CN110245556A - A kind of detection method, device, system and the storage medium of the state of leaving home of being in - Google Patents
A kind of detection method, device, system and the storage medium of the state of leaving home of being in Download PDFInfo
- Publication number
- CN110245556A CN110245556A CN201910371573.8A CN201910371573A CN110245556A CN 110245556 A CN110245556 A CN 110245556A CN 201910371573 A CN201910371573 A CN 201910371573A CN 110245556 A CN110245556 A CN 110245556A
- Authority
- CN
- China
- Prior art keywords
- event
- state
- currently
- leaving home
- type
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 81
- 230000033001 locomotion Effects 0.000 claims abstract description 60
- 238000012790 confirmation Methods 0.000 claims description 26
- 238000000034 method Methods 0.000 claims description 26
- 238000012545 processing Methods 0.000 claims description 7
- 238000009434 installation Methods 0.000 abstract description 6
- 230000004044 response Effects 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 9
- 230000009286 beneficial effect Effects 0.000 description 8
- 238000009826 distribution Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 238000012163 sequencing technique Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000001914 filtration Methods 0.000 description 3
- 230000003796 beauty Effects 0.000 description 2
- 230000002354 daily effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000000474 nursing effect Effects 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000009017 pursuit movement Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000007873 sieving Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
- G06V20/42—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items of sport video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/48—Matching video sequences
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses detection method, device, system and the storage mediums of a kind of state of leaving home of being in, including obtaining the motion profile of the moving target after carrying out motion target tracking to sequence of video images;According to the motion profile, the type of current kinetic event is determined;According to the type for the current kinetic event determined, the judgement for the state of leaving home that be in.It by using the solution of the present invention, not only reduces purchase input cost, save installation space, but also also improve detection system installs and uses convenience;In addition, the detection for the state of leaving home that carries out being in by using the embodiment of the present invention, accuracy in detection is high and fast response time.The present invention can be widely applied in daily life as detection method, device, system and the storage medium of a kind of state of leaving home of being in.
Description
Technical field
The present invention relates to data processing technique more particularly to a kind of detection method for the state of leaving home of being in, device, system and
Storage medium.
Background technique
The case where stepping into aging, old solitary people and Empty nest elderly gradually with today's society crowd is also more and more, together
Shi Zuowei Empty nest elderly children be in old man, neighbourhood committee's (or other unit) to old solitary people leave home situation understanding contradiction
Also it becomes increasingly conspicuous, therefore how daily nurse and understanding is carried out to the case where old solitary people and Empty nest elderly, this becomes people
One of (especially children and neighbourhood committee) concern.
Develop higher and higher now with scientific and technological level, various consumer electronic product popularization and applications are in various
Place, such as education, media, amusement, shopping etc., but the application in endowment nurse field still lacks.It is common to only at present
It occupies old man and nursing system that Empty nest elderly is nursed also is only resided within by installing multiple traditional detector (human body senses
Answer device) Lai Shixian old man's state-detection for being in or leaving home.However, needs are in for traditional nursing system
Multiple traditional peoplesniffers are installed inside front yard to be just able to achieve, the input cost for not only having equipment purchase so excessively high is asked
Topic, and it is different in home environment enough installation spaces are surely provided, and influence the aesthetic feeling of household internal environment.
Summary of the invention
In view of this, the purpose of the embodiment of the present invention is that providing detection method, device, the system of a kind of state of leaving home of being in
And storage medium, purchase input cost can be reduced, and save installation space.
In a first aspect, the embodiment of the invention provides a kind of detection methods of state of leaving home of being in, comprising the following steps:
Sequence of video images is obtained from the video that shooting obtains;
The motion profile of the moving target is obtained after carrying out motion target tracking to sequence of video images;
According to the motion profile, the type of current kinetic event is determined, wherein the type of the current kinetic event
Including event type and the event type of going out of entering;
According to the type for the current kinetic event determined, the judgement for the state of leaving home that be in.
Further, described according to the motion profile, the step for determining the type of current kinetic event comprising:
Determine that the motion profile enters the sequence in several detection image regions;
According to the sequence, determine that the type of current kinetic event is enter event type or event type of going out.
Further, the type for the current kinetic event that the basis is determined, the judgement of the state of leaving home that be in this
Step comprising:
When the type for determining current kinetic event is to enter event type, then the record number for the event of entering is added 1, and
Determine that current state is state of being in;
When the type for determining current kinetic event is event type of going out, then after the record number for the event of going out being added 1,
It is calculated and is worked as according to the record number of current resident population, the record number of current event of entering and current event of going out
Before be in number, according to the calculated number of being currently in, the judgement for the state of leaving home that be in.
Further, described according to the calculated number of being currently in, carry out this step of the judgement for the state of leaving home of being in
Suddenly comprising:
When it is described it is calculated be currently in number greater than 0, then determine that current state is state of being in;
When it is described it is calculated be currently in number equal to 0, then determine that current state is state of leaving home;
When it is described it is calculated be currently in number less than 0, then the calculated current number of being in is set to 0, and
Determine that current state is state of leaving home.
Further, the currently resident population is determined by following steps:
The record number and the record for event of currently going out of event of once currently entering are calculated every prefixed time interval
Absolute difference between number;
When reaching preset time point, then numerical value is that maximum difference is exhausted from choosing in several obtained absolute differences
To value;
When the absolute difference selected is identical as preceding primary calculated resident population, it is determined that the choosing
The absolute difference of taking-up is currently to reside population;
When the absolute difference selected and preceding primary calculated resident population be not identical, then confirmation is sent
It requests to client, and determines currently to reside population according to the confirmation message returned by client is received.
Further, the basis currently resides population, the record number for event of currently entering and thing of currently going out
The record number of part calculates the step for being currently in number comprising:
Calculate the difference between the record number for event of currently entering and the record number for event of currently going out;
The number of being currently in is obtained after the calculated difference is added with the currently resident population.
Second aspect, the embodiment of the invention provides a kind of detection systems of state of leaving home of being in, comprising:
Module is obtained, for obtaining sequence of video images from the video that shooting obtains;
Processing module, for obtaining the movement rail of the moving target after carrying out motion target tracking to sequence of video images
Mark;
Determining module, for determining the type of current kinetic event according to the motion profile, wherein described current
The type of motion event includes enter event type and event type of going out;
Judgment module, the judgement for the state of leaving home that according to the type of current kinetic event determined, carries out being in.
The third aspect, the embodiment of the invention provides a kind of detection devices of state of leaving home of being in, comprising:
At least one processor;
At least one processor, for storing at least one program;
When at least one described program is executed by least one described processor, so that at least one described processor is realized
A kind of detection method of the state of leaving home of being in.
Fourth aspect, the embodiment of the invention provides a kind of storage mediums, wherein it is stored with the executable instruction of processor,
The executable instruction of the processor is when executed by the processor for executing a kind of detection method of state of leaving home of being in.
5th aspect, the embodiment of the invention provides a kind of detection system of state of leaving home of being in, including controller and
For shooting the camera of video, the controller is connect with the camera;The controller includes:
At least one processor;
At least one processor, for storing at least one program;
When at least one described program is executed by least one described processor, so that at least one described processor is realized
A kind of detection method of the state of leaving home of being in.
Compared with prior art, the embodiment of the present invention is by obtaining sequence of video images from the video that shooting obtains
Afterwards, the motion profile of the moving target is obtained after carrying out motion target tracking to sequence of video images, then according to the fortune
Dynamic rail mark determines the type of current kinetic event, wherein the type of the current kinetic event include enter event type and
It gos out event type, the last type according to the current kinetic event determined, the judgement for the state of leaving home that be in, it is seen that logical
It crosses and uses the solution of the present invention, it is only necessary to the device for shooting video image be installed, not only reduce purchase investment in this way
Cost saves installation space, and also improve detection system installs and uses convenience;In addition, by using of the invention real
Example is applied come the detection for the state of leaving home that be in, accuracy in detection is high and fast response time.
Detailed description of the invention
Fig. 1 is a kind of step flow diagram of the detection method for the state of leaving home of being in of the embodiment of the present invention;
Fig. 2 is the first schematic diagram of the sequence that motion profile enters several detection image regions;
Fig. 3 is the second schematic diagram of the sequence that motion profile enters several detection image regions;
Fig. 4 be in leave home state judgement an idiographic flow schematic diagram;
Fig. 5 is the idiographic flow schematic diagram that current resident population determines;
Fig. 6 is a kind of first structure block diagram of the detection system for the state of leaving home of being in of the embodiment of the present invention;
Fig. 7 is a kind of second structural block diagram of the detection system for the state of leaving home of being in of the embodiment of the present invention.
Specific embodiment
The present invention is described in further detail in the following with reference to the drawings and specific embodiments.In for the examples below
Number of steps is arranged only for the purposes of illustrating explanation, does not do any restriction to the sequence between step, each in embodiment
The execution sequence of step can be adaptively adjusted according to the understanding of those skilled in the art.In addition, for following implementation
Several, are expressed as at least one described in example.
It should be noted that unless otherwise specified, when a certain feature referred to as " fixation ", " connection " are in another feature,
It can directly fix, be connected to another feature, and can also fix, be connected to another feature indirectly.In addition, this
The descriptions such as the upper and lower, left and right used in open are only the mutual alignment pass relative to each component part of the disclosure in attached drawing
For system.The "an" of used singular, " described " and "the" are also intended to including most forms in the disclosure, are removed
Non- context clearly expresses other meaning.In addition, unless otherwise defined, all technical and scientific terms used herein
It is identical as the normally understood meaning of those skilled in the art.Term used in the description is intended merely to describe herein
Specific embodiment is not intended to be limiting of the invention.Term as used herein "and/or" includes one or more relevant
The arbitrary combination of listed item.
It will be appreciated that though various elements, but this may be described using term first, second, third, etc. in the disclosure
A little elements should not necessarily be limited by these terms.These terms are only used to for same type of element being distinguished from each other out.For example, not departing from
In the case where disclosure range, first element can also be referred to as second element, and similarly, second element can also be referred to as
One element.The use of provided in this article any and all example or exemplary language (" such as ", " such as ") is intended merely to more
Illustrate the embodiment of the present invention well, and unless the context requires otherwise, otherwise the scope of the present invention will not be applied and be limited.
As shown in Figure 1, the embodiment of the invention provides a kind of detection methods of state of leaving home of being in, comprising the following steps:
S101, sequence of video images is obtained from the video that shooting obtains.
Specifically, obtained video is shot for described, generallys use camera and/or the video with camera is clapped
Device is taken the photograph to be shot.And for the sequence of video images of the acquisition, institute in the video that obtains to shooting can be passed through
The video frame images that include just obtain after carrying out effective video frame optical sieving, can also pass through institute in the video that obtains to shooting
The video frame images for including just obtain after carrying out image preprocessing, wherein described image pretreatment may include but be not limited to figure
As filtering, going hot-tempered, morphological images processing etc., sequence of video images directly can also be obtained from the video that shooting obtains.Its
In, the sequence of video images refers to: video frame images included in video are arranged successively in chronological order, therefore
Several video frame images successively acquired in chronological order from video, the as sequence of video images.
S102, the motion profile of the moving target is obtained after carrying out motion target tracking to sequence of video images.
Specifically, the motion profile of the moving target can use motion target tracking technology and carry out pursuit movement mesh
It is marked on generated motion profile in image sequence.Generally for the target following technology, the process realized can are as follows: first right
Video image carries out foreground moving object region and distinguishes with background area, to obtain the foreground moving object in video image
The foreground moving object region in region, several video images for then being included to sequence of video images tracks, thus
Obtain the motion profile of the moving target.In addition, the determination for the foreground moving object region, can directly by from
The foreground area divided in video image is preceding then due in video image as the foreground moving object region
Scape target not necessarily may be people, it may be possible to animal or other movement objects, therefore can by segmentation obtain before
Scene area carries out the qualifications such as center present position of size, region, thus to determine required foreground moving object area
Domain, this can according to the actual situation and/or demand is configured, and does not do excessive restriction herein.
S103, according to the motion profile, determine the type of current kinetic event, wherein the current kinetic event
Type include entering event type and to go out event type.
Specifically, if motion profile shows that moving target is moved to outdoors from family, then for event of going out, and if transporting
Dynamic rail mark shows moving target from being moved to inside family outdoors, then being then event of entering, therefore, just by the motion profile
It can determine that out that current kinetic event is event or the event of going out of entering.
S104, according to the type of current kinetic event determined, the judgement for the state of leaving home that be in.
In embodiments of the present invention, the state of being in refers to the state of at least 1 people of family, and the shape of leaving home
State is then the state that there is no one at home.In addition, state of being such as in or state of leaving home are also preferable when determining current state
It is pushed to client in real time, such user just can real time inspection recognize the state of leaving home of being in of main family.
As it can be seen that the embodiment of the present invention be in leave home condition detection method can be by the view acquired from video
After frequency image sequence carries out motion target tracking, according to the motion profile, the type of current kinetic event is determined, wherein
The type of the current kinetic event includes enter event type and event type of going out, then according to the current kinetic determined
The type of event, the judgement for the state of leaving home that be in, such user just only need to buy video equipment can realize be in from
The detection of family's state is just able to achieve without installing multiple conventional body's detectors, dramatically saves the investment of equipment purchase
Cost, and installation space is saved, reduce the influence to home environment visual sense of beauty;Furthermore through the embodiment of the present invention
For detection method come the detection for the state of leaving home that be in, accuracy in detection is high, and it is fast to detect response.
In some preferred embodiments, described according to the motion profile, determine current kinetic event type this
Step S103 comprising:
S1031, determine that the motion profile enters the sequence in several detection image regions;
S1032, according to the sequence, determine that the type of current kinetic event is enter event type or event class of going out
Type.
Specifically, as shown in Fig. 2, several described detection image regions are laid on the video images, i.e., in video image
Several detection image regions of upper determination, and each detection image region is all had into only one label, such as 1,2,
3 ..., n, the label can be used for carrying out identification differentiation to detection image region, then detect the motion profile in entire mistake
It is to be introduced into which detection image region, and when entering a detection image region, just record by motion profile in journey
The label in the detection image region of entrance, the time sequencing or sequencing of the label being recorded in this way are just movement rail
Mark enters the sequence in several detection image regions.So, mounted camera against the position on doorway and can be clapped
It takes the photograph from the region outside transition door mouth inside family, so successively lays in several described detection image regions from left to right, this
If the region sequencing (i.e. the sequencing of trigger area) that sample motion profile is entered be 1,2,3 ... n, and this indicate
For the event of entering, then if the region sequencing that motion profile is entered be n, n-1 ..., 2,1, this is expressed as going out
Event.As it can be seen that determining the type of current kinetic time in this way, accuracy in detection and place can be further improved
Manage efficiency.
In addition, can be completely not identical, not exactly the same for the size and shape in several detection image regions
Or it is identical, this can be configured with demand according to the actual situation.
In some preferred embodiments, the type for the current kinetic event that the basis is determined, carries out shape of leaving home of being in
The step for judgement of state S104 comprising:
S1041, when the type for determining current kinetic event is to enter event type, then by the record number for the event of entering
IN1 adds 1, and determines that current state is state of being in;And/or
S1042, when the type for determining current kinetic event is to go out event type, then by the record number for the event of going out
After OUT1 adds 1, according to the note of current resident population N, the record number IN1 of current event of entering and current event of going out
Record number OUT1 calculates the number M that is currently in, and according to the calculated number M that is currently in, carries out the state of leaving home of being in
Judgement.Due in embodiments of the present invention, the state of being in refers to the state of at least 1 people of family, and it is described from
Family's state is then the state that there is no one at home, thus using current resident population N, currently enter event record number with
And the record number for event of currently going out calculates number of being currently in (i.e. somebody's number inside current family), it can be accurately
Number of being currently in is calculated, to further improve accuracy in detection.
In some embodiments, the basis currently reside population, currently enter event record number and work as
Before go out event record number calculate currently be in number the step for comprising:
Calculate the difference between the record number IN1 for event of currently entering and the record number OUT1 for event of currently going out
Value, i.e. IN1-OUT1;
The number of being currently in is obtained after the calculated difference is added with the currently resident population.
In some preferred embodiments, described according to the calculated number of being currently in, carry out the state of leaving home of being in
Judgement the step for comprising:
When it is described it is calculated be currently in number M greater than 0, then determine that current state is state of being in;
When it is described it is calculated be currently in number M equal to 0, then determine that current state is state of leaving home;
When it is described it is calculated be currently in number M less than 0, then the calculated current number of being in is set to 0, and
Determine that current state is state of leaving home.As it can be seen that current determining current state is to leave home when as long as calculated M is less than or equal to 0
State.Wherein, the case where M is less than 0, it may be possible to have thief or other people not from doorway enter, at this time should also further by
The state of leaving home is pushed to except client, it is also necessary to be issued prompting message to client, be used the user of client can to enable
It is checked and is confirmed, can also achieve the effect that security protection in this way.
In some preferred embodiments, the currently resident population N is determined by following steps:
The record number IN1 and the record for event of currently going out of event of once currently entering are calculated every prefixed time interval
Absolute difference I between number OUT1, i.e. I=ABS (OUT1-IN1);For example, being opened from 5 points of every morning (i.e. 5:00a.m)
Begin, was calculated every 30 minutes between the record number IN1 for event of once currently entering and the record number OUT1 for event of currently going out
Absolute difference, then experience 24 as a child, i.e., 1 day, then obtained 48 absolute differences, I at this time1、I2、I3、……、
I48, wherein I1It is expressed as the 1st calculated absolute difference.As it can be seen that if prefixed time interval is different, and start to calculate
Time point and end calculate time point setting difference, then the number for then obtaining absolute difference then can be different, i.e. IkIt indicates
For kth time calculated absolute difference, then k=1,2,3 ..., K, the value of K can be according to different the case where actual set
And it is different;
When reaching preset time point, then numerical value is that maximum difference is exhausted from choosing in several obtained absolute differences
To value;For example, from I1、I2、I3、……、I48In, select the maximum absolute difference of numerical value, such as I35;In addition it is also possible to first
X period is preset, maximum absolute difference in each period is then first found out, then again from the absolute difference acquired
In select the maximum absolute difference of numerical value;
As absolute difference (such as I selected35) identical as preceding primary calculated resident population, it is determined that
The absolute difference selected is currently to reside population;
When the absolute difference selected and preceding primary calculated resident population be not identical, then confirmation is sent
It requests to client, and determines currently to reside population according to the confirmation message returned by client is received.
Specifically, it since the resident size of population of one family may generate variation because of various factors, then needs
Population is resided to family to confirm, and each finally confirms that obtained resident population can be stored in database
In, in this way as absolute difference (such as I selected35) and preceding primary calculated resident population Np-1It is identical, it is determined that institute
State absolute difference (such as I selected35) it is currently to reside population NpIf but I35With Np-1Not identical, then needing will be true
Recognize request and be sent to client, wherein the confirmation request contains I35With Np-1Not identical this case information, to enable client
The display of the confirmation request is carried out after receiving this confirmation request, and can further be issued prompt and (such as be enabled client
End vibration issues voice, prompt tone), remind user to check that after user checks and confirms situation, client can then be returned really
Then the information (the i.e. described confirmation message) recognized is determined current normal according to the confirmation message returned by client is received
In population.Wherein, the confirmation message that may include that there is something special in the confirmation message, at this point, then by I35As Np;Alternatively,
It is can wrap in the confirmation message containing the resident population confirmed, then will confirm that and reside population included in information
As Np;Finally, will confirm that good current resident population NpIt is stored in database profession.
The embodiment of the invention also provides a kind of detection method of state of leaving home of being in, the step of specifically including following institute
Show.
Step S201, the type of current kinetic event is determined.Wherein the step S201 includes following sub-step.
S2011, video capture is installed at home and carried out to camera, video figure is obtained from the video that shooting obtains
As sequence, that is, acquire several video images.
S2011, the motion profile of the moving target is obtained after carrying out motion target tracking to sequence of video images.
In the present embodiment, the specific embodiment of the motion target tracking is shown in steps are as follows.
1) mixture Gaussian background model, is established, then video image is carried out using the mixture Gaussian background model
Division between moving target and background.
It is close to one probability of the distribution maintaining of pixel point value each in scene for the mixed Gauss model algorithm
Function is spent, the value of specific pixel each in scene is gone to be fitted with several (generally taking 3~5) Gaussian Profiles, each
The case where Gaussian Profile has the weight of oneself, therefore it can handle multi-modal background distributions.On the other hand, mixed Gauss model
It is to parameterize, the parameter meeting adaptive updates of model, and do not have to cache past video frame, in this way in the mistake of algorithm process
Cheng Zhong can go the weight, mean value and variance that update each Gaussian Profile of each pixel with the arrival of new images.
Firstly, certain specific pixel point pixel value is X in moment ttProbability be:
Wherein, K illustrates the number of selected distribution, is mainly determined by hardware computing capability and available memory,
And in the present embodiment, K is preferably chosen for 3~5.ωn,tIt is the weight in n-th of Gauss model component of moment t, andIt is Gaussian probability-density function, specific as follows shown:
Wherein, μ heren,tWithThe respectively mean value and variance of n-th of Gauss model component of t moment.And it each newly arrives
The pixel value X cometK Gaussian Profile corresponding with it goes to match.If XtIn λ times of standard deviation of some distribution, it may be assumed that
|Xt-μn,t|<λσn,t
It is denoted herein as XtThe distribution is matched;λ is deviation factor.To the model component matched, its parameter can
It is updated by following equation:
ωn,t=(1- α) ωn,t-1+α
μt=(1- ρ) μt-1+ρXt
Herein, α and ρ is model component weight learning rate and model parameter learning rate respectively, and α can be defined by the user, i.e., certainly
Definition.It, can will be with if currently any one Gauss model component of the pixel value that newly arrives and the pixel does not all match
Current observation value XtFor mean value, given the larger value is that initial variance and an initial lower weight are new to construct one
Model replaces the model component minimum when previous belief (being defined as ω/σ).Remaining model component keeps original mean value side
Difference is constant, but their weight can decay by following equation:
ωn,t=(1- α) ωn,t-1
Then, in order to determine XtIt whether is foreground moving pixel, it can be first according to model confidence (ω/σ) to the pixel
K Gauss model component be ranked up, the forward component that sorts has lower variance and higher weight, they can be well
Reflect the characteristic of background.Therefore B distribution of front can be chosen as background model, wherein the value of B should meet following criterion:
Wherein, T is the measurement proportion threshold value of background, can specifically be adjusted according to scenario, and general value is 0.5
~1.If XtWhen matching any one of B distribution distribution, then it can determine that current XtFor background pixel, it is on the contrary then before being
Scape moves pixel.
2), divided from each video image that sequence of video images is included using above-mentioned motion target tracking technology
After belonging to the image-region of moving target out, then movement mesh is carried out according to the image-region for several moving targets being partitioned into
Target tracking, to obtain the motion profile of moving target.
It specifically, for the target following, is passed through between the target that sequence image different moments detect
The connection that certain features of target are established in time.And in the present embodiment, the connection between moving target is using more
Target following correlation technology realizes, then the trace flow realized based on the association is shown in steps are as follows.
STEP1, tracking, which start and arrange, has captured object chain.The specific sub-step that the STEP1 is included is as follows.
If STEP1.1, current frame number are first frame, then all targets detected are considered as fresh target, setting
One initial confidence level.
If STEP1.2, current frame number are not first frame, then checking the object listing captured.
STEP1.3, linear filtering is done to obtained target, to predict it in the position of present frame.Wherein, the position of target
It is indicated, and enabled with its connected region boundary rectangle centerFor position of i-th of target in k-th frame.
STEP2, target association.For this step, establishes a cost function matrix mainly to record and captured chain
In in each target and present frame detection object chain each target cost function value, while establishing an incidence matrix and determining mesh
Mark optimal corresponding relationship.For i-th of target in capture chain, can be found in current detection chain so that cost function E
(i, j) the smallest detection target j.Further compare centroid distance d (i, j) and a distance threshold dT simultaneously.
If STEP2.1, d (i, j) < dT, it is determined that j-th currently detected of target is to have captured i-th of target in chain
Temporal subsequent correspondence, and increase by i-th and captured the confidence level of target and updated its characterisitic parameter.
If STEP2.2, d (i, j) > dT, it is determined that captured for i-th target do not found in present frame it is temporal after
It is continuous to correspond to.So its confidence level can reduce, and if its location of already close to scene edge, then indicating it
Scene has been walked out, then it should have been removed from having captured in chained list.The numerical value after if its confidence level reduces is still higher, and
In the scape area of observation coverage on the scene of its position, then then still going to predict its next frame position using linear filtering described in STEP1.3.
STEP2.3, for currently detected target, if suitable associated objects are not found in object chain having captured,
It is then added to as fresh target and has captured target chained list, and assign an initial confidence level.
STEP3, next frame video is handled until image sequence terminates according to the process of STEP1 and STEP2.
As it can be seen that being capable of determining that the motion profile of the moving target in this image sequence by using aforesaid way.
S2013, according to the motion profile, determine the type of current kinetic event, wherein the current kinetic event
Type include entering event type and to go out event type.
Specifically, the step S2013 includes:
S20131, determine that the motion profile enters the sequence in several detection image regions;
S20132, according to the sequence, determine that the type of current kinetic event is event type or the event of going out of entering
Type.
Specifically, as shown in figure 3, two detection image regions are arranged in the present embodiment in video pictures, respectively marked as
1,2, then setting area triggering is regular, as 1-2 (after first triggering/entering detection image region 1, then triggers/enter detection figure
As region 2), this is expressed as entering, and 2-1 (after first triggering/entering detection image region 2, then trigger/enter detection image
Region 1), this is then expressed as going out.So determine that moving target triggers/enter the detection figure according to the motion profile determined
When as region, then its triggering/entrance region labeling is recorded, if the sequence of then label record meets the sequence of going out of setting,
It is then recorded as the event of going out, if meeting the sequence of entering of setting, is recorded as the event of entering.
Step S202, according to the type of current kinetic event determined, the judgement for the state of leaving home that be in, such as Fig. 4
It is shown.Wherein the step S202 includes following sub-step.
S2021, when the type for determining current kinetic event is to enter event type, then by the record number for the event of entering
IN1 adds 1, and determines that current state is state of being in.
S2022, when the type for determining current kinetic event is to go out event type, then by the record number for the event of going out
After OUT1 adds 1, according to the note of current resident population N, the record number IN1 of current event of entering and current event of going out
Record number OUT1 calculates the number M that is currently in, and according to the calculated number M that is currently in, carries out the state of leaving home of being in
Judgement.
Wherein, the basis currently reside population M, currently enter event record number IN1 and currently go out
The record number OUT1 of event calculates the step for being currently in number M comprising:
Calculate the difference between the record number for event of currently entering and the record number for event of currently going out;
The number of being currently in is obtained after the calculated difference is added with the currently resident population.It can
See, in the present embodiment, M=N+IN1-OUT1.
Then, for it is described according to it is described it is calculated be currently in number M, the judgement of the state of leaving home that be in this
Step comprising:
When it is described it is calculated be currently in number M greater than 0, then determine that current state is state of being in;
When it is described it is calculated be currently in number M equal to 0, then determine that current state is state of leaving home;
When it is described it is calculated be currently in number M less than 0, then the calculated current number of being in is set to 0, and
Determine that current state is state of leaving home.
In addition, being preferably determined, the currently resident population N such as Fig. 5 institute by following steps
Show:
The record number IN1 and the record for event of currently going out of event of once currently entering are calculated every prefixed time interval
Absolute difference between number OUT1;In the present embodiment, the prefixed time interval is 30 minutes;
When reaching preset time point, then numerical value is that maximum difference is exhausted from choosing in several obtained absolute differences
To value;In the present embodiment, starting the time point calculated is 3:00 AM, and the time point for terminating to calculate then is second day icepro
3 points of morning, it is in this way 1 period, then continuously has chosen 3 periods, that is to say, that have chosen 3 day time, then calculate
Every day, maximum absolute difference then selected numerical value most from 3 maximum absolute differences that this is calculated for 3 days again
Big absolute difference, that is to say, that 48 absolute differences can be obtained daily, then chosen from this 48 absolute differences
Numerical value is maximum absolute difference Imax out, and 2 days then to get off also so operation then obtains 3 Imax, then at this time
It is maximum absolute difference that the absolute difference corresponding to this 3 days, which selects numerical value,;
When the absolute difference selected is identical as preceding primary calculated resident population, it is determined that the choosing
The absolute difference of taking-up is currently to reside population;
When the absolute difference selected and preceding primary calculated resident population be not identical, then confirmation is sent
Request is to client to determine currently to reside population for reminding, and according to the confirmation message returned by client is received
Sum.Since the resident size of population of one family may generate variation because of various factors, then need normal to family
Confirmed in population, and each finally confirms that obtained resident population can store in the database, it is elected in this way
Absolute difference (such as I of taking-up35) and preceding primary calculated resident population Np-1It is identical, it is determined that described to select
Absolute difference (such as I35) it is currently to reside population NpIf but I35With Np-1It is not identical, then need will confirm that request is sent
To client, wherein the confirmation request contains I35With Np-1Not identical this case information, to enable client receive this
The display of the confirmation request is carried out after one confirmation request, and can further be issued prompt and (such as be enabled client vibration, issue
Voice, prompt tone), remind user to check that after user checks and confirms situation, client can then return to the information confirmed
(the i.e. described confirmation message), it is then total to determine currently to reside population according to the confirmation message returned by client is received
Number.Wherein, the confirmation message that may include that there is something special in the confirmation message, at this point, then by I35As Np;Alternatively, the confirmation
It is can wrap in information containing the resident population confirmed, then will confirm that and reside population included in information as Np;
Finally, will confirm that good current resident population NpIt is stored in database profession.
As it can be seen that the embodiment of the present invention be in leave home condition detection method can be by the view acquired from video
After frequency image sequence carries out motion target tracking, according to the motion profile, the type of current kinetic event is determined, wherein
The type of the current kinetic event includes enter event type and event type of going out, then according to the current kinetic determined
The type of event, the judgement for the state of leaving home that be in, such user just only need to buy video equipment can realize be in from
The detection of family's state is just able to achieve without installing multiple conventional body's detectors, dramatically saves the investment of equipment purchase
Cost, and installation space is saved, reduce the influence to home environment visual sense of beauty;Furthermore through the embodiment of the present invention
For detection method come the detection for the state of leaving home that be in, accuracy in detection is high, and it is fast to detect response.
As shown in fig. 6, the embodiment of the invention also provides a kind of detection systems of state of leaving home of being in, comprising:
Module is obtained, for obtaining sequence of video images from the video that shooting obtains;
Processing module, for obtaining the movement rail of the moving target after carrying out motion target tracking to sequence of video images
Mark;
Determining module, for determining the type of current kinetic event according to the motion profile, wherein described current
The type of motion event includes enter event type and event type of going out;
Judgment module, the judgement for the state of leaving home that according to the type of current kinetic event determined, carries out being in.
In some preferred embodiments, the determining module includes:
First determines submodule, for determining that the motion profile enters the sequence in several detection image regions;
Second determines from module, for determining that the type of current kinetic event is event class of entering according to the sequence
Type or event type of going out.
In some preferred embodiments, the judgment module includes:
First submodule, for being event type of entering when the type for determining current kinetic event, then will enter event
Record number add 1, and determine that current state is to be in state;
Second submodule, for being event type of going out when the type for determining current kinetic event, then will go out event
Record number add 1 after, according to current resident population, currently enter event record number and current event of going out
Record number, which calculates, is currently in number, and according to the calculated number of being currently in, the state of leaving home that be in is sentenced
It is disconnected.
In some preferred embodiments, described according to the calculated number of being currently in, carry out the state of leaving home of being in
Judgement comprising:
When it is described it is calculated be currently in number greater than 0, then determine that current state is state of being in;
When it is described it is calculated be currently in number equal to 0, then determine that current state is state of leaving home;
When it is described it is calculated be currently in number less than 0, then the calculated current number of being in is set to 0, and
Determine that current state is state of leaving home.
In some preferred embodiments, the currently resident population is determined by following steps:
The record number and the record for event of currently going out of event of once currently entering are calculated every prefixed time interval
Absolute difference between number;
When reaching preset time point, then numerical value is that maximum difference is exhausted from choosing in several obtained absolute differences
To value;
When the absolute difference selected is identical as preceding primary calculated resident population, it is determined that the choosing
The absolute difference of taking-up is currently to reside population;
When the absolute difference selected and preceding primary calculated resident population be not identical, then confirmation is sent
It requests to client, and determines currently to reside population according to the confirmation message returned by client is received.
In some preferred embodiments, the basis currently reside population, currently enter event record number with
And the record number for event of currently going out calculates the step for being currently in number comprising:
Calculate the difference between the record number for event of currently entering and the record number for event of currently going out;
The number of being currently in is obtained after the calculated difference is added with the currently resident population.
Suitable for this system embodiment, this system embodiment is implemented content in above method embodiment
Function is identical as above method embodiment, and the beneficial effect reached and above method embodiment beneficial effect achieved
It is identical.
The embodiment of the invention also provides a kind of detection devices of state of leaving home of being in, comprising:
At least one processor;
At least one processor, for storing at least one program;
When at least one described program is executed by least one described processor, so that at least one described processor is realized
A kind of detection method step of the state of leaving home of being in.As it can be seen that the content in above method embodiment is suitable for the present apparatus
In embodiment, the function that present apparatus embodiment is implemented is identical as above method embodiment, and the beneficial effect reached
It is also identical as above method embodiment beneficial effect achieved.
In addition, the embodiment of the invention also provides a kind of storage mediums, wherein being stored with the executable instruction of processor, institute
It states the executable instruction of processor and is used to execute any of above method embodiment technical solution institute when executed by the processor
A kind of detection method step for the state of leaving home of being in stated.It may include that high random access is deposited for the storage medium
Reservoir, can also include nonvolatile memory, a for example, at least disk memory, flush memory device or other are non-volatile
Property solid-state memory.It is also seen that the content in above method embodiment is originally deposited suitable for this storage medium embodiment
The function that storage media embodiment is implemented is identical as above method embodiment, and the beneficial effect and the above method reached
Embodiment beneficial effect achieved is also identical.
As shown in fig. 7, the embodiment of the invention also provides a kind of detection system of state of leaving home of being in, including controller with
And the camera for shooting video, the controller are connect with the camera;The controller includes:
At least one processor;
At least one processor, for storing at least one program;
When at least one described program is executed by least one described processor, so that at least one described processor is realized
A kind of detection method of the state of leaving home of being in.
Suitable for this system embodiment, this system embodiment is implemented content in above method embodiment
Function is identical as above method embodiment, and the beneficial effect reached and above method embodiment beneficial effect achieved
It is identical.
It should be appreciated that layer included in the embodiment of the present invention system, module, unit and/or platform etc. can be by
Computer hardware, the combination of hardware and software or the computer by being stored in non-transitory computer-readable memory
Instruction is to be effected or carried out.It includes the non-transitory configured with computer program that standard programming technology-, which can be used, in the method
Computer readable storage medium realizes in computer program, wherein configured in this way storage medium make computer with specific and
Predefined mode operates --- according to the method and attached drawing described in a particular embodiment.Each program can be with level process
Or the programming language of object-oriented is realized to communicate with computer system.However, if desired, the program can be with compilation or machine
Device language is realized.Under any circumstance, which can be the language of compiling or explanation.In addition, the program can for this purpose
It is run on the specific integrated circuit of programming.
In addition, the corresponding data executed of layer included in system of the embodiment of the present invention, module, unit and/or platform
Process flow can be performed in any suitable order, unless in addition instruction or otherwise significantly and up and down herein
Literary contradiction.The corresponding data processing stream executed of layer included in system of the embodiment of the present invention, module, unit and/or platform
Journey can execute under the control of one or more computer systems configured with executable instruction, and can be used as jointly one
The code executed on a or multiple processors is (for example, executable instruction, one or more computer programs or one or more are answered
With), realized by hardware or combinations thereof.The computer program includes the multiple fingers that can be performed by one or more processors
It enables.
Further, the system can be realized in being operably coupled to suitable any kind of computing platform, wrap
Include but be not limited to PC, mini-computer, main frame, work station, network or distributed computing environment, individual or integrated
Computer platform or communicated with charged particle tool or other imaging devices etc..Layer included in present system,
The corresponding flow chart of data processing executed of module, unit and/or platform can be to be stored in non-transitory storage medium or equipment
On machine readable code realize no matter to be moveable or be integrated to computing platform, such as hard disk, optically read and/or
Storage medium, RAM, ROM etc. is written, so that it can be read by programmable calculator, when storage medium or equipment are machine-readable by calculating
It can be used for configuration and operation computer when taking to execute process described herein.In addition, machine readable code, or part thereof can
To be transmitted by wired or wireless network.When such media include realizing institute above in conjunction with microprocessor or other data processors
When stating the instruction or program of step, invention as described herein includes that these and other different types of non-transitories are computer-readable
Storage medium.When methods and techniques according to the present invention programming, the invention also includes computers itself.
The above, only presently preferred embodiments of the present invention, the invention is not limited to above embodiment, as long as
It reaches technical effect of the invention with identical means, all within the spirits and principles of the present invention, any modification for being made,
Equivalent replacement, improvement etc., should be included within the scope of the present invention.Its technical solution within the scope of the present invention
And/or embodiment can have a variety of different modifications and variations.
Claims (10)
1. a kind of detection method for the state of leaving home of being in, which comprises the following steps:
Sequence of video images is obtained from the video that shooting obtains;
The motion profile of the moving target is obtained after carrying out motion target tracking to sequence of video images;
According to the motion profile, the type of current kinetic event is determined, wherein the type of the current kinetic event includes
It enters and event type and gos out event type;
According to the type for the current kinetic event determined, the judgement for the state of leaving home that be in.
2. a kind of detection method for the state of leaving home of being according to claim 1, which is characterized in that described according to the movement
Track, the step for determining the type of current kinetic event comprising:
Determine that the motion profile enters the sequence in several detection image regions;
According to the sequence, determine that the type of current kinetic event is enter event type or event type of going out.
3. a kind of detection method of state of leaving home of being according to claim 1 or claim 2, which is characterized in that described according to determination
The step for type of current kinetic event out, the judgement for the state of leaving home that be in comprising:
When the type for determining current kinetic event is to enter event type, then the record number for the event of entering is added 1, and determine
Current state is state of being in;
When the type for determining current kinetic event is event type of going out, then after the record number for the event of going out being added 1, according to
The record number of current resident population, the record number for event of currently entering and event of currently going out, which calculates, currently to exist
Family's number, according to the calculated judgement be currently in number, be in the state of leaving home.
4. a kind of detection method for the state of leaving home of being according to claim 3, which is characterized in that described according to the calculating
The step for being currently in number out, the judgement for the state of leaving home that be in comprising:
When it is described it is calculated be currently in number greater than 0, then determine that current state is state of being in;
When it is described it is calculated be currently in number equal to 0, then determine that current state is state of leaving home;
When it is described it is calculated be currently in number less than 0, then the calculated current number of being in is set to 0, and determine
Current state is state of leaving home.
5. a kind of detection method for the state of leaving home of being according to claim 3, which is characterized in that the currently resident population
Sum is determined by following steps:
Every prefixed time interval calculate once currently enter event record number and currently go out event record number it
Between absolute difference;
When reaching preset time point, then numerical value is that maximum difference is absolute from choosing in several obtained absolute differences
Value;
When the absolute difference selected is identical as preceding primary calculated resident population, it is determined that described to select
Absolute difference be current resident population;
When the absolute difference selected and preceding primary calculated resident population be not identical, then confirmation request is sent
It determines currently to reside population to client, and according to the confirmation message returned by client is received.
6. a kind of detection method for the state of leaving home of being according to claim 3, which is characterized in that the basis is currently resided
The record number of population, the record number for event of currently entering and event of currently going out calculate it is current be in number this
One step comprising:
Calculate the difference between the record number for event of currently entering and the record number for event of currently going out;
The number of being currently in is obtained after the calculated difference is added with the currently resident population.
7. a kind of detection system for the state of leaving home of being in characterized by comprising
Module is obtained, for obtaining sequence of video images from the video that shooting obtains;
Processing module, for obtaining the motion profile of the moving target after carrying out motion target tracking to sequence of video images;
Determining module, for determining the type of current kinetic event, wherein the current kinetic according to the motion profile
The type of event includes enter event type and event type of going out;
Judgment module, the judgement for the state of leaving home that according to the type of current kinetic event determined, carries out being in.
8. a kind of detection device for the state of leaving home of being in characterized by comprising
At least one processor;
At least one processor, for storing at least one program;
When at least one described program is executed by least one described processor, so that at least one described processor is realized as weighed
Benefit requires a kind of any one of 1-6 detection method for the state of leaving home of being in.
9. a kind of storage medium, wherein being stored with the executable instruction of processor, which is characterized in that the processor can be performed
Instruction is when executed by the processor for executing a kind of detection side for the state of leaving home of being in as described in claim any one of 1-6
Method.
10. a kind of detection system for the state of leaving home of being in, which is characterized in that the camera shooting including controller and for shooting video
Head, the controller are connect with the camera;The controller includes:
At least one processor;
At least one processor, for storing at least one program;
When at least one described program is executed by least one described processor, so that at least one described processor is realized as weighed
Benefit requires a kind of any one of 1-6 detection method for the state of leaving home of being in.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910371573.8A CN110245556A (en) | 2019-05-06 | 2019-05-06 | A kind of detection method, device, system and the storage medium of the state of leaving home of being in |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910371573.8A CN110245556A (en) | 2019-05-06 | 2019-05-06 | A kind of detection method, device, system and the storage medium of the state of leaving home of being in |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110245556A true CN110245556A (en) | 2019-09-17 |
Family
ID=67883708
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910371573.8A Pending CN110245556A (en) | 2019-05-06 | 2019-05-06 | A kind of detection method, device, system and the storage medium of the state of leaving home of being in |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110245556A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111652043A (en) * | 2020-04-15 | 2020-09-11 | 北京三快在线科技有限公司 | Object state identification method and device, image acquisition equipment and storage medium |
CN113256925A (en) * | 2021-05-08 | 2021-08-13 | 浙江宏海智能科技有限公司 | Intelligent building intrusion alarm method, system and storage medium thereof |
CN113359496A (en) * | 2021-06-28 | 2021-09-07 | 平安普惠企业管理有限公司 | Intelligent household data control method, device, equipment and storage medium |
CN114023037A (en) * | 2021-10-19 | 2022-02-08 | 深圳市中博科创信息技术有限公司 | User safety early warning method and device, electronic equipment and readable storage medium |
CN115035669A (en) * | 2022-08-10 | 2022-09-09 | 长沙朗源电子科技有限公司 | Security intelligent lock, control method and home security system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101231755A (en) * | 2007-01-25 | 2008-07-30 | 上海遥薇实业有限公司 | Moving target tracking and quantity statistics method |
CN102855475A (en) * | 2012-09-17 | 2013-01-02 | 广州杰赛科技股份有限公司 | School bus monitoring method and school bus monitoring system |
WO2015131734A1 (en) * | 2014-07-25 | 2015-09-11 | 中兴通讯股份有限公司 | Method, device, and storage medium for pedestrian counting in forward looking surveillance scenario |
CN107330386A (en) * | 2017-06-21 | 2017-11-07 | 厦门中控智慧信息技术有限公司 | A kind of people flow rate statistical method and terminal device |
-
2019
- 2019-05-06 CN CN201910371573.8A patent/CN110245556A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101231755A (en) * | 2007-01-25 | 2008-07-30 | 上海遥薇实业有限公司 | Moving target tracking and quantity statistics method |
CN102855475A (en) * | 2012-09-17 | 2013-01-02 | 广州杰赛科技股份有限公司 | School bus monitoring method and school bus monitoring system |
WO2015131734A1 (en) * | 2014-07-25 | 2015-09-11 | 中兴通讯股份有限公司 | Method, device, and storage medium for pedestrian counting in forward looking surveillance scenario |
CN107330386A (en) * | 2017-06-21 | 2017-11-07 | 厦门中控智慧信息技术有限公司 | A kind of people flow rate statistical method and terminal device |
Non-Patent Citations (1)
Title |
---|
郝鸿安等, 上海科学普及出版社 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111652043A (en) * | 2020-04-15 | 2020-09-11 | 北京三快在线科技有限公司 | Object state identification method and device, image acquisition equipment and storage medium |
CN113256925A (en) * | 2021-05-08 | 2021-08-13 | 浙江宏海智能科技有限公司 | Intelligent building intrusion alarm method, system and storage medium thereof |
CN113359496A (en) * | 2021-06-28 | 2021-09-07 | 平安普惠企业管理有限公司 | Intelligent household data control method, device, equipment and storage medium |
CN114023037A (en) * | 2021-10-19 | 2022-02-08 | 深圳市中博科创信息技术有限公司 | User safety early warning method and device, electronic equipment and readable storage medium |
CN115035669A (en) * | 2022-08-10 | 2022-09-09 | 长沙朗源电子科技有限公司 | Security intelligent lock, control method and home security system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110245556A (en) | A kind of detection method, device, system and the storage medium of the state of leaving home of being in | |
US20230316762A1 (en) | Object detection in edge devices for barrier operation and parcel delivery | |
CN108629791B (en) | Pedestrian tracking method and device and cross-camera pedestrian tracking method and device | |
US9147128B1 (en) | Machine learning enhanced facial recognition | |
Bazzani et al. | Decentralized particle filter for joint individual-group tracking | |
Francies et al. | A robust multiclass 3D object recognition based on modern YOLO deep learning algorithms | |
US11232327B2 (en) | Smart video surveillance system using a neural network engine | |
CN110399835B (en) | Analysis method, device and system for personnel residence time | |
CN101095149A (en) | Image comparison | |
CN102375537A (en) | Information processing apparatus, information processing method, and program | |
US20210182571A1 (en) | Population density determination from multi-camera sourced imagery | |
CN109508586B (en) | Passenger flow statistical method, device and equipment | |
US11423694B2 (en) | Methods and systems for dynamic and incremental face recognition | |
CN109902681B (en) | User group relation determining method, device, equipment and storage medium | |
CN110490697A (en) | Unmanned convenience store's settlement method, device, computer and storage medium | |
CN109597833A (en) | Event prediction method, apparatus, computer equipment and storage medium based on big data | |
CN113869274B (en) | Unmanned aerial vehicle intelligent tracking monitoring method and system based on city management | |
EP2447912B1 (en) | Method and device for the detection of change in illumination for vision systems | |
CN112800841B (en) | Pedestrian counting method, device and system and computer readable storage medium | |
CN110866175A (en) | Information recommendation method and device and electronic equipment | |
CN112329616A (en) | Target detection method, device, equipment and storage medium | |
CN112308045B (en) | Method and device for detecting dense crowd and electronic equipment | |
CN115457706A (en) | Control method and device of storage cabinet, electronic equipment and storage medium | |
CN114943336A (en) | Model pruning method, device, equipment and storage medium | |
CN110348422B (en) | Image processing method, image processing device, computer-readable storage medium and electronic equipment |
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 | ||
AD01 | Patent right deemed abandoned | ||
AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20240322 |