CN108683724A - A kind of intelligence children's safety and gait health monitoring system - Google Patents
A kind of intelligence children's safety and gait health monitoring system Download PDFInfo
- Publication number
- CN108683724A CN108683724A CN201810447694.1A CN201810447694A CN108683724A CN 108683724 A CN108683724 A CN 108683724A CN 201810447694 A CN201810447694 A CN 201810447694A CN 108683724 A CN108683724 A CN 108683724A
- Authority
- CN
- China
- Prior art keywords
- children
- gait
- safety
- supervision
- intelligent
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/112—Gait analysis
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6802—Sensor mounted on worn items
- A61B5/6804—Garments; Clothes
- A61B5/6807—Footwear
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G08B21/0202—Child monitoring systems using a transmitter-receiver system carried by the parent and the child
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2503/00—Evaluating a particular growth phase or type of persons or animals
- A61B2503/06—Children, e.g. for attention deficit diagnosis
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Theoretical Computer Science (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Surgery (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Computing Systems (AREA)
- Pathology (AREA)
- Animal Behavior & Ethology (AREA)
- Heart & Thoracic Surgery (AREA)
- Signal Processing (AREA)
- Physiology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Networks & Wireless Communication (AREA)
- Psychiatry (AREA)
- Child & Adolescent Psychology (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Dentistry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a kind of intelligent children's safety and gait health monitoring system and its monitoring method, which includes the supervision shoe worn for children and the intelligent terminal that guardian uses.Pressure sensitive, positioning system, gyroscope, voice input and humidity sensing device built in the supervision shoe, system also includes high in the clouds analysis process system, high in the clouds analysis process system carries out analyzing processing to the data that supervision shoe built-in acquires, and is then passed to the intelligent terminal that guardian uses.Intelligent terminal includes smartwatch, mobile terminal (mobile phone), home intelligent alarm speaker.The data of the supervision shoe built-in acquisition of the present invention may be implemented fence, children's gait analysis, the analysis of leg skeleton development, the judgement of danger sound, tumble, be taken in one's arms and overboard monitoring function.System is fallen, is taken in one's arms and send out alarm to guardian by vibrations or alarm tone when overboard state monitoring that children are in.
Description
Technical field
A kind of signal device technical field that the invention belongs to be monitored to abnormal conditions, and in particular to intelligence children peace
Complete and gait health monitoring system and its monitoring method.
Background technology
With the development of wireless communication technique and technology of Internet of things, the application of security monitoring is more and more extensive.Especially needle
There are many products to emerge the security monitoring of specific group such as children and old man.Existing this kind of product mainly uses
The intelligent mode location technology of GPS/LBS, i.e. satellite positioning and base station, guardian can be mobile eventually by online or mobile phone etc.
The real time information of children is checked at end, to determine whether in the region of safety；Once children under guardianship leave specified safety zone
Monitoring device will send information to guardian, or even can also include some emergency reliefs, such as dangerous or abnormal
When generation, distress call key can be pressed and dial preset cell-phone number.
In existing patent document, Patent No. CN201520711591.3, entitled " a kind of children old man intelligently succours
The utility model of body-worn equipment " discloses a kind of body-worn equipment that children old man intelligently succours comprising is worn on
Wearing gadget with old man or children, have receive, send, the high in the clouds of management function and mobile terminal, wherein wearing gadget
Include level-one acquisition sensor, two level acquisition sensor and central controller, level-one acquisition inductor includes gravitational equilibrium sensing
Device or sensor noise, level-one acquisition inductor are connect by analog-to-digital conversion module with central controller, and central controller is felt with two level acquisition
Device is answered to be bi-directionally connected, the central controller is also connect with high in the clouds, mobile terminal successively, and central controller includes localization tracker.The reality
With the novel abnormal movement or ambient enviroment that can be directed to children and old man, the environment or body being in old man and children are triggered
Body is judged, if it is in the hole, notify parent or guardian.But the shortcomings that body-worn equipment is to lay particular emphasis on pair
The protection of the elderly such as can acquire body temperature, blood pressure and fall down at the parameters, taken in one's arms for the running of children, jump, by stranger,
The danger for being common in children such as drowned can not carry out effective monitoring.Moreover, monitor device disclosed in the above-mentioned prior art all lacks
Weary learning functionality can not accomplish the improvement that system is realized based on existing data.
In machine learning, convolutional neural networks are a kind of depth feed forward-fuzzy controls, have been applied successfully to figure
As identification.Convolutional neural networks are a kind of feedforward neural networks, and artificial neuron can respond surrounding cells, can carry out large size
Image procossing.Convolutional neural networks include convolutional layer and pond layer.Convolutional neural networks are persistently had an effect in multiple directions in recent years,
Have in terms of speech recognition, recognition of face, generic object identification, motion analysis, natural language processing even brain wave analysis
It breaks through.Wherein, LSTM (the Long Short-Term Memory) one kind of algorithm as deep learning method, is a kind of specific shape
The RNN (Recurrent neural network, Recognition with Recurrent Neural Network) of formula, belongs to time recurrent neural network, is suitable for locating
It is spaced and postpones relatively long critical event in reason and predicted time sequence, LSTM has a variety of answer in numerous areas
With, such as language translation, control robot, image analysis, documentation summary, speech recognition image recognition, handwriting recognition, control chat
The tasks such as its robot, predictive disease, clicking rate and stock, composite music.
Tracking and judgement by convolutional neural networks and LSTM algorithms for the health status and developmental condition of children, at present
There are no disclosed disclosures.
Invention content
The present invention for the above-mentioned prior art the problem of propose a kind of intelligent children's safety and gait health supervision
System and its monitoring method, can solve problems with, the threat or different that can not including but not limited to differentiate for certain children
Reason condition or children will not be autonomous calling for help, guardian cannot know dangerous generation at the first time to take measures,
And in the case of moving outdoors, the children's situation of attention transfer and mobile phone in packet when there is parent to accompany and attend to, for
Certain potential threats from ambient outside, such as malice movement that stranger takes in one's arms, fall down, drown, can not be at the first time
It makes a response and protects in time.
In order to achieve the above objectives, technical solution proposed by the present invention is a kind of intelligent children's safety and gait health supervision system
System includes the supervision shoe worn for children and the intelligent terminal that guardian uses.It is pressure sensitive built in the supervision shoe, fixed
Position system, gyroscope, voice input and humidity sensing device, system also include high in the clouds analysis process system, high in the clouds analyzing processing
System carries out analyzing processing to the data that supervision shoe built-in acquires, and is then passed to the intelligent terminal that guardian uses and sets
It is standby.
Preferably, above-mentioned intelligent terminal includes smartwatch, mobile terminal (mobile phone), home intelligent alarm sound
Case.
Further, fence, children's gait analysis, leg may be implemented in the data of above-mentioned supervision shoe built-in acquisition
Skeleton development analysis, tumble, is taken in one's arms and overboard monitoring function the judgement of danger sound.
Preferably, system is fallen, is taken in one's arms and when overboard state by vibrations or alarm tone monitoring that children are in
Alarm is sent out to guardian.
The present invention is it is further proposed that a kind of carrying out children using above-mentioned intelligent children's safety and gait health monitoring system
The method of gait health supervision, specifically includes following steps：
(1) intelligent children's safety and gait health monitoring system acquire children to sole by the built-in of supervision shoe
Pressure spectrum data；
(2) data of above-mentioned acquisition are uploaded to high in the clouds analysis process system；
(3) high in the clouds analysis process system uses the above-mentioned pressure spectrum data of depth convolutional neural networks analyzing processing, and ties
It closes the children's gait health data trained early period to compare, judges children's gait health status；
(4) high in the clouds analysis process system utilizes the LSTM deep learning networks based on time series, to different time state
Foot pressure analyzed, whether child development state is well judged；
(5) user is allowed to give a mark above-mentioned child development status information feedback to user；
(6) convolutional neural networks and LSTM deep learning networks are finely adjusted according to field feedback, to obtain
More accurate analysis result；
(7) the continuous iteration above process, until result is stablized, process terminates.
Further, preferably, health status described in above-mentioned steps 3 comprises determining whether that X-type, O-shaped leg, sole are exerted oneself
Situation.
Preferably, child development state described in above-mentioned steps 4 includes weight, posture of walking.
Compared with prior art, the beneficial effects of the present invention are：
1, the present invention is monitored in real time by foot force, sound to children wearer, humidity, the parameters such as position,
And the data monitored are analyzed and judged by high in the clouds analysis process system, generate fence, electron trajectory, pressure
Collection of illustrative plates, so judge children be in static, walk thrust kick state in line state, running state, jump state, profundal zone water,
Tumble state, taken in one's arms state, shoes are taken off, the states such as lose, and when child state exception occur, in time to monitoring human hair
Alarm is sent, is succoured with timely implementation.
2, high in the clouds analysis process system acquires pressure spectrum data using depth convolutional neural networks analyzing processing, and combines
The children's gait health data trained early period compares, and judges children's gait health status, and using based on the time
The LSTM deep learning networks of sequence, can analyze the foot pressure of different time state, be to child development state
It is no well to be judged, the advantage of big data can be given full play in this way, improve the accuracy of judgement.
3, by being finely adjusted to convolutional neural networks and LSTM deep learning networks according to field feedback, can obtain
Go out more accurate analysis result.
Description of the drawings
Fig. 1 is the system construction drawing of intelligent children's safety and gait health monitoring system.
Fig. 2 is the method and step that children's gait health supervision is carried out using intelligent children's safety and gait health monitoring system
Figure.
Fig. 3 is VGGNet network structures and improved Structure Comparison figure.
Fig. 4 is ResNet residual error network element schematic diagrames.
Fig. 5 is the detailed structure view of Block in feature extraction network.
Specific implementation mode
The specific implementation mode of the present invention is further described in conjunction with attached drawing.
Intelligence children's safety of the present invention and the structure of gait health monitoring system are as shown in Figure 1, the monitoring worn for children
Multiple sensors built in shoe lining face, including pressure sensitive, positioning system, gyroscope, voice input and humidity sensing device, these
Sensor reasonable layout in shoe body avoids excessively centralized arrangement as possible, to prevent from interfering with each other.Pressure sensitive sensor can be adopted
With ultra-thin ultra-soft diaphragm pressure sensor, such as RFP-603-10kg thin-film pressure sensors, movement can with alignment sensor
Using IMU- Inertial Measurement Unit BMI055, three axis MEMS gyro sensor can be used in gyroscope.Voice input pickup can
To use high sensitivity microphone sensor module or highly sensitive sound control plate module, as long as humidity sensor satisfaction ± 5%RH
Precision.
The intelligent terminal that guardian uses includes that the various intelligence wearings that can carry computer applied algorithm app are set
It is standby, such as smartwatch, mobile terminal (mobile phone), home intelligent alarm speaker etc..
Fence, children's gait analysis, leg bone may be implemented based on the data that supervision shoe built-in acquires in system
Bone developmental analysis, tumble, is taken in one's arms and overboard monitoring function the judgement of danger sound.
The higher state of danger coefficient takes in one's arms (abducted including falling (being easy to happen fracture), by stranger for children
Sell possibility) and it is overboard (drowned), therefore system is monitoring that children are in a period of time of above-mentioned state and need through vibrations or alert
Report sound sends out alarm to guardian.
The sensor gathered data passes through the analysis process system of 4G network transmissions to high in the clouds, high in the clouds analysis process system
Analyzing processing is carried out to the data of supervision shoe built-in acquisition, is then passed to the intelligent terminal that guardian uses.Cloud
End is mounted in the GPU cluster application system on server.
The step of carrying out the method for children's gait health supervision using intelligent children's safety and gait health monitoring system is such as
Shown in Fig. 2：
1, pressure spectrum data of the children to sole is acquired by the built-in of supervision shoe；
2, high in the clouds analysis process system is uploaded to by 4G networks；
3, high in the clouds analysis process system uses depth convolutional neural networks analyzing processing pressure spectrum data, is instructed in conjunction with early period
The children's gait health data got compares, and judges children's gait health status.
Maximum difference lies in it to learn automatically to number from big data for deep learning and traditional mode identification method
According to feature, and it is the feature of engineer that traditional mode identification method, which mainly utilizes,.It is artificial in the past few decades
The feature of design is in leading position.It relies primarily on the priori of designer, dependent on manually adjusting parameter, and it is artificial
The parameter of design is less, is difficult with a large amount of data.Deep learning can from the data of magnanimity automatic learning characteristic, this its
In can include thousands of a parameters, and the parameter for only needing manual setting less just can obtain preferable result.This be by
Multilayer neural network is obtained by the training of mass data in deep learning, it is special that combination low-level feature forms more abstract high level
Sign indicates, to find that the distributed nature of data indicates.
The feature extraction network main thought used in the present invention derives from VGGNet and ResNet.VGGNet is big by Oxford
The depth convolutional network that computer vision group proposes is learned, by stacking 3x3 convolutional layers and 2x2 maximums pond layer, success structure repeatedly
16-19 layers of depth convolutional network has been built, the achievement of 2014 tournament sortings of ILSVRC the 2nd and positioning the 1st is obtained.Except this
Except, the extended capability of network is very strong, moves in other structures that Generalization Capability is preferable, is relatively common feature extraction
Network.Shown in VGG-16 network structures such as Fig. 3 (a).Although VGGNet achieves good effect in sorter network, due to dividing
Generic task input is whole pictures, is classified by extracting the feature of image content by depth convolutional network, due to picture
Characteristic area it is apparent and to account for the proportion of entire picture larger, so being abstracted to entire picture extraction feature and constantly and can obtain
Preferable result.But in specific objective identification mission (the O-shaped leg of X-type leg), although character network purpose is also for can
The feature of whole image is extracted, but due to the convolution sum pondization operation in network so that it takes up space in picture originally
Fewer target area is smaller and smaller during convolution sum pond, is finally difficult to extract feature.
The ResNet depth residual error networks that Kaiming He et al. are proposed are under the premise of ensureing neural network accuracy by network
Depth training has reached 152 layers, promotes the depth to 1000 layers again later.Author has used short connection in depth residual error network
Operation, concrete structure are as shown in Figure 4.In lower graph structure, inputs as X, be F (x) by the output of two layer operations, pass through short connection
The output of entire module is defined as F (X)+X, i.e. the module not only exports the characteristic pattern after forward calculation, also by input feature vector
Figure is added in output, and module exports abstract characteristics and rough features simultaneously.In addition to this, in the process for carrying out backpropagation
In, due to the presence of short connection so that gradient can be returned directly to input, this can be effectively in less too deep network
Gradient disappearance problem.
But residual error network is due to the complicated network structure, it is difficult to which training occupies a large amount of computing resource, is difficult in practice
Using.Therefore, VGGNet is improved according to the thought of residual error network in the present invention, especially with the operation of short connection.And
Mainly there are two effects for short attended operation in the present invention：
(1) training speed is improved.In the training process due to the presence of short connection, gradient disappearance problem, gradient are alleviated
Can fast propagation in a network, and the experiment of the present invention also demonstrates improved network compared to primitive network training speed
Faster.
(2) primitive character and abstract characteristics are combined.In the process, feature extraction is not to extract the spy of whole image
Sign, it is preferred that emphasis is the character representation of target area.Since target area is reduced rapidly by convolution pond operational size so that most
Whole character representation is inaccurate, so the present invention blends primitive character and abstract characteristics, solves target area size
Reduce the insufficient problem of the characteristic present brought.
Shown in network proposed by the present invention such as Fig. 3 (b).The feature extraction network of the present invention is broadly divided into two parts：It is main
Dry network and branching networks.Core network is similar to VGGNet, and feature is extracted by concatenated convolutional.It is wherein different from VGGNet
It is not use pond layer in core network, but the 3x3 convolution that step-length is 2 is used to substitute, this is because pond is to entire
The feature in region carry out can not inverse operation, and many information are lost during pond, especially the smaller feelings in target area
Under condition, pondization operation carries out unified operation to target area and background area so that the target area of very little is smaller, follow-up
Convolution operation in be more difficult to extract feature.
In addition to this, 1x1 convolution operations are also added in core network, this be in order to the characteristic pattern of input into row of channels
Fusion and reduce the quantity in channel, to reduce calculation amount.In core network, convolution is packaged into Block modules, a side
Face is succinct for network, is on the other hand in order to which the thought for being multiplexed feature is used for network local shape factor.Block has
Body structure is as shown in Figure 5.Branching networks are made of pondization operation and 1x1 convolution operations, and pondization operation is for uniform characteristics figure
Size, 1x1 convolution operations are in order to reduce the number of active lanes of input feature vector, to reduce the calculation amount of next step convolution.
In order to more effectively evaluate children's danger classes state, the special data to sensor acquisition carry out children's danger classes
It scores (1~5).The present invention scores to video by four layers of fully-connected network, and training input is a bit of time gyroscope
Information, GPS information, humidity sensor information are exported and are scored (1~5 point) for video, and process is as follows：
(1) data normalization
In this step, it is normalized between 0~1 by maximum value.
(2) training network
The present invention scores to video by four layers of fully-connected network, specifically inputs a bit of time gyroscope letter
Breath, GPS information, humidity sensor information export 5 kinds of classifications, respectively corresponding scoring 1-5.It is 500 times that iterations, which are arranged,
Batch size are 16, learning rate 0.001, and verification collection is 10%.After 500 iteration, network convergence, training set rate of accuracy reached
To 92%, test set rate of accuracy reached to 91%.
4, using LSTM (Long Short-Term Memory) deep learning network based on time series, to it is different when
Between the foot pressure of state analyzed, whether child development state is well judged.
We can judge whether children are O-shaped leg or X-type leg by depth convolutional neural networks in the present invention.But it is deep
Convolutional neural networks are unpredictable will develop into the bad development such as O-shaped leg or X-type leg and children in a period for degree
Whether the inside physically well develops.We use depth convolutional neural networks (CNN) to be used as feature extractor first, it is assumed that CNN is extracted
Characteristic dimension be N (this feature is exactly the full articulamentum that network is last in step 3).Then for the tonogram of K periods
Spectrum just constitutes the N-dimensional characteristic sequence that sequential length is K.Then using this sequence as the input of LSTM, obtained LSTM's
Output remains the sequence that a length is K (dimension should be the number of action classification).Then by training, differentiate the time
Whether children's gait of sequence K physically well develops.
5, by above-mentioned child development status information feedback to user, user's marking；
6, convolutional neural networks and LSTM deep learning networks are finely adjusted according to field feedback, to obtain more
For accurate analysis result；
7, the continuous iteration above process, until result is stablized, process terminates.
It should be noted that above-described embodiment provided by the present invention only has schematically, does not have and limit the present invention's
The effect of the range of specific implementation.Protection scope of the present invention should for those of ordinary skill in the art be shown including those
And the transformation being clear to or alternative solution.
Claims (7)
- Include that the supervision shoe worn for children and guardian use 1. a kind of intelligence children's safety and gait health monitoring system Intelligent terminal, it is characterised in that pressure sensitive, positioning system, gyroscope, voice input and humidity built in the supervision shoe Sensing device, system also include high in the clouds analysis process system, the number that high in the clouds analysis process system acquires supervision shoe built-in According to analyzing processing is carried out, it is then passed to the intelligent terminal that guardian uses.
- 2. intelligence children's safety according to claim 1 and gait health monitoring system, it is characterised in that the intelligence is eventually End equipment includes smartwatch, mobile terminal, home intelligent alarm speaker.
- 3. intelligence children's safety according to claim 1 and gait health monitoring system, it is characterised in that pass through the prison Fence, children's gait analysis, the analysis of leg skeleton development, danger sound may be implemented in the data of shield shoes built-in acquisition Judgement, tumble, taken in one's arms and overboard monitoring function.
- 4. intelligence children's safety according to claim 3 and gait health monitoring system, it is characterised in that system is monitoring It is in children and falls, taken in one's arms and send out alarm to guardian by vibrations or alarm tone when overboard state.
- 5. a kind of carrying out children's gait health prison using intelligent children's safety described in claim 1 and gait health monitoring system The method of shield, which is characterized in that include the following steps：(1) intelligent children's safety and gait health monitoring system acquire pressure of the children to sole by the built-in of supervision shoe Spectrum data；(2) data of above-mentioned acquisition are uploaded to high in the clouds analysis process system；(3) high in the clouds analysis process system uses the above-mentioned pressure spectrum data of depth convolutional neural networks analyzing processing, and before combination Children's gait health data that phase trains compares, and judges children's gait health status；(4) high in the clouds analysis process system utilizes the LSTM deep learning networks based on time series, to the foot of different time state Whether portion's pressure is analyzed, well judge child development state；(5) user is allowed to give a mark above-mentioned child development status information feedback to user；(6) convolutional neural networks and LSTM deep learning networks are finely adjusted according to field feedback, to obtain more Accurate analysis result；(7) the continuous iteration above process, until result is stablized, process terminates.
- 6. according to claim 5 carry out children's gait health prison using intelligent children's safety and gait health monitoring system The method of shield, it is characterised in that health status described in step 3 comprises determining whether X-type, O-shaped leg, and sole is exerted oneself situation.
- 7. according to claim 5 carry out children's gait health prison using intelligent children's safety and gait health monitoring system The method of shield, it is characterised in that child development state described in step 4 includes weight, on foot posture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810447694.1A CN108683724A (en) | 2018-05-11 | 2018-05-11 | A kind of intelligence children's safety and gait health monitoring system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810447694.1A CN108683724A (en) | 2018-05-11 | 2018-05-11 | A kind of intelligence children's safety and gait health monitoring system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108683724A true CN108683724A (en) | 2018-10-19 |
Family
ID=63805524
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810447694.1A Pending CN108683724A (en) | 2018-05-11 | 2018-05-11 | A kind of intelligence children's safety and gait health monitoring system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108683724A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109394229A (en) * | 2018-11-22 | 2019-03-01 | 九牧厨卫股份有限公司 | A kind of fall detection method, apparatus and system |
| CN109783619A (en) * | 2018-12-14 | 2019-05-21 | 广东创我科技发展有限公司 | A kind of data filtering method for digging |
| CN109998508A (en) * | 2019-04-18 | 2019-07-12 | 北京工业大学 | The intelligent shoe of detectable and prediction physical condition and positioning and warning function with self-charging |
| CN110163175A (en) * | 2019-05-28 | 2019-08-23 | 杭州电子科技大学 | A kind of gait recognition method and system based on improvement VGG-16 network |
| CN111436944A (en) * | 2020-04-20 | 2020-07-24 | 电子科技大学 | Falling detection method based on intelligent mobile terminal |
| CN111967400A (en) * | 2020-08-19 | 2020-11-20 | 顾晓东 | Campus safety monitoring and early warning management system based on artificial intelligence |
| CN111973193A (en) * | 2020-08-20 | 2020-11-24 | 中山大学 | Gait recognition method and system based on silicone-nickel nano sensor |
| CN112101472A (en) * | 2020-09-21 | 2020-12-18 | 长沙昱旻信息科技有限公司 | Shoe identification method, device and system, computer storage medium and electronic equipment |
| CN112419669A (en) * | 2020-11-19 | 2021-02-26 | 重庆电子工程职业学院 | Wisdom endowment system |
| CN112784987A (en) * | 2019-01-29 | 2021-05-11 | 武汉星巡智能科技有限公司 | Target nursing method and device based on multistage neural network cascade |
| CN112965488A (en) * | 2021-02-05 | 2021-06-15 | 重庆大学 | Baby monitoring mobile machine trolley based on transfer learning neural network |
| CN113436426A (en) * | 2021-06-22 | 2021-09-24 | 北京时空数智科技有限公司 | Personnel behavior warning system based on video AI analysis |
| CN113780223A (en) * | 2021-09-09 | 2021-12-10 | 北京信息科技大学 | Gait recognition method and device for artificial limb and storage medium |
| CN116304596A (en) * | 2023-05-26 | 2023-06-23 | 深圳市明源云科技有限公司 | Indoor child safety monitoring method and device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103961109A (en) * | 2014-05-05 | 2014-08-06 | 北京航空航天大学 | Human body posture detection device based on acceleration signals and angular speed signals |
| CN105631195A (en) * | 2015-12-18 | 2016-06-01 | 合肥工业大学 | Wearable multi-information fusion gait analysis system and method thereof |
| CN106264544A (en) * | 2016-07-28 | 2017-01-04 | 江苏人之初母婴用品有限公司 | A kind of intelligence children's safety monitor system and method |
| CN107277164A (en) * | 2017-07-21 | 2017-10-20 | 重庆市端峰科技有限公司 | A kind of children's long-distance intelligent monitor system |
-
2018
- 2018-05-11 CN CN201810447694.1A patent/CN108683724A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103961109A (en) * | 2014-05-05 | 2014-08-06 | 北京航空航天大学 | Human body posture detection device based on acceleration signals and angular speed signals |
| CN105631195A (en) * | 2015-12-18 | 2016-06-01 | 合肥工业大学 | Wearable multi-information fusion gait analysis system and method thereof |
| CN106264544A (en) * | 2016-07-28 | 2017-01-04 | 江苏人之初母婴用品有限公司 | A kind of intelligence children's safety monitor system and method |
| CN107277164A (en) * | 2017-07-21 | 2017-10-20 | 重庆市端峰科技有限公司 | A kind of children's long-distance intelligent monitor system |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109394229A (en) * | 2018-11-22 | 2019-03-01 | 九牧厨卫股份有限公司 | A kind of fall detection method, apparatus and system |
| CN109783619A (en) * | 2018-12-14 | 2019-05-21 | 广东创我科技发展有限公司 | A kind of data filtering method for digging |
| CN112784987B (en) * | 2019-01-29 | 2024-01-23 | 武汉星巡智能科技有限公司 | Target nursing method and device based on multistage neural network cascade |
| CN112784987A (en) * | 2019-01-29 | 2021-05-11 | 武汉星巡智能科技有限公司 | Target nursing method and device based on multistage neural network cascade |
| CN109998508A (en) * | 2019-04-18 | 2019-07-12 | 北京工业大学 | The intelligent shoe of detectable and prediction physical condition and positioning and warning function with self-charging |
| CN110163175A (en) * | 2019-05-28 | 2019-08-23 | 杭州电子科技大学 | A kind of gait recognition method and system based on improvement VGG-16 network |
| CN111436944A (en) * | 2020-04-20 | 2020-07-24 | 电子科技大学 | Falling detection method based on intelligent mobile terminal |
| CN111967400B (en) * | 2020-08-19 | 2021-02-26 | 青岛伟东云教育集团有限公司 | Campus safety monitoring and early warning management system based on artificial intelligence |
| CN111967400A (en) * | 2020-08-19 | 2020-11-20 | 顾晓东 | Campus safety monitoring and early warning management system based on artificial intelligence |
| CN111973193A (en) * | 2020-08-20 | 2020-11-24 | 中山大学 | Gait recognition method and system based on silicone-nickel nano sensor |
| CN112101472A (en) * | 2020-09-21 | 2020-12-18 | 长沙昱旻信息科技有限公司 | Shoe identification method, device and system, computer storage medium and electronic equipment |
| CN112419669A (en) * | 2020-11-19 | 2021-02-26 | 重庆电子工程职业学院 | Wisdom endowment system |
| CN112965488A (en) * | 2021-02-05 | 2021-06-15 | 重庆大学 | Baby monitoring mobile machine trolley based on transfer learning neural network |
| CN112965488B (en) * | 2021-02-05 | 2022-06-03 | 重庆大学 | Baby monitoring mobile machine trolley based on transfer learning neural network |
| CN113436426A (en) * | 2021-06-22 | 2021-09-24 | 北京时空数智科技有限公司 | Personnel behavior warning system based on video AI analysis |
| CN113780223A (en) * | 2021-09-09 | 2021-12-10 | 北京信息科技大学 | Gait recognition method and device for artificial limb and storage medium |
| CN116304596A (en) * | 2023-05-26 | 2023-06-23 | 深圳市明源云科技有限公司 | Indoor child safety monitoring method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108683724A (en) | A kind of intelligence children's safety and gait health monitoring system | |
| Wang et al. | Fall detection based on dual-channel feature integration | |
| CN107153871B (en) | Falling detection method based on convolutional neural network and mobile phone sensor data | |
| Avola et al. | Deep temporal analysis for non-acted body affect recognition | |
| CN109009145A (en) | A kind of tumble judgment method based on wearable device | |
| Xu et al. | Elders’ fall detection based on biomechanical features using depth camera | |
| Javeed et al. | Body-worn hybrid-sensors based motion patterns detection via bag-of-features and Fuzzy logic optimization | |
| Sun et al. | Fall detection algorithm for the elderly based on human posture estimation | |
| Cai et al. | GBDT‐Based Fall Detection with Comprehensive Data from Posture Sensor and Human Skeleton Extraction | |
| CN115116127A (en) | Fall detection method based on computer vision and artificial intelligence | |
| Hua et al. | Falls prediction based on body keypoints and seq2seq architecture | |
| Hernandez et al. | Machine learning techniques for motion analysis of fatigue from manual material handling operations using 3D motion capture data | |
| Casilari et al. | A wearable fall detection system using deep learning | |
| CN115586834A (en) | Intelligent cardio-pulmonary resuscitation training system | |
| Beily et al. | A sensor based on recognition activities using smartphone | |
| CN108717548B (en) | Behavior recognition model updating method and system for dynamic increase of sensors | |
| Du et al. | RETRACTED: Research on the intelligent model of progress in physical education training based on motion sensor | |
| Jiang et al. | Deep learning algorithm based wearable device for basketball stance recognition in basketball | |
| Yuan et al. | Adaptive recognition of motion posture in sports video based on evolution equation | |
| Kong et al. | Design of computer interactive system for sports training based on artificial intelligence and improved support vector | |
| CN113780223A (en) | Gait recognition method and device for artificial limb and storage medium | |
| Leite et al. | Three-stream convolutional neural network for human fall detection | |
| Chen et al. | Improve the Accuracy of Fall Detection Based on Artificial Intelligence Algorithm. | |
| Feng et al. | Genetic algorithm based optimization of deep neural network ensemble for personal identification in pedestrians behaviors | |
| CN108648417A (en) | Raspberry Pi old man based on deep learning falls down detecting system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181019 |
|
| RJ01 | Rejection of invention patent application after publication |