CN110047249A - Human body attitude identification sensor, human posture recognition method and system - Google Patents
Human body attitude identification sensor, human posture recognition method and system Download PDFInfo
- Publication number
- CN110047249A CN110047249A CN201910443623.9A CN201910443623A CN110047249A CN 110047249 A CN110047249 A CN 110047249A CN 201910443623 A CN201910443623 A CN 201910443623A CN 110047249 A CN110047249 A CN 110047249A
- Authority
- CN
- China
- Prior art keywords
- human body
- neural network
- human
- microwave
- identification sensor
- 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
- 238000000034 method Methods 0.000 title claims abstract description 17
- 238000013528 artificial neural network Methods 0.000 claims abstract description 58
- 238000012544 monitoring process Methods 0.000 claims abstract description 39
- 238000001514 detection method Methods 0.000 claims abstract description 12
- 230000036387 respiratory rate Effects 0.000 claims abstract description 11
- 238000001914 filtration Methods 0.000 claims description 41
- 238000013135 deep learning Methods 0.000 claims description 24
- 230000005856 abnormality Effects 0.000 claims description 15
- 238000004891 communication Methods 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 5
- 210000000988 bone and bone Anatomy 0.000 claims description 4
- 230000008859 change Effects 0.000 abstract description 8
- 238000009434 installation Methods 0.000 abstract description 4
- 201000010099 disease Diseases 0.000 description 10
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 10
- 230000000694 effects Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 230000001276 controlling effect Effects 0.000 description 5
- 230000000873 masking effect Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000036772 blood pressure Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000010304 firing Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012806 monitoring device Methods 0.000 description 2
- 230000000474 nursing effect Effects 0.000 description 2
- 230000037081 physical activity Effects 0.000 description 2
- 206010008190 Cerebrovascular accident Diseases 0.000 description 1
- 208000006011 Stroke Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 210000000746 body region Anatomy 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0438—Sensor means for detecting
- G08B21/0461—Sensor means for detecting integrated or attached to an item closely associated with the person but not worn by the person, e.g. chair, walking stick, bed sensor
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0438—Sensor means for detecting
- G08B21/0476—Cameras to detect unsafe condition, e.g. video cameras
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Emergency Management (AREA)
- Business, Economics & Management (AREA)
- Gerontology & Geriatric Medicine (AREA)
- Multimedia (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Computer Interaction (AREA)
- Alarm Systems (AREA)
Abstract
This application provides a kind of human body attitude identification sensor, human posture recognition method and systems, by microwave signal and infrared image, carry out behavior monitoring to the human body in monitoring range.The indexs such as respiratory rate, HR Heart Rate that human body is monitored by microwave signal the case where establishing human skeleton model in conjunction with infrared image, monitor human body by skeleton model, when occurring falling, squatting long, can timely be found and warning reminding.The sensor uses embedded neural network processor, and face can be rapidly completed and body state detection, the sensor do not record video data, be not related to privacy of user, be not required to configuration local server, storage equipment, save cost.Sensor installation is simple, is that non-intervention type passively monitors, does not change monitored personnel's daily habits, can obtain more data, improves monitoring accuracy.To network bandwidth no requirement (NR), no network delay, system upgrade only needs to increase and decrease sensor, can carry out system update by long-range OTA.
Description
Technical field
This application involves sensor technical fields, in particular to a kind of human body attitude identification sensor, human body attitude
Recognition methods and system.
Background technique
As population tends to the phenomenon aggravation of aging, the endowment of old man is looked after situation also sternness increasingly, has been had at present
More and more old man's selections are in community endowment.Community center etc. is the place of the daily centralized activity of old man, for old man
Nurse can manually be carried out by caregiver, also can be used the devices such as heart rate monitor apparatus or blood pressure monitoring device into
Row, but quantity due to caregiver and energy are all limited, and having a single function for monitoring device is likely to occur old man at any time
Risk is difficult to effectively be monitored.For example, in certain places, it is difficult to timely and effectively find the old man to fall.Therefore, urgently
Need a kind of device that gesture recognition can be carried out to old man.
Summary of the invention
In order at least overcome above-mentioned deficiency in the prior art, the first purpose of the application is to provide a kind of human body attitude
Identification sensor, human posture recognition method and system.
Technical solution provided by the present application is as follows:
A kind of human body attitude identification sensor, human body gesture recognition sensor include:
Bottom cover and fixed frame, the bottom cover and fixed frame are fixed and are enclosed accommodating chamber；
Circuit board, the circuit board are fixed in the accommodating chamber；
Neural network processor on the circuit board is set；
Microwave transmitting antenna, microwave antenna and the binocular infrared camera being connect with the neural network processor,
The microwave transmitting antenna is for emitting microwave signal, and the microwave antenna is for described in receiving and reflecting through other objects
Microwave signal, the binocular infrared camera is for acquiring infrared image；It is provided on the fixed frame described micro- for fixing
Wave transmitting antenna, the first fixation hole of microwave antenna and the second fixation for fixing the binocular infrared camera
Hole；
Filtering cover, the filtering, which is covered, is fixed on side of the fixed frame far from the accommodating chamber, on the filtering cover
Offer the first through hole to match with first fixation hole；
Baffle, the baffle plate setting offers on the separate fixed frame side of filtering cover, the baffle and institute
State the second through-hole that first through hole matches, planar movement where the baffle can be covered along the filtering, by adjusting described
The overlapping area of second through-hole and the first through hole changes the direction of the launch and receiving direction of the microwave signal.
Further, filtering cover both ends offer sliding slot respectively, are provided on the baffle and the sliding slot phase
The sliding block matched, the baffle pass through the relatively described sliding slot sliding of the sliding block.
Further, human body gesture recognition sensor further include:
It is arranged on the circuit board, the communication module being connected with the neural network processor, the communication module
For the processing result of the neural network processor to be sent to other external equipments.
Further, human body gesture recognition sensor further include:
Side of the fixed frame far from the accommodating chamber is arranged in RF transmitter, the RF transmitter,
And be connected with the neural network processor, for launching outward infrared ray.
Further, the filter to match with the binocular infrared camera, the filter are provided on the filtering cover
For filtering out the ambient light into the binocular infrared camera.
Present invention also provides a kind of human posture recognition methods, are applied to above-mentioned human body attitude identification sensor, described
Human body attitude identification sensor pre-establishes the communication connection with Cloud Server, and the Cloud Server is for establishing deep learning mind
The deep learning neural network is trained through network, and using default tumble image, human body gesture recognition method packet
It includes:
According to the microwave signal received, determine in monitoring range with the presence or absence of human body；
When determining in the monitoring range there are when human body, according to the infrared image, determine that target body is sat default
Human body depth information in mark system；
Recognition of face is carried out to the target body, determines the personal information of the target body；
The target body is drawn according to the human body depth information based on default human body prediction model
Skeleton model；
The skeleton model is monitored using the deep learning neural network that training terminates；
When monitoring the skeleton model is abnormality, warning message corresponding with the personal information is generated, and
It is sent to the Cloud Server, the abnormality includes falling, squat down for a long time or helping for a long time to hold other objects.
Further, whether according to the microwave signal received, determining in monitoring range has the step of human body to include:
According to the microwave signal received, determine whether there are objects moving in detection range；
When detecting that there are objects moving, the HR Heart Rate and breathing speed of mobile object are determined based on the microwave signal
Rate；
When the HR Heart Rate or respiratory rate that determine mobile object meet preset condition, someone in detection range is determined
Body.
Further, when determining in monitoring range there are when human body, according to the infrared image, determine target body pre-
If the step of human body depth information in coordinate system, includes:
Location information based on same target point in the infrared image on each infrared camera image plane calculates
D coordinates value of the target point in the preset coordinate system；
The D coordinates value is converted into the coordinate value in two dimensional image, determines multiple target points in the two dimensional image
In depth information, the human body depth information as the target body.
It further, include the connection relationship of the multiple crucial skeleton points of human body in the default human body prediction model,
The skeleton model of the target body is drawn according to the human body depth information based on default human body prediction model
The step of include:
According to the infrared image, at least one crucial skeleton point of the target body is determined；
Institute is obtained according to the connection relationship of the multiple crucial skeleton point and the human body depth information, drafting
State the skeleton model of target body.
Present invention also provides a kind of human body attitude identifying system, including above-mentioned human body attitude identification sensor, Yi Jiyu
The Cloud Server of the human body attitude identification sensor connection, the Cloud Server are used to establish depth for default tumble image
Learning neural network, and the deep learning neural network is trained so that the human body attitude identification sensor according to
The training result of the deep learning neural network carries out abnormality detection monitoring human body.
In terms of existing technologies, the application has the advantages that human body attitude provided by the embodiments of the present application
Identification sensor includes bottom cover, fixed frame and filtering cover, is provided with neural network sensor, micro- in accommodating chamber inside it
Wave transmitting antenna, microwave antenna and binocular infrared camera, filtering, which is covered, is additionally provided with baffle, and baffle is relative to filtering
The position of cover can move, and baffle is allowed to carry out the masking of different area to the microwave signal that microwave transmitting antenna issues,
To change the coverage area of microwave signal, allow the coverage area of the microwave signal of human body gesture recognition sensor according to reality
Border needs to adjust, and improves the scope of application of human body attitude gesture sensor, meets the needs of human body gesture recognition under different scenes.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only some embodiments of the application, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 is a kind of structural schematic diagram of human body attitude identification sensor provided by the embodiments of the present application.
Fig. 2 is a kind of another structural schematic diagram of human body attitude identification sensor provided by the embodiments of the present application.
Fig. 3 is the structural schematic diagram of fixed frame in a kind of human body attitude identification sensor provided by the embodiments of the present application.
Fig. 4 to fig. 6 is the knot of filtering cover and baffle in a kind of human body attitude identification sensor provided by the embodiments of the present application
Structure schematic diagram.
Fig. 7 is a kind of schematic diagram of human posture recognition method provided by the embodiments of the present application.
Fig. 8 is a kind of schematic diagram of human body attitude identifying system provided by the embodiments of the present application.
Icon: 100- human body attitude identifying system；10- human body attitude identification sensor；20- Cloud Server；101- bottom cover；
102- fixed frame；103- circuit board；104- neural network processor；105- microwave transmitting antenna；106- microwave antenna；
107- binocular infrared camera；108- filtering cover；109- baffle；110- communication module；The first fixation hole of 111-；112- second is solid
Determine hole；181- first through hole；182- sliding slot；The second through-hole of 191-；192- sliding block.
Specific embodiment
To keep the purposes, technical schemes and advantages of the embodiment of the present application clearer, below in conjunction with the embodiment of the present application
In attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is
Some embodiments of the present application, instead of all the embodiments.The application being usually described and illustrated herein in the accompanying drawings is implemented
The component of example can be arranged and be designed with a variety of different configurations.
Therefore, the detailed description of the embodiments herein provided in the accompanying drawings is not intended to limit below claimed
Scope of the present application, but be merely representative of the selected embodiment of the application.Based on the embodiment in the application, this field is common
Technical staff's every other embodiment obtained without making creative work belongs to the model of the application protection
It encloses.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.
In the description of the present application, it is also necessary to which explanation is unless specifically defined or limited otherwise, term " setting ",
" installation ", " connected ", " connection " shall be understood in a broad sense, for example, it may be fixedly connected, may be a detachable connection or one
Connect to body；It can be mechanical connection, be also possible to be electrically connected；It can be directly connected, it can also be indirect by intermediary
It is connected, can be the connection inside two elements.For the ordinary skill in the art, on being understood with concrete condition
State the concrete meaning of term in this application.
Existing senior activity place there are problems that following risk and: due to the generally open place in senior activity place,
It is easy to be waited by people without fixed duties or criminal is mixed into, including health care product sales force etc.；It is difficult to carry out user data statistics, including people
Traffic statistics etc.；Old man is easy to appear the risks such as disease burst in activity, is difficult to find in time by existing means；It is difficult to lead to
The early stage behavioral data for crossing old man's disease incidence carries out Disease Warning Mechanism.To sum up, old man has both sides wind in community center
Danger, one is security risk, including identification and data statistics.Another is active risks, including tumble or disease are dashed forward
Hair.
Efficiency and level is looked after to looking after for community center old man in order to improve, a variety of improvement means occurs.Example
Such as, increase the quantity of community center nursing staff；It is equipped with information collecting device for every old man, these information collecting devices can
The location information and vital sign information (blood pressure, heart rate etc.) of old man are acquired, thus administrative staff can be allowed simply to grasp always
The status of people.But all there is disadvantage in these countermeasures and management means.The pass refined by increasing operation personnel
The risk of note safety and activity aspect, can have higher operation cost and can not replicate popularization on a large scale.By wearing wearing
In the means of equipment, certain old men can resist wearing, while be easy to forget to charge, so that old man can not be hard during actual operation
It holds and gives up.Importantly, these facilities and administration systems lack the monitoring of the sudden abnormal conditions such as the tumble of old man at present
Weary low cost, effective monitoring method.
In view of this, the embodiment of the present application provides a kind of human body attitude identification sensor 10, it as depicted in figs. 1 and 2, should
Human body attitude identification sensor 10 includes bottom cover 101, fixed frame 102, circuit board 103, neural network processor 104, microwave hair
Penetrate antenna 105, microwave antenna 106, binocular infrared camera 107, filtering cover 108 and baffle 109.
The bottom cover 101 and fixed frame 102 are fixed and are enclosed accommodating chamber, and bottom cover 101 and fixed frame 102 can lead to
It crosses screw or other fixed forms is fixed, the bottom cover 101 in the application can make to be formed with cuboid sky open at one end
Chamber, fixed frame 102 can be fixed on the opening of cuboid cavity.The circuit board 103 can be fixed on the accommodating chamber
It is interior.
Neural network processor 104 can be set on the circuit board 103, microwave transmitting antenna 105, microwave receiving day
Line 106 and binocular infrared camera 107 are connect with the neural network processor 104.The microwave transmitting antenna 105 is for sending out
Microwave signal is penetrated, the microwave antenna 106 is for receiving the microwave signal reflected through other objects.At neural network
Signal firing order can be generated in reason device 104, and microwave transmitting antenna 105 can launch outward microwave letter according to signal firing order
Number.In the embodiment of the present application, neural network sensor can use the embedded neural network processor of model CK860MP
104。
The binocular infrared camera 107 is for acquiring infrared image.It can be subsequent by binocular infrared camera 107
It carries out human body depth information on image and available infrared image is provided.Binocular infrared camera 107 can be by two
Omnivision OV16860 camera module composition.
As shown in figure 3, being provided on the fixed frame 102 for fixing the microwave transmitting antenna 105, microwave receiving day
First fixation hole 111 of line 106 and the second fixation hole 112 for fixing the binocular infrared camera 107.It is micro- in order to make
The microwave signal that wave transmitting antenna 105 issues as much as possible can take a walk into environment, and the first fixation hole 111 can be provided with
The both ends of fixed frame 102, the quantity of the first fixation hole 111 are two, one of them is for fixing microwave transmitting antenna 105, separately
One for fixing microwave antenna 106.Second fixation hole 112 can be located at the intermediate position of fixed frame 102, and second is fixed
Hole 112 can also open up two, realize the fixation to each camera module.
As shown in figure 4, the filtering cover 108 is fixed on the side of the fixed frame 102 far from the accommodating chamber, it is described
The first through hole 181 to match with first fixation hole 111 is offered on filtering cover 108.First opened up on filtering cover 108
Through-hole 181 can allow microwave signal to pass through, and the position of first through hole 181 can be determined according to the position of the first fixation hole 111.
As shown in Figure 4 and Figure 5, the setting of baffle 109 is in filtering cover 108 far from 102 side of fixed frame, institute
The second through-hole 191 for offering on baffle 109 and matching with the first through hole 181 is stated, the baffle 109 can be along the filtering
108 place planar movements are covered, with the overlapping area by adjusting second through-hole 191 and the first through hole 181, change institute
State the direction of the launch and receiving direction of microwave signal.
Human body attitude identification sensor 10 in the embodiment of the present application can be installed in indoor roof in use
Or other positions, after installation fixes, integral position is difficult to flexibly be moved human body gesture recognition sensor 10.
And since the microwave signal that microwave transmitting antenna 105 issues has certain range and directive property, it identifies and senses in human body attitude
In the case that the integral position of device 10 is unable to adjust, first through hole 181 and second can be made by adjusting the position of baffle 109
The overlapping area of through-hole 191 changes, when the overlapping area of first through hole 181 and the second through-hole 191 is larger, fixed frame
Part as much as possible is just had in the microwave signal that microwave transmitting antenna 105 on 102 issues passes through first through hole 181 and the
Two through-holes 191.And when the overlapping area of first through hole 181 and the second through-hole 191 becomes smaller, what microwave transmitting antenna 105 issued
Microwave signal will be reduced by the signal of first through hole 181 and the second through-hole 191.Pass through first through hole 181 and the second through-hole
191 microwave signal number can influence the spread scope of microwave signal, while the position of baffle 109 can also change microwave letter
Number propagation angle.By the way that baffle 109 is arranged, the direction of propagation to microwave signal and the control of propagation angle may be implemented.
Again as shown in figure 5,108 both ends are covered in the filtering can offer sliding slot 182 respectively, it is arranged on the baffle 109
There is the sliding block 192 to match with the sliding slot 182, the baffle 109 is connected by the sliding block 192 and the sliding slot 182 sliding
It connects.Sliding slot 182 can be provided with the both ends of filtering cover 108 respectively, and in the embodiment of the present application, the extending direction of sliding slot 182 can
With the length direction perpendicular to filtering cover 108, in this way, baffle 109 can after baffle 109 is connect by sliding block 192 with sliding slot 182
It can be with manual toggle sliding block when needing the position of regulating fender 109 to do the movement for covering 108 length directions perpendicular to filtering
192 position in sliding slot 182.Dial up sliding block 192,182 upward sliding along the chute of sliding block 192, the also Xiang Shangyi of baffle 109
It is dynamic.In the embodiment of the present application, the width of sliding slot 182 can be slightly less than the width of sliding block 192, and sliding block 192 can in sliding slot 182
With sliding, but when not having external force, sliding block 192 can be fixed in sliding slot 182, to realize the position of baffle 109
It is fixed.
In the embodiment of the present application, the area of the second through-hole 191 on baffle 109 can be less than or equal to first through hole
181 area, baffle 109 can select the material that can absorb microwave signal to be made, in baffle 109 by the second through-hole of part
After 191 maskings, the part microwave signal that microwave transmitting antenna 105 issues can be absorbed by baffle 109, not hidden by baffle 109
The microwave signal blocked can be emitted in external environment by the second through-hole 191.By adjusting the position of baffle 109, make
The position that second through-hole 191 and first through hole 181 are overlapped changes, and the shielded area of first through hole 181 can also be therewith
Transformation, to realize the adjustment of the microwave signal direction of propagation and spread scope.Cannot or inconvenient human body attitude is identified pass
When the integral position of sensor 10 is adjusted, the coverage area of microwave signal can be adjusted by adjusting the position of baffle 109.
In the embodiment of the present application, in order to realize that transmitting range and range of receiving to microwave signal carry out more flexible tune
It is whole, can baffle 109 be provided only on the corresponding position of microwave transmitting antenna 105, can also be provided only on according to actual needs
The corresponding position of microwave antenna 106, can also be in microwave transmitting antenna 105 and the corresponding position of microwave antenna 106
It is respectively provided with baffle 109.It, can be only again as shown in figure 5, when baffle 109 is provided only on 105 corresponding position of microwave transmitting antenna
The transmitting range of microwave signal is adjusted.It, can when baffle 109 is provided only on 105 corresponding position of microwave transmitting antenna
To be only adjusted to the transmitting range of microwave signal.When baffle 109 is provided only on the corresponding position of microwave antenna 106
When, only the range of receiving of microwave signal can be adjusted.As shown in fig. 6, working as microwave transmitting antenna 105 and microwave receiving day
When line 106 is respectively provided with baffle 109, so that it may which transmitting range and range of receiving to microwave signal are all adjusted.
In actual use, in an indoor environment, in order to improve the accuracy of human body attitude identification, Ke Yipei
Set more human body attitude identification sensors 10, if not to every human body attitude be bound to sensor sending microwave signal covering
Range is adjusted, it is more likely that case occur that the Microwave emission day in a certain human body attitude identification sensor 10
The microwave signal that line 105 issues has entered directly into the range of receiving of an other human body attitude identification sensor 10, so that
The microwave signal that second human body attitude identification sensor 10 receives is the microwave signal not reflected by other objects, the
The microwave signal of two human body attitude identification sensors 10 in this way is not can determine that human body attitude.But if adjusting again again
The complete machine position of whole human body attitude identification sensor 10, workload can be very big.
It therefore, both can be by adjusting in the environment at one configured with more human body attitude identification sensors 10
The position of the corresponding baffle 109 of microwave antenna 106 in human body attitude identification sensor 10 passes a human body attitude identification
Sensor 10 will not directly receive the microwave signal of other human body gesture recognition sensors 10 sending.It is also possible to adjust it
The position of the corresponding baffle 109 of microwave transmitting antenna 105 on his human body gesture recognition sensor 10 passes human body gesture recognition
The microwave signal that sensor 10 issues is to cover the range of physical activity, without being directly connected to other human body gesture recognitions sensing
In range of receiving on device 10.Staff can be according to actual environment, only by adjusting different human body gesture recognition sensor
The position of 10 overhead gages 109, so that it may so that the microwave signal of multiple human body attitude identification sensors 10 as much as possible will be practical
The physical activity range of environment covers, while can directly be connect by other microwave antennas 106 to avoid the microwave signal of sending
The case where receipts, improves the association that multiple human body attitude identification sensors 10 work while reducing staff's maintenance workload
Tonality.
Optionally, then as depicted in figs. 1 and 2, human body gesture recognition sensor 10 further include: be arranged in the circuit board
On 103, the communication module 110 being connected with the neural network processor 104, the communication module 110 is used for the mind
Processing result through network processing unit 104 is sent to other external equipments.
In the embodiment of the present application, neural network sensor can judge according to the microwave signal and infrared image received
Whether there is human body in a falling state, after being determined that someone falls, can be sent to this result by communication module 110
Other external equipments, so as to the dangerous situation that staff can know to have human body to fall as soon as possible, so as to as soon as possible into
Row disposition.
Optionally, human body gesture recognition sensor 10 further include: RF transmitter, the RF transmitter setting
Be connected in the side of the fixed frame 102 far from the accommodating chamber, and with the neural network processor 104, for
Outer transmitting infrared ray.
Binocular infrared camera 107 can acquire the infrared image in environment, when place environment is in dark state or light
When line intensity is lower, binocular infrared camera 107 is just difficult to take enough clearly infrared images.Lead in the embodiment of the present application
Setting RF transmitter is crossed, infrared ray can be emitted, enough infrared rays is provided for binocular infrared camera 107, improves red
The clarity of outer image.
In the embodiment of the present application, it is also provided with and 107 phase of binocular infrared camera on the filtering cover 108
Matched filter, the filter are used to filter out the ambient light into the binocular infrared camera 107.Filtering 108 tables of cover
Face can coat one layer of filter coating or filter only is arranged in the corresponding position of binocular infrared camera 107, outside filtering cover 108
Side can not observe 10 internal structure of human body attitude identification sensor.
In conclusion human body attitude identification sensor 10 provided by the embodiments of the present application includes bottom cover 101, fixed frame 102
And filtering cover 108, neural network sensor, microwave transmitting antenna 105, microwave receiving are provided in accommodating chamber inside it
Antenna 106 and binocular infrared camera 107, filtering, which is covered, is additionally provided with baffle 109 on 108, baffle 109 is covered relative to filtering
108 position can move, and baffle 109 is allowed to carry out different area to the microwave signal that microwave transmitting antenna 105 issues
Masking make the coverage area of the microwave signal of human body gesture recognition sensor 10 to change the coverage area of microwave signal
It can adjust according to actual needs, improve the scope of application of human body attitude gesture sensor, meet human body attitude under different scenes and know
Other needs.
The embodiment of the present application also provides a kind of human posture recognition methods, are applied to above-mentioned human body attitude identification sensing
Device, the human body attitude identification sensor and Cloud Server communicate to connect, and the Cloud Server is for establishing deep learning nerve
Network, and the deep learning neural network is trained using default tumble image, as shown in fig. 7, human body posture is known
Other method includes the following steps.
Step S101 is determined in monitoring range according to the microwave signal received with the presence or absence of human body.
In actual use, multiple human body attitude identification sensors, multiple human bodies can be arranged under same usage scenario
The detection range of gesture recognition sensor is different.It is detecting whether there are when human body, it can be first according to the microwave received
Signal determines whether there are objects moving in detection range.When detecting that there are objects moving, is determined and moved based on the microwave signal
The HR Heart Rate and respiratory rate of dynamic object.When the HR Heart Rate or respiratory rate that determine mobile object meet preset condition
When, determining has human body in detection range.
HR Heart Rate and the corresponding preset condition of respiratory rate can be configured according to the difference of usage scenario, for example,
The HR Heart Rate of the elderly compares that young people is lower with respiratory rate, in the place of the elderly's frequent activity, HR Heart Rate and
Lower threshold value can be arranged in the corresponding preset condition of respiratory rate.Meanwhile the HR Heart Rate and respiratory rate of human body and its
His life entity also different from, can determine whether mobile object is human body by the microwave signal received.
Step S102, according to the infrared image, determines target body when determining in the monitoring range there are when human body
Human body depth information in preset coordinate system.
Human body attitude identification sensor is also shot by binocular infrared camera while transmitting and receive microwave signal
Image in monitoring range has determined in monitoring range there are after human body when by microwave signal, can further pass through red
Outer image determines the depth image of the target body.It is detailed, it can be first based on same target point in the infrared image every
Location information on a infrared camera image plane calculates the target point in the D coordinates value of the preset coordinate system.
Then the D coordinates value is converted into the coordinate value in two dimensional image, determines multiple target points in the two dimensional image
Depth information, the human body depth information as the target body.
Target body may be not stop movement in monitoring range, be also to the monitoring of target body by infrared image
In real time.The human body depth information being calculated is also real-time change.The step of detecting human body by microwave is more
It is rough, it can only position where rough determining human body.Since human body has certain temperature, in infrared signature and environment
Object infrared signature it is different, by infrared image can accurately determine monitoring environment in target body.
In the embodiment of the present application, human body attitude identification sensor includes using binocular infrared camera, the binocular infrared camera
Two camera modules, and there is a certain distance between two camera modules, form the binocular framework of similar human eye.Based in this way
Binocular infrared camera, position of a certain target point on two camera module image planes in infrared image can be calculated
Deviation is set, target point three-dimensional coordinate in three dimensions can be calculated according to the position deviation.Reconvert is into two dimensional image
Afterwards, the depth information of the target point can be recovered.Target point in the embodiment of the present application is the certain point in target body, is led to
Cross the calculating to target point depth informations multiple in infrared image, the overall depth information of available target body.
Step S103 carries out recognition of face to the target body, determines the personal information of the target body.
During being monitored to human body, it is thus necessary to determine that the relevant information of the target body, neural network processor can
To be prerecorded with the face information and corresponding personal information that often enter and leave all personnel in monitoring range, at neural network
Recognition of face can be carried out to target body, determine the personal information of the target body according to the infrared image taken by managing device.
Certainly, in some open usage scenarios, neural network sensor can not pre-recorded personnel face information and personnel
Information, neural network processor can carry out recognition of face to all personnel in monitoring range, and only for each personnel assignment
Vertical identification information, the personal information as the personnel.
Step S104 draws the mesh according to the human body depth information based on default human body prediction model
Mark human skeleton model.
Neural network processor, can be with base after the human body depth information for calculating target body by infrared image
In preset human body prediction model, the skeleton model of target body is drawn out.The people's body region prediction model can be with
It is that convolutional neural networks are instructed using the image for largely including human body depth information on Cloud Server in advance
Practice, human body prediction model is obtained by training, neural network sensor can be got from Cloud Server by training
Human body prediction model, and the skeleton model of target body can be drawn out according to human body depth information.
In the embodiment of the present application, skeleton model describes some major joint points and major joint point in skeleton
Connection relationship, can be with the current state of rough determining human body, such as human body by the connection relationship of these major joint points
It is different squatting down with the state of the artis under standing state.
Step S105 is monitored the skeleton model using the deep learning neural network that training terminates.
Deep learning neural network is that training is completed on Cloud Server, can pre-establish depth on Cloud Server
Neural network is practised, the deep learning neural network is trained using a large amount of sample skeleton model, sample skeleton model can
Think the human skeleton model under abnormality, in the embodiment of the present application, the skeleton model of abnormality includes but unlimited
Other objects are held in falling, squatting down for a long time or helping for a long time.It, can during being trained to deep learning neural network
First using being largely in different tumble states, different squat down state and different help holds the skeleton models of other object states
Image is trained, and is helped for a long time and is held other objects and can be long-time sustaining wall or help for a long time and hold the states such as railing.Cloud service
The deep learning neural network that training is completed can be sent to human body appearance after the training for completing deep learning neural network by device
State identification sensor, by human body attitude identification sensor locally carry out using.
Step S106 generates report corresponding with the personal information when monitoring the skeleton model is abnormality
Alert information, and it is sent to the Cloud Server, the abnormality includes falling, squat down for a long time or helping for a long time to hold other objects
Body.
After human body attitude identification sensor recognizes human body in monitoring range, so that it may be carried out to the human body real-time
Monitoring, and judge whether the skeleton model of the target body enters abnormality in real time.When the bone for monitoring target body
When model is tumble state, it can determine that target body has been fallen.It is shape of squatting down in the skeleton model for monitoring target body
State is helped when holding other object states, and neural network processor can star timer, holds other to entering state of squatting down or helping
The duration of object state carries out timing, when squatting down duration or the duration for holding other objects being helped to be more than preset duration, that is, can determine
Target body time of squatting down holds the states of other objects too long or in long-time sustaining wall or help, that is, determines that target body is in different
Normal state.
Determining that target body is when in an abnormal state, neural network processor can be according to the predetermined target person
The information that the target body is in abnormality is sent to Cloud Server by the personal information of body, and Cloud Server can will in this way
Exceptional state alarm information timely feedback to the mobile terminal for pre-establishing communication connection.For example, in family's use environment,
Human body attitude identification sensor can be first passed through in advance to record the face for the personnel that needs monitor, while passing through Cloud Server
The corresponding relationship for establishing the personnel of at least one mobile terminal and needs monitoring is determining that it is different that the personnel of needs monitoring occur
When normal state, warning message can be sent to the mobile terminal for pre-establishing connection by Cloud Server.For example, live in old man
In house, can in old man, often multiple human body attitude identification sensors are arranged in movable region, and pass through Cloud Server foundation
The corresponding relationship of human body attitude identification sensor, Cloud Server are connecing in the mobile terminal and house of old man children or staff
After receiving the exceptional state alarm information that human body attitude identification sensor is sent in house, so that it may be sent to warning message always
The mobile terminal of people children or staff timely to find the abnormality of old man, and are in time disposed old man.
In addition, the abnormality in the embodiment of the present application can also include the state that will be fallen ill.Cloud Server can be with
Some diseases with typical skeleton character learn deep learning neural network, for example, for diseases such as apoplexy,
The connection status of general joint and the connection status of normal human can be distinct before morbidity by patient.Certain can be acquired in advance
A little sample skeleton models of the disease before morbidity are trained deep learning neural network, so that deep learning neural network
Have the recognition capability to certain diseases before morbidity, it can be by target body bone by human body attitude identification sensor
The identification of bone model determines that target body whether there is the disease that will be fallen ill, and can determine target person by skeleton model
Body generates prompt information, reminds the caregiver of target body there are when onset risk.
Optionally, the human body attitude identification sensor in the embodiment of the present application can also be to a certain target person in a certain period
The number that body occurs is recorded.For example, nursing place in certain old men, the old man nursed generally has fixed activity rule
Rule.At least 5 times are such as had in a certain position old man one week in nurse floor activity, human body attitude identification sensor will be recorded always
Number of activities in people one week.If the number of activities of the old man are significantly lower than 5 times in a certain week, human body attitude identification sensor
The prompt information for the old man can also be generated, phone is carried out to old man in time by staff or is paid a return visit face to face, with determination
The old man whether there is disease risk.
Optionally, the human body attitude identification sensor in the embodiment of the present application can also preset need to monitor it is all
The information of personnel, and according to face recognition result count each personnel it is movable in monitoring range at the time of, within a certain period
The information such as the number of appearance are directed to personal behavior database so as to establish, so as to carry out subsequent safety and wind
Dangerous early warning.The result that human body attitude identification sensor is calculated can upload Cloud Server and be stored, and guarantee depositing for data
Storage safety.
In another embodiment, it in the place configured with multiple human body attitude identification sensors, can also configure
One or more controlling terminals, controlling terminal can be connected by Cloud Server, configured with one or more aobvious in controlling terminal
The real-time recognition result of multiple human body attitude identification sensors can be shown by display screen, and controlling terminal can also be to multiple
The human bioequivalence result of human body attitude identification sensor is counted, for example, the total number of persons currently recognized, each can be counted
The number that personnel occur in predetermined period.After human body attitude identification sensor generates warning message, Cloud Server can also be incited somebody to action
Warning message is shown by controlling terminal and is reminded with acousto-optic, so that staff has found dangerous situation in time.
Human body attitude identification sensor provided by the embodiments of the present application can be by microwave signal and infrared image, to monitoring
Human body in range carries out behavior monitoring.The indexs such as respiratory rate, the HR Heart Rate of human body can be monitored by microwave signal, together
When combine infrared image the case where can establish human skeleton model, human body can be monitored by skeleton model, falling
, when squatting long, it can timely find and carry out warning reminding.Human body gesture recognition sensor uses embedded mind
Through network processing unit, the detection of face and body state can be quickly finished, which not will record video data, no
It is related to privacy of user, is not required to configure relevant local server, ontology storage equipment, saves system cost of layout.Meanwhile it sensing
Device installation is simple, is the passive monitoring to the non-intervention type of human body, does not change the daily habits of monitored personnel, while can obtain
The more data in family are taken, monitoring accuracy is improved.Human body gesture recognition sensor does not require network bandwidth, no network
Delay, follow-up system upgrading only needs to increase and decrease sensor, and can pass through long-range OTA (Over the Air
Technology, over the air) carry out system function update.
As shown in figure 8, present invention also provides a kind of human body attitude identifying system 100, including the identification of multiple human body attitudes
Sensor 10 and the Cloud Server 20 being connect with the human body attitude identification sensor 10, wherein
The human body attitude identification sensor 10 includes:
Bottom cover and fixed frame, the bottom cover and fixed frame are fixed and are enclosed accommodating chamber；
Circuit board, the circuit board are fixed in the accommodating chamber；
Neural network processor on the circuit board is set；
Microwave transmitting antenna, microwave antenna and the binocular infrared camera being connect with the neural network processor,
The microwave transmitting antenna is for emitting microwave signal, and the microwave antenna is for described in receiving and reflecting through other objects
Microwave signal, the binocular infrared camera is for acquiring infrared image；It is provided on the fixed frame described micro- for fixing
Wave transmitting antenna, the first fixation hole of microwave antenna and the second fixation for fixing the binocular infrared camera
Hole；
Filtering cover, the filtering, which is covered, is fixed on side of the fixed frame far from the accommodating chamber, on the filtering cover
Offer the first through hole to match with first fixation hole；
Baffle, the baffle plate setting offers on the separate fixed frame side of filtering cover, the baffle and institute
State the second through-hole that first through hole matches, planar movement where the baffle can be covered along the filtering, by adjusting described
The overlapping area of second through-hole and the first through hole changes the direction of the launch and receiving direction of the microwave signal；
The Cloud Server 20 is used to establish deep learning neural network for default tumble image, and to the depth
It practises neural network to be trained, so that training of the human body attitude identification sensor 10 according to the deep learning neural network
As a result monitoring human body is carried out abnormality detection.
The above, the only various embodiments of the application, but the protection scope of the application is not limited thereto, it is any
Those familiar with the art within the technical scope of the present application, can easily think of the change or the replacement, and should all contain
Lid is within the scope of protection of this application.Therefore, the protection scope of the application shall be subject to the protection scope of the claim.
Claims (10)
1. a kind of human body attitude identification sensor, which is characterized in that human body gesture recognition sensor includes:
Bottom cover and fixed frame, the bottom cover and fixed frame are fixed and are enclosed accommodating chamber；
Circuit board, the circuit board are fixed in the accommodating chamber；
Neural network processor on the circuit board is set；
Microwave transmitting antenna, microwave antenna and the binocular infrared camera being connect with the neural network processor, it is described
Microwave transmitting antenna is for emitting microwave signal, and the microwave antenna is for receiving the microwave reflected through other objects
Signal, the binocular infrared camera is for acquiring infrared image；It is provided on the fixed frame for fixing the microwave hair
Penetrate antenna, the first fixation hole of microwave antenna and the second fixation hole for fixing the binocular infrared camera；
Filtering cover, the filtering cover are fixed on side of the fixed frame far from the accommodating chamber, open up on the filtering cover
There is the first through hole to match with first fixation hole；
Baffle, the baffle plate setting are covered in the filtering far from the fixed frame side, are offered on the baffle and described the
The second through-hole that one through-hole matches, planar movement where the baffle can be covered along the filtering, by adjusting described second
The overlapping area of through-hole and the first through hole changes the direction of the launch and receiving direction of the microwave signal.
2. human body attitude identification sensor according to claim 1, which is characterized in that filtering cover both ends open up respectively
There is sliding slot, the sliding block to match with the sliding slot is provided on the baffle, the baffle passes through the relatively described cunning of the sliding block
Slot sliding.
3. human body attitude identification sensor according to claim 1, which is characterized in that human body gesture recognition sensor is also
Include:
It is arranged on the circuit board, the communication module being connected with the neural network processor, the communication module is used for
The processing result of the neural network processor is sent to other external equipments.
4. human body attitude identification sensor according to claim 1, which is characterized in that human body gesture recognition sensor is also
Include:
RF transmitter, the RF transmitter are arranged in side of the fixed frame far from the accommodating chamber, and with
The neural network processor is connected, for launching outward infrared ray.
5. human body attitude identification sensor according to claim 1, which is characterized in that be provided on filtering cover and institute
The filter that binocular infrared camera matches is stated, the filter is used to filter out the environment light into the binocular infrared camera
Line.
6. a kind of human posture recognition method, which is characterized in that applied to human body appearance described in claim 1 to 5 any one
State identification sensor, the human body attitude identification sensor pre-establish the communication connection with Cloud Server, the Cloud Server
The deep learning neural network is trained for establishing deep learning neural network, and using default tumble image, it should
Human posture recognition method includes:
According to the microwave signal received, determine in monitoring range with the presence or absence of human body；
When determining in the monitoring range there are when human body, according to the infrared image, determine target body in preset coordinate system
Interior human body depth information；
Recognition of face is carried out to the target body, determines the personal information of the target body；
The bone of the target body is drawn according to the human body depth information based on default human body prediction model
Model；
The skeleton model is monitored using the deep learning neural network that training terminates；
When monitoring the skeleton model is abnormality, warning message corresponding with the personal information is generated, and send
To the Cloud Server, the abnormality includes falling, squat down for a long time or helping for a long time to hold other objects.
7. human posture recognition method according to claim 6, which is characterized in that according to the microwave signal received, really
Determine whether to have in monitoring range the step of human body include:
According to the microwave signal received, determine whether there are objects moving in detection range；
When detecting that there are objects moving, the HR Heart Rate and respiratory rate of mobile object are determined based on the microwave signal；
When the HR Heart Rate or respiratory rate that determine mobile object meet preset condition, determining has human body in detection range.
8. human posture recognition method according to claim 6, which is characterized in that when determining that there are human bodies in monitoring range
When, according to the infrared image, the step of determining human body depth information of the target body in preset coordinate system, includes:
Location information based on same target point in the infrared image on each infrared camera image plane, described in calculating
D coordinates value of the target point in the preset coordinate system；
The D coordinates value is converted into the coordinate value in two dimensional image, determines multiple target points in the two dimensional image
Depth information, the human body depth information as the target body.
9. human posture recognition method according to claim 6, which is characterized in that the default human body prediction model
In include the multiple crucial skeleton points of human body connection relationship, based on default human body prediction model, according to the human body
Depth information, the step of drawing the skeleton model of the target body include:
According to the infrared image, at least one crucial skeleton point of the target body is determined；
The mesh is obtained according to the connection relationship of the multiple crucial skeleton point and the human body depth information, drafting
Mark human skeleton model.
10. a kind of human body attitude identifying system, which is characterized in that including human body attitude described in claim 1 to 5 any one
Identification sensor, and the Cloud Server connecting with the human body attitude identification sensor, the Cloud Server are used for for pre-
If tumble image establishes deep learning neural network, and is trained to the deep learning neural network, so that the human body
Gesture recognition sensor carries out abnormality detection monitoring human body according to the training result of the deep learning neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910443623.9A CN110047249A (en) | 2019-05-27 | 2019-05-27 | Human body attitude identification sensor, human posture recognition method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910443623.9A CN110047249A (en) | 2019-05-27 | 2019-05-27 | Human body attitude identification sensor, human posture recognition method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110047249A true CN110047249A (en) | 2019-07-23 |
Family
ID=67283617
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910443623.9A Pending CN110047249A (en) | 2019-05-27 | 2019-05-27 | Human body attitude identification sensor, human posture recognition method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110047249A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112132055A (en) * | 2020-09-24 | 2020-12-25 | 天津锋物科技有限公司 | Home security management method and system |
| WO2021174414A1 (en) * | 2020-03-03 | 2021-09-10 | 苏州七星天专利运营管理有限责任公司 | Microwave identification method and system |
| CN114267153A (en) * | 2021-12-29 | 2022-04-01 | 苏州英鹏信息科技有限公司 | Home safety monitoring management system |
| CN115690897A (en) * | 2022-08-31 | 2023-02-03 | 北京夕阳无忧科技有限公司 | Accidental behavior processing method, device, equipment and storage medium for preventing privacy leakage |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090040099A1 (en) * | 2006-10-12 | 2009-02-12 | Young James E | Coaxial bi-modal imaging system for combined microwave and optical imaging |
| CN204087403U (en) * | 2014-09-02 | 2015-01-07 | 深圳市美安科技有限公司 | A kind of structure of intrusion detector |
| CN105741492A (en) * | 2016-04-27 | 2016-07-06 | 华中科技大学 | Detecting and alarming system and method thereof |
| CN105856187A (en) * | 2016-05-05 | 2016-08-17 | 宁波华狮智能科技有限公司 | Microwave security inspection robot capable of automatically identifying hazardous articles |
| CN108827899A (en) * | 2018-05-24 | 2018-11-16 | 江苏大学 | Infrared and microwave complex probe imaging platform and its imaging method |
| CN209785215U (en) * | 2019-05-27 | 2019-12-13 | 中侨健康管理（上海）有限公司 | Human body posture recognition sensor and human body posture recognition system |
-
2019
- 2019-05-27 CN CN201910443623.9A patent/CN110047249A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090040099A1 (en) * | 2006-10-12 | 2009-02-12 | Young James E | Coaxial bi-modal imaging system for combined microwave and optical imaging |
| CN204087403U (en) * | 2014-09-02 | 2015-01-07 | 深圳市美安科技有限公司 | A kind of structure of intrusion detector |
| CN105741492A (en) * | 2016-04-27 | 2016-07-06 | 华中科技大学 | Detecting and alarming system and method thereof |
| CN105856187A (en) * | 2016-05-05 | 2016-08-17 | 宁波华狮智能科技有限公司 | Microwave security inspection robot capable of automatically identifying hazardous articles |
| CN108827899A (en) * | 2018-05-24 | 2018-11-16 | 江苏大学 | Infrared and microwave complex probe imaging platform and its imaging method |
| CN209785215U (en) * | 2019-05-27 | 2019-12-13 | 中侨健康管理（上海）有限公司 | Human body posture recognition sensor and human body posture recognition system |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021174414A1 (en) * | 2020-03-03 | 2021-09-10 | 苏州七星天专利运营管理有限责任公司 | Microwave identification method and system |
| CN112132055A (en) * | 2020-09-24 | 2020-12-25 | 天津锋物科技有限公司 | Home security management method and system |
| CN114267153A (en) * | 2021-12-29 | 2022-04-01 | 苏州英鹏信息科技有限公司 | Home safety monitoring management system |
| CN114267153B (en) * | 2021-12-29 | 2024-03-26 | 苏州英鹏信息科技有限公司 | Household safety monitoring management system |
| CN115690897A (en) * | 2022-08-31 | 2023-02-03 | 北京夕阳无忧科技有限公司 | Accidental behavior processing method, device, equipment and storage medium for preventing privacy leakage |
| CN115690897B (en) * | 2022-08-31 | 2023-10-20 | 北京夕阳无忧科技有限公司 | Unexpected behavior processing method, device and equipment for preventing privacy leakage and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110047249A (en) | Human body attitude identification sensor, human posture recognition method and system | |
| US11875656B2 (en) | Virtual enhancement of security monitoring | |
| US11062589B2 (en) | Drone-augmented emergency response services | |
| ES2551099T3 (en) | Position Tracking System | |
| US11727518B2 (en) | Systems and methods for location fencing within a controlled environment | |
| KR101210783B1 (en) | Node management system and node managing method using sensing system | |
| KR20150092545A (en) | Warning method and system using prompt situation information data | |
| KR101514696B1 (en) | Safety Accident Management System Of Bridge | |
| CN108638082A (en) | Security robot system based on Internet of Things | |
| EP3374925B1 (en) | Image processing system | |
| CN110647797B (en) | Visitor detection method and device | |
| US20220139199A1 (en) | Accurate digital security system, method, and program | |
| CN106444587B (en) | A kind of system and method that human life's body characteristics are monitored based on unmanned plane | |
| CN106850796A (en) | Wireless tracking monitoring method and system based on Internet of Things | |
| CN110974158B (en) | Community patient identification and distress calling method, system method and system based on deep learning | |
| CN109378067A (en) | Terminal control method and healthy and safe educational management all-in-one machine | |
| US11729372B2 (en) | Drone-assisted sensor mapping | |
| JP6565061B2 (en) | Viewing system | |
| CN110675595B (en) | Fall detection system based on vision and artificial intelligence | |
| KR102211723B1 (en) | Smart Management System | |
| CN110177254A (en) | A kind of wisdom building site system for managing video and its control method | |
| US10915733B1 (en) | Temperature determination from brain thermal tunnel | |
| Lock et al. | Improving operational processes for COVID-19 ready smart campus | |
| CN106448058A (en) | Usage method of intelligent nursing terminal | |
| CN111666794A (en) | Pet monitoring method and pet monitoring 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 |