CN110400446A - A method of it is detected for swimming pool drowning - Google Patents
A method of it is detected for swimming pool drowning Download PDFInfo
- Publication number
- CN110400446A CN110400446A CN201910711521.0A CN201910711521A CN110400446A CN 110400446 A CN110400446 A CN 110400446A CN 201910711521 A CN201910711521 A CN 201910711521A CN 110400446 A CN110400446 A CN 110400446A
- Authority
- CN
- China
- Prior art keywords
- value
- change rate
- oxygen saturation
- sliding window
- maximum absolute
- 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.)
- Granted
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
- A61B5/02438—Detecting, measuring or recording pulse rate or heart rate with portable devices, e.g. worn by the patient
-
- 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
- G08B21/08—Alarms for ensuring the safety of persons responsive to the presence of persons in a body of water, e.g. a swimming pool; responsive to an abnormal condition of a body of water
- G08B21/088—Alarms for ensuring the safety of persons responsive to the presence of persons in a body of water, e.g. a swimming pool; responsive to an abnormal condition of a body of water by monitoring a device worn by the person, e.g. a bracelet attached to the swimmer
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Cardiology (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Animal Behavior & Ethology (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Physiology (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- General Physics & Mathematics (AREA)
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
The present invention provides a kind of methods for swimming pool drowning detection, including carry out data prediction, characteristic vector pickup, feature vector pretreatment, decision device calculating, result treatment to acquisition data;Processing analysis is carried out by the physiological parameter to human body, completes the automatic detection judgement of swimming pool drowning situation;Pass through the extraction of feature vector and the use of One Class SVM model, comprehensive reasonable utilizes human body items physiological parameter, and when so that normally swimming situation false detection rate being 2%, drowned recall rate is up to 99.9%, to fully ensure that the accurate detection of drowned situation in the lower situation of erroneous detection;This method realizes continuous, the real-time detection on time shaft using the methods of time shaft sliding window fragment, less to resource consumption, conducive to mobile terminal is deployed to.
Description
Technical field
The present invention relates to information analyses, and in particular to a method of it is detected for swimming pool drowning.
Background technique
With the construction of community in urban areas swimming pool.Swimming pool drown causing death the phenomenon that also gradually highlighted.Currently, it swims
Swimming pool strengthens rescue measure one after another, such as increases lifeguard, sets up " anti-diving ", " water depth " warning sign, but this brings drop again
Low swimmer's freedom degree, recreational drawback;The wearable products such as numerous bracelets current simultaneously can be to every physiology of human body
Parameter is accurately measured, but currently shortage one is rationally, effectively, timely method is carried out using the physiological parameter measured
Drowned judgement.
Summary of the invention
It is an object of that present invention to provide a kind of methods for swimming pool drowning detection, divide human body physiological parameter
Analysis, to it is drowned the occurrence of carry out in time, accurate judgement;
To achieve the above object, the present invention provides it is a kind of for swimming pool drowning detection method, to acquisition data into
Line number Data preprocess, characteristic vector pickup, eigen vector pretreatment, decision device calculating, result treatment;
The acquisition data include age, gender, rhythm of the heart data, oxygen saturation value;
The data prediction includes gender digitlization, time shaft sliding window fragment, rhythm of the heart data calculate and oxygen saturation
Angle value calculates;
The gender is digitized as, and indicates male using 1, indicates female using -1;
The time shaft sliding window fragment, on a timeline, Fixed Time Interval data are as a time window, i.e. conduct
One group of calculating data, time window move on a timeline, and interception multiple groups calculate data;
The rhythm of the heart data calculate, calculate minimum value of the rhythm of the heart data in a time shaft sliding window fragment,
The direction of mean value, maximum value and change rate of heartbeat maximum absolute value value and change rate of heartbeat maximum absolute value value;
The change rate of heartbeat is heart rate in unit time variable quantity, and it is positive value that heart rate, which increases, and heart rate is reduced to negative value;
The change rate of heartbeat maximum absolute value value is change rate of heartbeat maximum absolute value value in a time shaft sliding window;
The direction of the change rate of heartbeat maximum absolute value value is the change rate of heartbeat of change rate of heartbeat maximum absolute value value
It is positive and negative, be timing, direction 1, when being negative, direction be -1;
The oxygen saturation value calculates, calculate minimum value of the oxygen saturation value in a time shaft sliding window fragment, mean value,
The direction of maximum value and oxygen saturation value change rate maximum absolute value value and oxygen saturation value change rate maximum absolute value value;
The oxygen saturation change rate is oxygen saturation value in unit time variable quantity, and oxygen saturation value is in the unit time
Increasing is positive value, is reduced to negative value;
The oxygen saturation value change rate maximum absolute value value is in a time shaft sliding window, and oxygen saturation value change rate is exhausted
To value maximum value;
The direction of the oxygen saturation value change rate maximum absolute value value is oxygen saturation value change rate maximum absolute value value
Oxygen saturation value change rate it is positive and negative, be timing, direction 1, when being negative, direction be -1;
Described eigenvector is extracted, by the age in a time shaft sliding window, gender, time shaft sliding window time interval, the heart
Rate monitoring data calculate and oxygen saturation checkout result composition;
Described eigenvector composition form are as follows:
V={ a, g, Δ t, hmin,hmean,hmax,|h'|max,dh,omin,omean,omax,|o'|max,do}
In formula, a is the age, g is gender, Δ t is time shaft sliding window time interval, hmin、hmean、hmaxRespectively time shaft
The minimum value of heart rate monitoring data, mean value and maximum value in sliding window fragment;|h'|maxFor change rate of heartbeat maximum absolute value value, dh
For the direction of change rate of heartbeat maximum absolute value value;omin、omean、omaxOxygen saturation value respectively in time shaft sliding window fragment
Minimum value, mean value and maximum value;|o'|maxFor oxygen saturation value interconversion rate maximum absolute value value, doFor oxygen saturation value change rate
The direction of maximum absolute value value;
Described eigenvector pretreatment, each dimension of feature vector normalizes respectively, as feature vector subtract mean value to
Amount, point remove variance vectors, obtain normalization characteristic vector, formula is as follows:
In formula, M is mean vector,For variance vectors;
The mean vector and variance vectors are calculated by each dimension element of the feature vector of all samples, are fixed
Value;
The decision device calculates, and pretreated normalization characteristic vector is sent into swimming pool drowning detection model, output
Operation result;
The swimming pool drowning detection model is One Class SVM model, gaussian kernel function is selected, by having trained under line
At;
The operation result is that input normalization characteristic vector is normal (+1) or exception (- 1);
The result treatment, according to decision device calculated result, output whether be it is drowned,
When decision device output is normal (- 1), the detection for carrying out a time shaft sliding window fragment determines;
When decision device output is abnormal (- 1), stop detection, output is judged to drowning.
Further, in described eigenvector pretreatment, for mean vector and variance vectors,
m0=m1=m2=m7=m12=0
δ0=δ1=δ2=δ7=δ12=1
In formula, the mean value and variance result correspond respectively to age, gender, time shaft sliding window time interval, heart rate change
The direction of rate maximum absolute value value and the direction of oxygen saturation value change rate maximum absolute value value;
The invention has the following beneficial effects:
The present invention provides a kind of methods for swimming pool drowning detection, are handled by the physiological parameter to human body
The automatic detection judgement of swimming pool drowning situation is completed in analysis;Extraction and One Class SVM model by feature vector
It uses, comprehensive reasonable utilizes human body items physiological parameter, and when so that normally swimming situation false detection rate being 2%, recall rate of drowning is high
Up to 99.9%, to fully ensure that the accurate detection of drowned situation in the lower situation of erroneous detection;This method uses time shaft sliding window
The methods of fragment realizes continuous, real-time detection on time shaft, less to resource consumption, conducive to mobile terminal is deployed to.
Detailed description of the invention
The method flow diagram that Fig. 1, a kind of swimming pool drowning detect.
Fig. 2, time shaft sliding window fragment schematic diagram.
Specific embodiment
The present invention is specifically described below by embodiment, it is necessary to which indicated herein is that following embodiment is only used
In invention is further explained, it should not be understood as limiting the scope of the invention, person skilled in art can
To make some nonessential modifications and adaptations to the present invention according to aforementioned present invention content.
Embodiment 1
A method of for swimming pool drowning detect, as shown in Figure 1, to acquisition data and carry out data prediction,
Characteristic vector pickup, feature vector pretreatment, decision device calculating, result treatment;
Wherein, acquisition data include age, gender, rhythm of the heart data, oxygen saturation value;
Data prediction completes gender digitlization, the calculating of time shaft sliding window fragment, heart rate monitoring data and oxygen saturation value
It calculates;
Wherein, gender is digitized as, and indicates male using 1, indicates female using -1;And time shaft sliding window fragment as shown in Fig. 2,
For on a timeline, Fixed Time Interval data are as a time window, and as one group of calculating data, time window is in the time
It is moved on axis, interception multiple groups calculate data;A length of Δ t when time window, a length of stepping when mobile of adjacent time axis sliding window fragment
ts;
Rhythm of the heart data are calculated as calculating minimum value of the rhythm of the heart data in a time shaft sliding window fragment,
The method of value, maximum value and change rate of heartbeat maximum absolute value value and change rate of heartbeat maximum absolute value value;
Change rate of heartbeat is variable quantity of the heart rate in the unit time, and it is positive value that heart rate, which increases, and heart rate is reduced to negative value;Into one
The change rate of heartbeat maximum absolute value value of step is change rate of heartbeat maximum absolute value value in a time shaft sliding window;And heart rate becomes
The direction of rate maximum absolute value value is the positive and negative of the change rate of heartbeat of change rate of heartbeat maximum absolute value value, is timing, direction
It is 1, when being negative, direction is -1;
Oxygen saturation value calculates, and calculates minimum value, mean value, maximum of the oxygen saturation value in a time shaft sliding window fragment
The direction of value and oxygen saturation value change rate maximum absolute value value and oxygen saturation value change rate maximum absolute value value;
Oxygen saturation change rate is oxygen saturation value in unit time variable quantity, and oxygen saturation value increases in the unit time
For positive value, it is reduced to negative value;
Oxygen saturation value change rate maximum absolute value value is oxygen saturation value change rate absolute value in a time shaft sliding window
Maximum value;
The direction of oxygen saturation value change rate maximum absolute value value is the oxygen of oxygen saturation value change rate maximum absolute value value
Intensity value change rate it is positive and negative, be timing, direction 1, when being negative, direction be -1;
Characteristic vector pickup is supervised by the age in a time shaft sliding window, gender, time shaft sliding window time interval, heart rate
Measured data calculates and oxygen saturation checkout result composition;Its feature vector composition form are as follows:
V={ a, g, Δ t, hmin,hmean,hmax,|h'|max,dh,omin,omean,omax,|o'|max,do}
In formula, a is the age, g is gender, Δ t is time shaft sliding window time interval, hmin、hmean、hmaxRespectively time shaft
The minimum value of heart rate monitoring data, mean value and maximum value in sliding window fragment;|h'|maxFor change rate of heartbeat maximum absolute value value, dh
For the direction of change rate of heartbeat maximum absolute value value;omin、omean、omaxOxygen saturation value respectively in time shaft sliding window fragment
Minimum value, mean value and maximum value;|o'|maxFor oxygen saturation value interconversion rate maximum absolute value value, doFor oxygen saturation value change rate
The direction of maximum absolute value value;
For feature vector preprocessing process, need that each dimension of feature vector is normalized respectively, i.e., feature to
Amount subtracts mean vector, and point removes variance vectors, obtains normalization characteristic vector, such as following formula:
In formula, M is mean vector,For variance vectors;Mean vector and variance vectors by all samples feature vector
Each dimension element be calculated, it is related with training sample distribution, be in use definite value;
Consider the age, the quantized value of gender, time shaft sliding window time interval, change rate of heartbeat maximum absolute value value direction
And the direction of oxygen saturation value change rate maximum absolute value value is not suitable for normalization quantization, so fixing its corresponding mean value and side
Difference is set as follows:
m0=m1=m2=m7=m12=0
δ0=δ1=δ2=δ7=δ12=1
Decision device calculates, and pretreated normalization characteristic vector is sent into swimming pool drowning detection model, exports operation
As a result;Swimming pool drowning detection model is One Class SVM model, selects gaussian kernel function, is completed by training under line;Operation
It as a result is normal (+1) or exception (- 1) for input normalization characteristic vector;
Final result processing, according to decision device calculated result, whether output is drowned, concrete operations are as follows:
When decision device output is normal (- 1), the detection for carrying out a time shaft sliding window fragment determines;
When decision device output is abnormal (- 1), stop detection, output is judged to drowning.
Embodiment 2:
Training under a kind of line of the drowned detection model of method middle reaches swimming pool for swimming pool drowning detection;Main use is received
Collection, the mode for arranging, marking training set are trained disaggregated model, are desirably to obtain the swimming pool drowning detection mould of better performances
Type;
Wherein positive sample is the normalization characteristic vector that acquisition data are extracted when swimmer's swimming pool is normally swum;Negative sample
It swims for swimmer and acquires the normalization characteristic vector that data are extracted when trip pond has drowned situation;Positive sample is labeled as 1, negative sample
It is labeled as -1;
During training set data is compiled, positive sample data are more, and are easy to obtain, and negative sample data are based on
The idiopathic and risk of the drowned situation in swimming trip pond, acquisition is more difficult, and it is less to obtain sample data;There are positive sample numbers for training set
The case where amount is much larger than negative sample quantity;
For the particularity of the drowned detection model above situation in training set data in swimming trip pond, model is selected as One
Class SVM model, select gaussian kernel function, training its positive sample distribution, thus make model have judge input feature vector to
Amount whether be positive sample ability;
In the training process, it is first determined training set;According to the feature vector of samples all in training set
V={ a, g, Δ t, hmin,hmean,hmax,|h'|max,dh,omin,omean,omax,|o'|max,do}
Feature vector is calculated separately per one-dimensional mean value and variance, obtains mean vector M and variance vectors
In formula,For the i-th dimension element of the feature vector of j-th of sample in training set;
Consider the particularity of specific dimension in feature vector, setting:
m0=m1=m2=m7=m12=0
δ0=δ1=δ2=δ7=δ12=1
Feature vector is normalized again, so that more rapidly and being easy to restrain, obtaining property during model training
It can preferable model;
About the setting method of negative sample proportion during One Class SVM model training,
It is r that statistics, which calculates negative sample proportion in training set,-1;The setting of negative sample proportion is such as in training process
Under:
Wherein in fixed training set, the setting of negative sample proportion is bigger, and negative sample is judged as positive sample
Probability is smaller, and the setting of negative sample proportion is excessive, and positive sample is caused to be determined as negative sample by maximum probability;So comprehensively considering
The practicability and functionality of the drowned detection model in swimming trip pond, and harm severity when erroneous judgement generation, as negative sample miss
When being judged to positive sample generation, the occurrence of drowned cannot be found to endanger drowning person's life in time;It is arranged shared by negative sample
Proportionality coefficient increases by 0.3 than true value, is mistaken for that negative sample is limited increasing positive sample, cuts the not high feelings of the erroneous judgement extent of injury
Condition, reduces the probability that negative sample is mistaken for positive sample, reduces the possibility that drowning person is not had found in time to greatest extent.
Claims (2)
1. a kind of method for swimming pool drowning detection, it is characterised in that: including carrying out data prediction, spy to acquisition data
Levy vector extraction, feature vector pretreatment, decision device calculating, result treatment;
The acquisition data include age, gender, rhythm of the heart data, oxygen saturation value;
The data prediction includes gender digitlization, time shaft sliding window fragment, rhythm of the heart data calculate and oxygen saturation value
It calculates;
The gender is digitized as, and indicates male using 1, indicates female using -1;
The time shaft sliding window fragment, on a timeline, Fixed Time Interval data are used as one group as a time window
Data are calculated, time window moves on a timeline, and interception multiple groups calculate data;
The rhythm of the heart data calculate, calculate minimum value of the rhythm of the heart data in a time shaft sliding window fragment, mean value,
The direction of maximum value and change rate of heartbeat maximum absolute value value and change rate of heartbeat maximum absolute value value;
The change rate of heartbeat is heart rate in unit time variable quantity, and it is positive value that heart rate, which increases, and heart rate is reduced to negative value;
The change rate of heartbeat maximum absolute value value is change rate of heartbeat maximum absolute value value in a time shaft sliding window;
The direction of the change rate of heartbeat maximum absolute value value be change rate of heartbeat maximum absolute value value change rate of heartbeat just
It is negative, it is timing, direction 1, when being negative, direction is -1;
The oxygen saturation value calculates, and calculates minimum value, mean value, maximum of the oxygen saturation value in a time shaft sliding window fragment
The direction of value and oxygen saturation value change rate maximum absolute value value and oxygen saturation value change rate maximum absolute value value;
The oxygen saturation change rate is oxygen saturation value in unit time variable quantity, and oxygen saturation value increases in the unit time
For positive value, it is reduced to negative value;
The oxygen saturation value change rate maximum absolute value value is oxygen saturation value change rate absolute value in a time shaft sliding window
Maximum value;
The direction of the oxygen saturation value change rate maximum absolute value value is the oxygen of oxygen saturation value change rate maximum absolute value value
Intensity value change rate it is positive and negative, be timing, direction 1, when being negative, direction be -1;
Described eigenvector is extracted, and is supervised by the age in a time shaft sliding window, gender, time shaft sliding window time interval, heart rate
Measured data calculates and oxygen saturation checkout result composition;
Described eigenvector composition form are as follows:
V={ a, g, Δ t, hmin,hmean,hmax,|h'|max,dh,omin,omean,omax,|o'|max,do}
In formula, a is the age, g is gender, Δ t is time shaft sliding window time interval, hmin、hmean、hmaxRespectively time shaft sliding window
The minimum value of heart rate monitoring data, mean value and maximum value in fragment;|h'|maxFor change rate of heartbeat maximum absolute value value, dhFor the heart
The direction of rate change rate maximum absolute value value;omin、omean、omaxThe minimum of oxygen saturation value respectively in time shaft sliding window fragment
Value, mean value and maximum value;|o'|maxFor oxygen saturation value interconversion rate maximum absolute value value, doIt is absolute for oxygen saturation value change rate
It is worth the direction of maximum value;
Described eigenvector pretreatment, each dimension of feature vector normalize respectively, and as feature vector subtracts mean vector, point
Except variance vectors, normalization characteristic vector is obtained, formula is as follows:
In formula, M is mean vector,For variance vectors;
The mean vector and variance vectors are calculated by each dimension element of the feature vector of all samples, are definite value;
The decision device calculates, and pretreated normalization characteristic vector is sent into swimming pool drowning detection model, exports operation
As a result;
The swimming pool drowning detection model is One Class SVM model, selects gaussian kernel function, is completed by training under line;
The operation result is that input normalization characteristic vector can be normal (+1) or exception (- 1);
The result treatment, according to decision device calculated result, output whether be it is drowned,
When decision device output is normal (- 1), the detection for carrying out a time shaft sliding window fragment determines;
When decision device output is abnormal (- 1), stop detection, output is judged to drowning.
2. a kind of method for swimming pool drowning detection as described in claim 1, it is characterised in that: described eigenvector is located in advance
In reason, for mean vector and variance vectors,
m0=m1=m2=m7=m12=0
δ0=δ1=δ2=δ7=δ12=1
In formula, the mean value and variance result correspond respectively to age, gender, time shaft sliding window time interval, change rate of heartbeat
The direction of maximum absolute value value and the direction of oxygen saturation value change rate maximum absolute value value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910711521.0A CN110400446B (en) | 2019-08-02 | 2019-08-02 | Method for detecting drowning of swimming pool |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910711521.0A CN110400446B (en) | 2019-08-02 | 2019-08-02 | Method for detecting drowning of swimming pool |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110400446A true CN110400446A (en) | 2019-11-01 |
CN110400446B CN110400446B (en) | 2021-03-26 |
Family
ID=68327176
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910711521.0A Active CN110400446B (en) | 2019-08-02 | 2019-08-02 | Method for detecting drowning of swimming pool |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110400446B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111833567A (en) * | 2020-05-30 | 2020-10-27 | 北京方德泳安体育文化有限公司 | Motion detection system based on UWB and nine-axis gyroscope |
CN114896568A (en) * | 2022-07-13 | 2022-08-12 | 杭州光粒科技有限公司 | Swimming data statistical method, device, equipment and medium |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103006223A (en) * | 2012-12-13 | 2013-04-03 | 中国人民解放军第四军医大学 | Household non-contact sleeping monitoring device and method |
CN103584840A (en) * | 2013-11-25 | 2014-02-19 | 天津大学 | Automatic sleep stage method based on electroencephalogram, heart rate variability and coherence between electroencephalogram and heart rate variability |
CN105678963A (en) * | 2016-03-15 | 2016-06-15 | 陕西师范大学 | Intelligent anti-drowning device |
CN105877766A (en) * | 2016-06-21 | 2016-08-24 | 东北大学 | Mental state detection system and method based on multiple physiological signal fusion |
US20160314675A1 (en) * | 2015-04-23 | 2016-10-27 | Asma Saleh HAWSAH | Swimming aid to prevent drowning |
CN108010273A (en) * | 2017-12-29 | 2018-05-08 | 辽宁工程技术大学 | A kind of drowned detection alarm based on Cloud Server and machine learning algorithm |
CN108113657A (en) * | 2016-11-30 | 2018-06-05 | 天津职业技术师范大学 | A kind of Driver Fatigue Detection based on D-S evidence theory |
CN108376456A (en) * | 2018-02-05 | 2018-08-07 | 广东小天才科技有限公司 | A kind of drowned monitoring method and wearable device |
CN108446733A (en) * | 2018-03-19 | 2018-08-24 | 上海玺翎智能科技有限公司 | A kind of human body behavior monitoring and intelligent identification Method based on multi-sensor data |
-
2019
- 2019-08-02 CN CN201910711521.0A patent/CN110400446B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103006223A (en) * | 2012-12-13 | 2013-04-03 | 中国人民解放军第四军医大学 | Household non-contact sleeping monitoring device and method |
CN103584840A (en) * | 2013-11-25 | 2014-02-19 | 天津大学 | Automatic sleep stage method based on electroencephalogram, heart rate variability and coherence between electroencephalogram and heart rate variability |
US20160314675A1 (en) * | 2015-04-23 | 2016-10-27 | Asma Saleh HAWSAH | Swimming aid to prevent drowning |
CN105678963A (en) * | 2016-03-15 | 2016-06-15 | 陕西师范大学 | Intelligent anti-drowning device |
CN105877766A (en) * | 2016-06-21 | 2016-08-24 | 东北大学 | Mental state detection system and method based on multiple physiological signal fusion |
CN108113657A (en) * | 2016-11-30 | 2018-06-05 | 天津职业技术师范大学 | A kind of Driver Fatigue Detection based on D-S evidence theory |
CN108010273A (en) * | 2017-12-29 | 2018-05-08 | 辽宁工程技术大学 | A kind of drowned detection alarm based on Cloud Server and machine learning algorithm |
CN108376456A (en) * | 2018-02-05 | 2018-08-07 | 广东小天才科技有限公司 | A kind of drowned monitoring method and wearable device |
CN108446733A (en) * | 2018-03-19 | 2018-08-24 | 上海玺翎智能科技有限公司 | A kind of human body behavior monitoring and intelligent identification Method based on multi-sensor data |
Non-Patent Citations (3)
Title |
---|
冯靖杰: "穿戴式核心体温和连续血压测量关键技术的研究", 《中国博士学位论文全文数据库 医药卫生科技辑》 * |
华晶: "可穿戴式系统心电信号压缩感知与心律失常分类研究", 《中国博士学位论文全文数据库 医药卫生科技辑》 * |
颜红博: "基于多生理信息融合的情绪识别方法", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111833567A (en) * | 2020-05-30 | 2020-10-27 | 北京方德泳安体育文化有限公司 | Motion detection system based on UWB and nine-axis gyroscope |
CN111833567B (en) * | 2020-05-30 | 2022-02-18 | 北京方德泳安体育文化有限公司 | Motion detection system based on UWB and nine-axis gyroscope |
CN114896568A (en) * | 2022-07-13 | 2022-08-12 | 杭州光粒科技有限公司 | Swimming data statistical method, device, equipment and medium |
Also Published As
Publication number | Publication date |
---|---|
CN110400446B (en) | 2021-03-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Patil et al. | Machine learning approach in melanoma cancer stage detection | |
CN110969626B (en) | Method for extracting hippocampus of human brain nuclear magnetic resonance image based on 3D neural network | |
US9336486B2 (en) | Attribute value estimation device, attribute value estimation method, program, and recording medium | |
Song et al. | Simultaneous cell detection and classification in bone marrow histology images | |
Schmid et al. | The LOKI underwater imaging system and an automatic identification model for the detection of zooplankton taxa in the Arctic Ocean | |
CN108446733A (en) | A kind of human body behavior monitoring and intelligent identification Method based on multi-sensor data | |
CN102302370A (en) | Method and device for detecting tumbling | |
CN110400446A (en) | A method of it is detected for swimming pool drowning | |
CN109902558A (en) | A kind of human health deep learning prediction technique based on CNN-LSTM | |
CN102163283A (en) | Method for extracting face characteristic based on local three-value mode | |
CN102938054A (en) | Method for recognizing compressed-domain sensitive images based on visual attention models | |
CN109977930A (en) | Method for detecting fatigue driving and device | |
CN111681209A (en) | Intelligent detection method and system for cleavage state of blastomere | |
CN108717548B (en) | Behavior recognition model updating method and system for dynamic increase of sensors | |
CN109567832A (en) | A kind of method and system of the angry driving condition of detection based on Intelligent bracelet | |
CN111862160B (en) | Target tracking method, medium and system based on ARM platform | |
Prabhakar et al. | A comprehensive analysis of alcoholic EEG signals with detrend fluctuation analysis and post classifiers | |
Li et al. | Spike detection and spike sorting with a hidden Markov model improves offline decoding of motor cortical recordings | |
CN112668362B (en) | Human evidence comparison model training method for dynamic optimization class proxy | |
CN109086704A (en) | A kind of physical activity recognition methods classified based on rarefaction representation and Softmax | |
McKenna et al. | Cascade-correlation neural networks for the classification of cervical cells | |
Kumari et al. | NeuroAid: Emotion-Based EEG Analysis for Parkinson's Disease Identification | |
CN117814163B (en) | Method for distinguishing male and female of silver drum fish based on morphological analysis | |
Mittal et al. | Prediction of Alopecia Areata using CNN | |
Zambales et al. | A model for classification of skin lesions using image processing techniques and neural network |
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 |