CN110793539A - Drowning detection method and system based on motion component threshold and machine learning - Google Patents
Drowning detection method and system based on motion component threshold and machine learning Download PDFInfo
- Publication number
- CN110793539A CN110793539A CN201911100919.7A CN201911100919A CN110793539A CN 110793539 A CN110793539 A CN 110793539A CN 201911100919 A CN201911100919 A CN 201911100919A CN 110793539 A CN110793539 A CN 110793539A
- Authority
- CN
- China
- Prior art keywords
- data
- motion
- machine learning
- swimming
- drowning
- 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
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C23/00—Combined instruments indicating more than one navigational value, e.g. for aircraft; Combined measuring devices for measuring two or more variables of movement, e.g. distance, speed or acceleration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Software Systems (AREA)
- Radar, Positioning & Navigation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Remote Sensing (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Medical Informatics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a drowning detection method and system based on motion component threshold and machine learning. And setting a horizontal displacement threshold value as a condition for judging possible drowning. When the pattern matching is unsuccessful and the early warning state is not drowned, the characteristic data is extracted by adopting a machine learning method and is stored in a pattern library in a classified manner, and the pattern library is continuously updated, so that the accuracy of the detection algorithm can be continuously improved.
Description
Technical Field
The invention relates to a swimming posture identification method based on a wearable sensor, in particular to a drowning state detection method based on combination of a motion component threshold value and machine learning.
Background
Swimming is a body-building sport which is very beneficial to body health and is widely popular with people. Swimming sports, however, also present a danger and drowning events occur every year in swimming pools around the world. With the development of sensor technology and computer technology, various sports wearable devices are appearing on the market. Wearable equipment based on sensor can be used for monitoring record human motion state, helps people to master self motion information. Such as a sports bracelet, can be used for running fitness, recording sports data and simple physiological information, such as heart rate and the like. The sports bracelet can record and analyze swimming data such as distance, swimming stroke, swimming speed and the like. At present, no sports bracelet can give an early warning to a drowning state possibly appearing in a swimming pool.
The method adopts a multi-sensor data fusion technology, filters noise interference, obtains human body posture data with high reliability and high precision, analyzes data characteristics, and judges the swimming posture of a human body, thereby identifying possible drowning tendency, and is very challenging work.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a drowning detection algorithm based on the combination of a motion component threshold and machine learning.
A first aspect of the invention provides a drowning detection method based on motion component thresholds and machine learning, comprising the steps of:
s1, collecting swimming motion data, including raw data of a nine-axis sensor in a wearable device worn on a swimmer;
s2, filtering and normalizing the swimming motion data, and extracting features to obtain a feature value;
s3, matching the characteristic value with a standard characteristic value in a swimming stroke data pattern library, if the matching is successful, no prompt is given, and the steps S1-S3 are repeated to continue data acquisition and matching; if the matching fails, go to step S4;
s4, carrying out fusion processing on the original data of the nine-axis sensor to obtain the motion components of the sensor data in different directions of a coordinate system;
s5, performing zero-speed correction on the motion component, and calculating the displacement in the horizontal direction based on the acceleration;
s6, comparing the calculated displacement with a preset displacement threshold, if the calculated displacement is smaller than the preset displacement threshold, judging that the swimming is in an abnormal swimming motion state, and giving an early warning signal; otherwise, go to step S7;
and S7, learning and classifying the characteristic value data acquired in the step S2 for multiple times by adopting a machine learning method, and storing the result as a standard characteristic value into the swimming stroke data pattern library.
Further, in step S1, the swimming motion data includes four swimming strokes of standard breaststroke, butterfly stroke, free stroke and backstroke, and nine-axis sensor raw data simulating a drowning state.
Further, in step S2, the obtained characteristic values include a peak value, a period, an average value, a standard deviation, a frequency peak value and a signal energy.
Further, in step S4, an extended kalman filter algorithm, a Madgwick algorithm, or a Mahony algorithm is used to perform fusion processing on the raw data of the nine-axis sensor, so as to obtain motion components of the sensor data in different directions of the coordinate system.
Further, in step S4, performing fusion processing on the raw data of the nine-axis sensor by using a Madgwick algorithm, and acquiring motion components of the sensor data in different directions of the coordinate system, specifically including:
s41, representing the motion posture of the human body by quaternion:
q=q0+q1i+q2j+q3k (1)
where q is a quaternion value, where q represents the azimuthal relationship of one coordinate system relative to another coordinate system, q0,q1,q2,q3Is four real numbers, i, j, k being units of imaginary numbers, i.e. i2=-1,j2=-1,k2=-1;
S42, calculating an estimated value of a quaternion at the time t:
qest,t=qest,t-1+kest,tΔt (2)
wherein q isest,tIs the quaternion estimate at time t, qest,t-1Is an estimate of time t-1, kest,tIs the scaling factor, Δ t is the sampling time interval;
s43, calculating the attitude inclination angle:
wherein, Yaw, Pitch and Roll are respectively a rolling angle, a Pitch angle and a horizontal angle;
and S44, acquiring acceleration components of the moving object in the directions of the three axes of x, y and z based on the calculated attitude dip angle data.
Further, in step S5, the method for performing zero velocity correction includes detecting acceleration moving variance, detecting acceleration amplitude, detecting angular velocity energy, or detecting acceleration angular velocity combination.
Further, in step S7, the Machine learning method includes a k-NN (k-Nearest Neighbor, k-Neighbor algorithm), LR (Logistic Regression), or SVM (Support Vector Machine) method.
A second aspect of the invention provides a drowning detection system based on motion component thresholds and machine learning, comprising the following modules:
the data acquisition module comprises a nine-axis sensor worn on a wearable device of a swimmer and is used for acquiring original swimming motion data;
the data processing module is used for carrying out filtering and normalization preprocessing on the original swimming motion data, carrying out feature extraction and obtaining a motion data feature value;
the database module is used for storing a swimming stroke data mode database formed by standard characteristic values corresponding to standard swimming sport data;
the pattern matching module is used for matching the motion data characteristic value with a standard characteristic value in a swimming stroke data pattern library, if the matching is successful, the processing is not carried out, and if the matching is unsuccessful, the drowning judgment processing is carried out;
the drowning judgment module is used for acquiring the displacement of the swimmer in the horizontal direction based on the original data, comparing the displacement with a preset displacement threshold value, judging that the swimmer is in an abnormal swimming motion state if the calculated displacement is smaller than the preset displacement threshold value, and giving an early warning signal; otherwise, a signal is sent to a machine learning and pattern library updating module;
and the machine learning and pattern library updating module starts a machine learning function after receiving a signal given by the drowning judging module, performs classified learning on the motion data characteristic value acquired by the data processing module, and updates the motion data characteristic value into the swimming stroke data pattern library of the database module.
Further, the system also comprises an early warning indication module which is used for receiving the early warning signal given by the drowning judgment module and sending out acousto-optic warning information.
Further, the system also comprises a communication module used for communicating with the external device for data interaction.
The invention has the following advantages:
1) the method is used for carrying out abnormal swimming stroke identification based on motion component decomposition and threshold values, and is simple in algorithm and small in calculated amount.
2) The invention adopts the nine-axis sensor to acquire the human motion data, and the data dimension can make the result more accurate.
3) The invention uses simple characteristic values such as peak value, average value, period and the like to store the swimming stroke mode, and has low requirements on the calculation capability and the storage capacity of the wearable equipment.
Drawings
FIG. 1 is a functional diagram of a device in which the method and system of the present invention may be implemented.
Fig. 2 is a flow diagram of a drowning detection method of the present invention.
Fig. 3 is a schematic flow chart of an embodiment of a drowning detection method of the present invention.
FIG. 4 shows the raw data collected by the sensors for four swimming gestures and simulated drowning in the embodiment of the present invention; wherein, (a) corresponds to the breaststroke state, (b) corresponds to the butterfly stroke state, (c) corresponds to the free stroke state, (d) corresponds to the backstroke state, and (e) corresponds to the simulated drowning state.
Fig. 5 shows four swimming strokes and horizontal displacement distances simulating drowning every 10s time interval in the embodiment of the present invention.
Detailed Description
For a further understanding of the invention, reference will now be made to the preferred embodiments of the invention by way of example, and it is to be understood that the description is intended to further illustrate features and advantages of the invention, and not to limit the scope of the claims.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
FIG. 1 is a functional diagram of an apparatus in which the method and system of the present invention may be implemented. The device mainly comprises five functional modules: the early warning system comprises a sensor module, a memory module, a processor module, a communication unit and an early warning module. Wherein the sensor module comprises an accelerometer, a gyroscope and a magnetometer, i.e. a nine-axis sensor module.
Based on the above device, the first embodiment of the present invention discloses a drowning detection method based on motion component threshold and machine learning, as shown in fig. 2 and 3, which includes the following steps:
step S201: and finishing swimming data acquisition to obtain original nine-axis sensor data. In the present embodiment, the data sampling frequency is 400 Hz. An example of the sampled data is shown in fig. 4, where (a) corresponds to a breaststroke state, (b) corresponds to a butterfly stroke state, (c) corresponds to a free-swimming state, (d) corresponds to a backstroke state, and (e) corresponds to a simulated drowning state. The volunteers participating in data acquisition are adult males and national second-level athletes, carry out data acquisition on the standard four swimming postures, and also carry out drowning simulation and data collection.
Step S202: and (5) data processing, namely performing filtering and normalization processing on the raw data acquired in the step S201. In the present embodiment, gaussian smoothing filtering with a window of 50 is performed, and Fast Fourier Transform (FFT) is performed to acquire frequency components.
Step S203: and (3) feature extraction, namely performing feature extraction on the data processed in the step S202. In this embodiment, the extracted characteristic values include a peak value, a period, an average value, a standard deviation, a frequency peak value, and signal energy.
Step S204: pattern matching, namely pre-storing feature data of four standard swimming gestures in a pattern library S210, matching the feature values obtained in the step 203 with the feature values in the pattern library, and continuing to acquire data without any prompt if the matching is successful; if the matching fails, the process proceeds to step S205.
Step S205: and (4) decomposing the motion components, namely fusing the nine-axis sensor data by adopting EKF (extended Kalman Filter) or Madgwick or Mahony algorithm to obtain the motion components of the sensor data in different directions of a coordinate system. The embodiment adopts a Madgwick method. Expressing the motion posture of the human body by quaternion:
q=q0+q1i+q2j+q3k (1)
where q is a quaternion value, where q represents the azimuthal relationship of one coordinate system relative to another coordinate system, q0,q1,q2,q3Is four real numbers, i, j, k being units of imaginary numbers, i.e. i2=-1,j2=-1,k2=-1;
And (3) calculating an estimated value of a quaternion at the time t:
qest,t=qest,t-1+kest,tΔt (2)
qest,tis the quaternion estimate at time t, qest,t-1Is an estimate of time t-1, kest,tIs the scaling factor, at is the sampling interval, which in this embodiment takes 25 ms.
The calculation formula of the attitude angle is as follows:
yaw, Pitch, Roll are Roll, Pitch and Yaw angles, respectively.
After the attitude dip angle is obtained, the acceleration components of the moving object in the directions of the three axes of x, y and z can be obtained.
Step S206: zero-speed correction, speed and displacement are generated by integration, the error becomes larger along with the increase of time, and an auxiliary correction algorithm is required to correct the error. There are various methods for detecting the zero velocity, such as acceleration moving variance detection, acceleration amplitude detection, angular velocity energy detection, acceleration angular velocity combination detection, etc. Acceleration amplitude detection is adopted in the embodiment (reference document: Chen Guang, Yang Zhou. step counting algorithm research based on acceleration measurement amplitude zero-speed detection [ J ]. Wuhan university Committee (information science edition), 2017,42 (06)).
Step S207: and (4) displacement calculation, namely obtaining the displacement by carrying out secondary integration on the acceleration. The present embodiment calculates the displacement in the horizontal direction for determining whether the human body movement corresponds to the swimming motion.
Step S208: and (5) judging a threshold value, and setting a threshold value range according to the normal swimming speed of the ordinary people. As shown in fig. 5, the lower limit of 2 meters for every 10s of moving distance is set in the present embodiment, and the abnormal swimming motion state is determined by less than 2 meters, and the algorithm gives the warning signal, step 209, and it is a preferable scheme to give a warning in the swimming pool by using a light signal.
Step S210: machine learning, when the pattern matching is unsuccessful and the horizontal displacement exceeds the threshold, indicates that the human body is in a normal swimming state, but the gesture is not in the pattern library (step S211). The data obtained in step S203 is learned and classified a plurality of times by using a machine learning method, and the result is stored in a pattern library. Many machine learning methods can be used, such as k-NN (k-nearest neighbor, k-neighbor algorithm), LR (Logistic Regression), or SVM (Support vector machine); this example employs the k-means method.
Based on the device in fig. 1, the first embodiment of the invention discloses a drowning detection system based on motion component threshold and machine learning, which comprises the following modules:
the data acquisition module comprises a nine-axis sensor worn on a wearable device of a swimmer and is used for acquiring original swimming motion data;
the data processing module is used for carrying out filtering and normalization preprocessing on the original swimming motion data, carrying out feature extraction and obtaining a motion data feature value;
the database module is used for storing a swimming stroke data mode database formed by standard characteristic values corresponding to standard swimming sport data;
the pattern matching module is used for matching the motion data characteristic value with a standard characteristic value in a swimming stroke data pattern library, if the matching is successful, the processing is not carried out, and if the matching is unsuccessful, the drowning judgment processing is carried out;
the drowning judgment module is used for acquiring the displacement of the swimmer in the horizontal direction based on the original data, comparing the displacement with a preset displacement threshold value, judging that the swimmer is in an abnormal swimming motion state if the calculated displacement is smaller than the preset displacement threshold value, and giving an early warning signal; otherwise, a signal is sent to a machine learning and pattern library updating module;
and the machine learning and pattern library updating module starts a machine learning function after receiving a signal given by the drowning judging module, performs classified learning on the motion data characteristic value acquired by the data processing module, and updates the motion data characteristic value into the swimming stroke data pattern library of the database module.
In a further preferred embodiment, the system further comprises an early warning indication module, which is used for receiving the early warning signal given by the drowning judgment module and sending out acousto-optic warning information.
In a further preferred embodiment, the system further comprises a communication module for communicating with an external device for data interaction.
The above description of the embodiments is only intended to facilitate the understanding of the method of the invention and its core idea. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
Claims (10)
1. A drowning detection method based on motion component threshold and machine learning is characterized by comprising the following steps:
s1, collecting swimming motion data, including raw data of a nine-axis sensor in a wearable device worn on a swimmer;
s2, filtering and normalizing the swimming motion data, and extracting features to obtain a feature value;
s3, matching the characteristic value with a standard characteristic value in a swimming stroke data pattern library, if the matching is successful, no prompt is given, and the steps S1-S3 are repeated to continue data acquisition and matching; if the matching fails, go to step S4;
s4, carrying out fusion processing on the original data of the nine-axis sensor to obtain the motion components of the sensor data in different directions of a coordinate system;
s5, performing zero-speed correction on the motion component, and calculating the displacement in the horizontal direction based on the acceleration;
s6, comparing the calculated displacement with a preset displacement threshold, if the calculated displacement is smaller than the preset displacement threshold, judging that the swimming is in an abnormal swimming motion state, and giving an early warning signal; otherwise, go to step S7;
and S7, learning and classifying the characteristic value data acquired in the step S2 for multiple times by adopting a machine learning method, and storing the result as a standard characteristic value into the swimming stroke data pattern library.
2. The drowning detection method based on motion component threshold and machine learning according to claim 1, wherein in step S1, the swimming motion data includes four swimming strokes of standard breaststroke, butterfly stroke, freestyle stroke and backstroke stroke and nine-axis sensor raw data simulating a drowning state.
3. The drowning detection method based on motion component threshold and machine learning according to claim 1, characterized in that in step S2, the obtained characteristic values include peak value, period, average value, standard deviation, frequency peak value and signal energy.
4. The drowning detection method based on motion component threshold and machine learning of claim 1, characterized in that in step S4, the extended kalman filter algorithm, the Madgwick algorithm, or the Mahony algorithm is used to perform the fusion processing on the raw data of the nine-axis sensor, so as to obtain the motion components of the sensor data in different directions of the coordinate system.
5. The drowning detection method based on motion component threshold and machine learning of claim 4, wherein in step S4, the method for obtaining the motion components of the sensor data in different directions of the coordinate system by fusing the raw data of the nine-axis sensor with the Madgwick algorithm specifically comprises:
s41, representing the motion posture of the human body by quaternion:
q=q0+q1i+q2j+q3k (1)
where q is a quaternion value, where q represents the azimuthal relationship of one coordinate system relative to another coordinate system, q0,q1,q2,q3Is four real numbers, i, j, k being units of imaginary numbers, i.e. i2=-1,j2=-1,k2=-1;
S42, calculating an estimated value of a quaternion at the time t:
qest,t=qest,t-1+kest,tΔt (2)
wherein q isest,tIs the quaternion estimate at time t, qest,t-1Is an estimate of time t-1,kest,tIs the scaling factor, Δ t is the sampling time interval;
s43, calculating the attitude inclination angle:
wherein, Yaw, Pitch and Roll are respectively a rolling angle, a Pitch angle and a horizontal angle;
and S44, acquiring acceleration components of the moving object in the directions of the three axes of x, y and z based on the calculated attitude dip angle data.
6. The drowning detection method based on motion component threshold and machine learning according to claim 1, characterized in that in step S5, the method for performing zero velocity correction includes acceleration moving variance detection, acceleration amplitude detection, angular velocity energy detection or acceleration angular velocity combination detection.
7. The drowning detection method based on motion component threshold and machine learning of claim 1, characterized in that in step S7, the machine learning method comprises k-neighbor method, logistic regression method or support vector machine method.
8. A drowning detection system based on motion component thresholds and machine learning, comprising the following modules:
the data acquisition module comprises a nine-axis sensor worn on a wearable device of a swimmer and is used for acquiring original swimming motion data;
the data processing module is used for carrying out filtering and normalization preprocessing on the original swimming motion data, carrying out feature extraction and obtaining a motion data feature value;
the database module is used for storing a swimming stroke data mode database formed by standard characteristic values corresponding to standard swimming sport data;
the pattern matching module is used for matching the motion data characteristic value with a standard characteristic value in a swimming stroke data pattern library, if the matching is successful, the processing is not carried out, and if the matching is unsuccessful, the drowning judgment processing is carried out;
the drowning judgment module is used for acquiring the displacement of the swimmer in the horizontal direction based on the original data, comparing the displacement with a preset displacement threshold value, judging that the swimmer is in an abnormal swimming motion state if the calculated displacement is smaller than the preset displacement threshold value, and giving an early warning signal; otherwise, a signal is sent to a machine learning and pattern library updating module;
and the machine learning and pattern library updating module starts a machine learning function after receiving a signal given by the drowning judging module, performs classified learning on the motion data characteristic value acquired by the data processing module, and updates the motion data characteristic value into the swimming stroke data pattern library of the database module.
9. The drowning detection system based on motion component threshold and machine learning of claim 8, further comprising an early warning indication module for receiving the early warning signal from the drowning determination module and emitting an audible and visual alarm message.
10. The motion component threshold and machine learning based drowning detection system of claim 8, further comprising a communication module for communicating data interaction with an external device.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911100919.7A CN110793539A (en) | 2019-11-12 | 2019-11-12 | Drowning detection method and system based on motion component threshold and machine learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911100919.7A CN110793539A (en) | 2019-11-12 | 2019-11-12 | Drowning detection method and system based on motion component threshold and machine learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110793539A true CN110793539A (en) | 2020-02-14 |
Family
ID=69444239
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911100919.7A Pending CN110793539A (en) | 2019-11-12 | 2019-11-12 | Drowning detection method and system based on motion component threshold and machine learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110793539A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112754473A (en) * | 2021-01-08 | 2021-05-07 | 华中科技大学鄂州工业技术研究院 | Spontaneous heating downloading and method for falling detection by fusing transfer learning algorithm |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20010017599A1 (en) * | 2000-02-24 | 2001-08-30 | U.S. Philips Corporation | GPS receiver and mobile unit incorporating the same |
CN102542301A (en) * | 2011-12-28 | 2012-07-04 | 浙江大学 | Early-stage drowning behavior detection method based on videos |
CN103295366A (en) * | 2013-04-18 | 2013-09-11 | 浙江大学城市学院 | Drowning alarming device based on body feeling and method thereof |
CN103413114A (en) * | 2013-05-17 | 2013-11-27 | 浙江大学 | Near-drowning behavior detection method based on support vector machine |
CN104517410A (en) * | 2015-01-21 | 2015-04-15 | 苏州大学 | Intelligent drowning-prevention system and method |
CN104537273A (en) * | 2015-01-21 | 2015-04-22 | 苏州大学 | Intelligent reasoning system and method for drowning mode |
CN105575043A (en) * | 2016-01-04 | 2016-05-11 | 广东小天才科技有限公司 | Reminding method and system for eliminating danger |
CN106251584A (en) * | 2016-10-14 | 2016-12-21 | 山西大学 | Multifunctional intellectual swimming bracelet and swimming state monitoring device and method |
CN106237604A (en) * | 2016-08-31 | 2016-12-21 | 歌尔股份有限公司 | Wearable device and the method utilizing its monitoring kinestate |
CN107566797A (en) * | 2017-09-07 | 2018-01-09 | 青岛博晶微电子科技有限公司 | A kind of drowned monitor and detection device of swimming pool |
CN107773931A (en) * | 2017-10-09 | 2018-03-09 | 共青城鑫航体育用品有限公司 | A kind of intelligent swimming mirror |
CN109347725A (en) * | 2018-10-31 | 2019-02-15 | 广东小天才科技有限公司 | Instant messaging method, intelligent device and storage medium |
CN110246306A (en) * | 2019-06-13 | 2019-09-17 | 西安电子科技大学 | Underwater exercise safety monitoring system design method based on wireless signal transmission |
-
2019
- 2019-11-12 CN CN201911100919.7A patent/CN110793539A/en active Pending
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20010017599A1 (en) * | 2000-02-24 | 2001-08-30 | U.S. Philips Corporation | GPS receiver and mobile unit incorporating the same |
CN102542301A (en) * | 2011-12-28 | 2012-07-04 | 浙江大学 | Early-stage drowning behavior detection method based on videos |
CN103295366A (en) * | 2013-04-18 | 2013-09-11 | 浙江大学城市学院 | Drowning alarming device based on body feeling and method thereof |
CN103413114A (en) * | 2013-05-17 | 2013-11-27 | 浙江大学 | Near-drowning behavior detection method based on support vector machine |
CN104517410A (en) * | 2015-01-21 | 2015-04-15 | 苏州大学 | Intelligent drowning-prevention system and method |
CN104537273A (en) * | 2015-01-21 | 2015-04-22 | 苏州大学 | Intelligent reasoning system and method for drowning mode |
CN105575043A (en) * | 2016-01-04 | 2016-05-11 | 广东小天才科技有限公司 | Reminding method and system for eliminating danger |
CN106237604A (en) * | 2016-08-31 | 2016-12-21 | 歌尔股份有限公司 | Wearable device and the method utilizing its monitoring kinestate |
CN106251584A (en) * | 2016-10-14 | 2016-12-21 | 山西大学 | Multifunctional intellectual swimming bracelet and swimming state monitoring device and method |
CN107566797A (en) * | 2017-09-07 | 2018-01-09 | 青岛博晶微电子科技有限公司 | A kind of drowned monitor and detection device of swimming pool |
CN107773931A (en) * | 2017-10-09 | 2018-03-09 | 共青城鑫航体育用品有限公司 | A kind of intelligent swimming mirror |
CN109347725A (en) * | 2018-10-31 | 2019-02-15 | 广东小天才科技有限公司 | Instant messaging method, intelligent device and storage medium |
CN110246306A (en) * | 2019-06-13 | 2019-09-17 | 西安电子科技大学 | Underwater exercise safety monitoring system design method based on wireless signal transmission |
Non-Patent Citations (1)
Title |
---|
王海涛等: "《卫星应用技术》", 31 March 2018, 北京理工大学出版社 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112754473A (en) * | 2021-01-08 | 2021-05-07 | 华中科技大学鄂州工业技术研究院 | Spontaneous heating downloading and method for falling detection by fusing transfer learning algorithm |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109643499B (en) | System and method for swimming analysis | |
CN109145696B (en) | Old people falling detection method and system based on deep learning | |
EP2850392B1 (en) | Method for step detection and gait direction estimation | |
US10486049B2 (en) | Device and method for monitoring swimming performance | |
Jensen et al. | Classification of kinematic swimming data with emphasis on resource consumption | |
CN104297519B (en) | Human motion gesture recognition method and mobile terminal | |
CN106981174A (en) | A kind of Falls Among Old People detection method based on smart mobile phone | |
CN113450539B (en) | Fall detection method | |
GB2529082A (en) | Device and method for monitoring swimming performance | |
CN111643092A (en) | Epilepsia alarm device and epilepsia detection method | |
CN110793539A (en) | Drowning detection method and system based on motion component threshold and machine learning | |
CN111435083A (en) | Pedestrian track calculation method, navigation method and device, handheld terminal and medium | |
CN116465412A (en) | Improved PDR indoor positioning method based on LSTM and attention mechanism | |
CN116092193A (en) | Pedestrian track reckoning method based on human motion state identification | |
CN106485889B (en) | Inmate based on intelligent spire lamella is detached from chain store method of real-time and system | |
CN112527118B (en) | Head posture recognition method based on dynamic time warping | |
CN115204240A (en) | Fall detection method and device based on millimeter wave radar and FPGA | |
CN114674317A (en) | Self-correcting dead reckoning system and method based on activity recognition and fusion filtering | |
CN114120252A (en) | Method and device for identifying state of automatic driving vehicle, electronic equipment and vehicle | |
Wang et al. | Human motion phase segmentation based on three new features | |
CN113095321A (en) | Roller bearing temperature measurement and fault early warning method and device for belt conveyor | |
Khan et al. | A novel hybrid fall detection technique using body part tracking and acceleration | |
Han et al. | Motion mode recognition in multi-storey buildings based on the naive Bayes method | |
CN116439692B (en) | Fall alarm detection method of wearable equipment | |
Gomes et al. | Detection and characterization of surfing events with smartphones’ embedded sensors |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200214 |