all gait cycles in the set of training dataAverage value of wave peak value of accelerationValley value a of acceleration1Peak value a of acceleration2Average value of time therebetweenPeak value a of acceleration2And the wave valley value a 'of the next acceleration'1Average value of time therebetween
C. By mean of valley values of accelerationAverage value of wave peak value of accelerationValley value a of acceleration1Peak value a of acceleration2Average value of time therebetweenPeak value a of acceleration2And the wave valley value a 'of the next acceleration'1Average value of time therebetweenConstructing an SVM classifier as a characteristic parameter for training;
D. acquiring sensing data through a sensor, wherein the sensor comprises an acceleration sensor, a gyroscope, a geomagnetic instrument and a gas pressure sensor;
E. testing the acceleration sensor data by adopting a trained SVM classifier, if the result is healthy human gait, not prompting, otherwise, sending a prompting signal and executing the step F;
F. and detecting whether the device is in a falling state according to the acquired sensing data, and if the device is in the falling state, sending an alarm signal.
2. The fall detection and early warning method based on the intelligent wearable device as claimed in claim 1, wherein the fall detection and early warning method comprises the following steps: the step F specifically includes the following substeps:
f1, determining the vertical position variation of the user according to the atmospheric pressure of the air pressure sensor, taking the movement direction detected by the geomagnetic instrument as the reference direction, and determining the inclination angle in the movement direction according to the data detected by the gyroscope;
f2, judging whether the acceleration value and the movement direction detected by the acceleration sensor are both the reference direction and the vertical position variation of the user exceed the set threshold value, if so, executing the step
F3, otherwise, returning to the step D;
f3, collecting positioning data through a positioning module;
and F4, sending the positioning data and the alarm signal through the communication module.
3. The fall detection and early warning method based on the intelligent wearable device as claimed in claim 2, wherein the fall detection and early warning method comprises the following steps: the step F3 is further followed by the steps of:
f5, collecting heart rate data collected by a heart rate sensor, judging whether the heart rate of the user in preset time exceeds a set threshold value, if so, executing a step F6, otherwise, returning to the step D;
f6, judging whether the user presses a help-seeking key, if so, executing a step F7, otherwise, returning to the step D;
f7, controlling the voice communication module to start a speed dialing function;
f8, judging whether the call is successful, if so, entering a voice communication mode, otherwise, returning to the step F7.
4. The fall detection and early warning method based on the intelligent wearable device as claimed in claim 1, wherein the fall detection and early warning method comprises the following steps: step length l of each gait cycle is also calculated in the step B1And step size averageThe characteristic parameters in the step C also comprise step length average values
5. The fall detection and early warning method based on the intelligent wearable device as claimed in claim 1, wherein the fall detection and early warning method comprises the following steps: in the step B, a trough value a of acceleration in each gait cycle is calculated1Peak value a of acceleration2Time t in between1Peak value a of acceleration2And the wave valley value a 'of the next acceleration'1Time t in between2Ratio t of1/t2The characteristic parameters in the step C also include a valley a of the acceleration1Peak value a of acceleration2Time t in between1Peak value a of acceleration2And the wave valley value a 'of the next acceleration'1Time t in between2Average value of the ratio of (1)
6. The fall detection and early warning method based on the intelligent wearable device as claimed in claim 1, wherein the fall detection and early warning method comprises the following steps: in the step B, the velocity v corresponding to the first zero point of the acceleration data of each gait cycle is calculated1Velocity v corresponding to the second zero point of the acceleration data2Average value of velocity corresponding to first zero point of acceleration dataAverage value of velocity corresponding to second zero point of acceleration dataThe characteristic parameter in the step C further includes an average value of the velocity corresponding to the first zero point of the acceleration dataAverage value of velocity corresponding to second zero point of acceleration data
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711489128.9A CN108257352B (en) | 2017-12-30 | 2017-12-30 | Fall detection and early warning method based on intelligent wearable equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711489128.9A CN108257352B (en) | 2017-12-30 | 2017-12-30 | Fall detection and early warning method based on intelligent wearable equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108257352A CN108257352A (en) | 2018-07-06 |
| CN108257352B true CN108257352B (en) | 2019-12-20 |
Family
ID=62725277
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711489128.9A Active CN108257352B (en) | 2017-12-30 | 2017-12-30 | Fall detection and early warning method based on intelligent wearable equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108257352B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109166275B (en) * | 2018-09-25 | 2020-08-11 | 山东科技大学 | Human body falling detection method based on acceleration sensor |
| CN109620250B (en) * | 2019-02-22 | 2024-02-02 | 北京大学深圳医院 | Tremor detection prompting parkinsonian risk bracelet and use method thereof |
| CN111657918B (en) * | 2020-06-12 | 2021-09-24 | 电子科技大学 | Falling detection method and system combining electrocardio and inertial sensing data |
| CN112669568A (en) * | 2020-12-18 | 2021-04-16 | 浙江工商大学 | Multi-mode human body falling detection method |
| CN113662535B (en) * | 2021-09-14 | 2022-07-01 | 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) | Gait detection method, device, equipment and storage medium |
| CN114061616A (en) * | 2021-10-22 | 2022-02-18 | 北京自动化控制设备研究所 | Self-adaptive peak detection step counting method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177530A (en) * | 2013-02-25 | 2013-06-26 | 宁波中科集成电路设计中心有限公司 | Human fall detection method based on triaxial acceleration |
| US8909497B1 (en) * | 2010-04-06 | 2014-12-09 | Keynetik, Inc. | System and method for fall detection |
| CN105286874A (en) * | 2015-11-03 | 2016-02-03 | 长安大学 | System and method for predicting tumble risk of old people |
| CN105513277A (en) * | 2016-01-10 | 2016-04-20 | 南京邮电大学 | Old people falling alarm device |
| CN105686833A (en) * | 2016-01-13 | 2016-06-22 | 上海交通大学 | Wearable multinode three dimensional motion monitoring system based on wireless body area network |
-
2017
- 2017-12-30 CN CN201711489128.9A patent/CN108257352B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8909497B1 (en) * | 2010-04-06 | 2014-12-09 | Keynetik, Inc. | System and method for fall detection |
| CN103177530A (en) * | 2013-02-25 | 2013-06-26 | 宁波中科集成电路设计中心有限公司 | Human fall detection method based on triaxial acceleration |
| CN105286874A (en) * | 2015-11-03 | 2016-02-03 | 长安大学 | System and method for predicting tumble risk of old people |
| CN105513277A (en) * | 2016-01-10 | 2016-04-20 | 南京邮电大学 | Old people falling alarm device |
| CN105686833A (en) * | 2016-01-13 | 2016-06-22 | 上海交通大学 | Wearable multinode three dimensional motion monitoring system based on wireless body area network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108257352A (en) | 2018-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108257352B (en) | Fall detection and early warning method based on intelligent wearable equipment | |
| Noor et al. | Adaptive sliding window segmentation for physical activity recognition using a single tri-axial accelerometer | |
| US10306135B2 (en) | Automatic detection by a wearable camera | |
| CN106887115B (en) | A kind of Falls Among Old People monitoring device and fall risk appraisal procedure | |
| CN205041401U (en) | Equipment is worn detection device and is had and wears guardianship device that detects function | |
| US8812258B2 (en) | Identifying a type of motion of an object | |
| CN103810817B (en) | A kind of detection alarm method of the wearable human paralysis device of falling detection alarm | |
| Lan et al. | SmartFall: an automatic fall detection system based on subsequence matching for the SmartCane | |
| CN105118236A (en) | Paralysis falling detection and prevention device and processing method thereof | |
| CN106572814B (en) | Device and method for classifying user activity and/or counting user steps | |
| CN104814728B (en) | A kind of running bootstrap technique and running guide device | |
| Song et al. | Speed estimation from a tri-axial accelerometer using neural networks | |
| CN105051799A (en) | Method for detecting falls and a fall detector. | |
| RU2018121883A (en) | DETECTION OF AN ACCIDENT OF Drowsiness | |
| US20170241797A1 (en) | Pedometer with Accelerometer and Foot Motion Distinguishing Method | |
| CN103961109A (en) | Human body posture detection device based on acceleration signals and angular speed signals | |
| CN109166275A (en) | A kind of tumble detection method for human body based on acceleration transducer | |
| CN106725445B (en) | A kind of the portable body injury gained in sports monitor system and method for brain wave control | |
| CN107693022A (en) | A kind of method and device for detecting falling over of human body | |
| Zhao et al. | Recognition of human fall events based on single tri-axial gyroscope | |
| Merrouche et al. | Depth camera based fall detection using human shape and movement | |
| CN105654666A (en) | Old people falling monitoring method based on MEMS sensor | |
| Yang et al. | A wearable real-time fall detector based on Naive Bayes classifier | |
| CN203931101U (en) | A kind of wearable human paralysis device of falling detection alarm | |
| CN112233374B (en) | Fall detection method, system, terminal and storage medium |
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 | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 510663 unit 1106, C1, innovation building, 182 science Avenue, Science City, Guangzhou high tech Industrial Development Zone, Guangdong Province Patentee after: Guangzhou AI care Digital Technology Co.,Ltd. Address before: 510663 unit 1106, C1, innovation building, 182 science Avenue, Science City, Guangzhou high tech Industrial Development Zone, Guangdong Province Patentee before: GUANGZHOU BAIYI INFORMATION TECHNOLOGY Co.,Ltd. |