CN102074095A - System and method for monitoring infant behaviors - Google Patents
System and method for monitoring infant behaviors Download PDFInfo
- Publication number
- CN102074095A CN102074095A CN2010105365013A CN201010536501A CN102074095A CN 102074095 A CN102074095 A CN 102074095A CN 2010105365013 A CN2010105365013 A CN 2010105365013A CN 201010536501 A CN201010536501 A CN 201010536501A CN 102074095 A CN102074095 A CN 102074095A
- Authority
- CN
- China
- Prior art keywords
- moving target
- image
- presumptive area
- height
- described moving
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000012544 monitoring process Methods 0.000 title abstract description 17
- 230000006399 behavior Effects 0.000 title abstract 7
- 238000001514 detection method Methods 0.000 claims abstract description 39
- 238000005259 measurement Methods 0.000 claims abstract description 17
- 230000003542 behavioural effect Effects 0.000 claims description 17
- 230000033001 locomotion Effects 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 8
- 238000007792 addition Methods 0.000 claims description 7
- 238000010586 diagram Methods 0.000 description 14
- 238000005516 engineering process Methods 0.000 description 7
- 238000001914 filtration Methods 0.000 description 7
- 231100001261 hazardous Toxicity 0.000 description 6
- 230000000877 morphologic effect Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 206010020675 Hypermetropia Diseases 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000003930 cognitive ability Effects 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000009017 pursuit movement Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses a system for monitoring infant behaviors. The system for monitoring the infant behaviors comprises an image acquisition module, a target detection module, a behavior analysis module, a target height measurement module and an alarm prompt module, wherein the image acquisition module is used for acquiring an image; the target detection module is used for detecting a moving target from the image and tracking the moving target; the behavior analysis module is used for judging whether the moving target enters a preset region or not; the target height measurement module is used for calculating the height of the moving target and judging whether the height is smaller than a preset threshold value or not; and the alarm prompt module is used for sending out a prompt and/or alarm when the moving target enters the preset region and the height of the moving target is less than the preset threshold value. The system for monitoring the infant behaviors monitors infants by an image identification technique, sends out alarm after the infants do the preset dangerous behaviors, can be singly implemented only by utilizing a camera, and can also be integrated into the conventional monitoring cameral system conveniently.
Description
[technical field]
The present invention relates to image processing field, particularly a kind of video monitoring system and method based on pattern-recognition and intrusion detection.
[background technology]
In recent years, the child of taking place in the kindergarten wandered away or the child swarms into the hazardous location and constantly taken place by the accident of injury and so on, and child's protection simultaneously also is the hot issue of the daily concern of people.Because the child does not have self-control and to the cognitive ability of danger, so the child is very easy to have an accident under the ordering about of curiosity and rebel's psychology.And in the kindergarten, the teacher of kindergarten need look after several childs usually simultaneously, in case carelessness appears in the teacher of kindergarten, just may cause the generation of accident.Wherein, child or children run out of a kind of major accident that the kindergarten just belongs to the kindergarten privately.And the technological means that does not also have good impersonal force in the prior art prevents the generation of this accident.The user wishes and can provide certain child to detect and the function of reminding alarm and so on based on the first-class monitoring means of shooting.In addition in some occasions, include the public arena in adventurous zone for the child such as crossing, supermarket, warehouse etc., same demand is also arranged.
Therefore be necessary to provide a kind of new technical scheme to solve above-mentioned shortcoming.
[summary of the invention]
One object of the present invention is to provide a kind of infant behavior supervisory system, utilizes image recognition technology to carry out the child and monitors and can be provided in the function of reporting to the police behind the predetermined hazardous act that detects the child.
Another object of the present invention is to provide a kind of infant behavior method for supervising, utilize image recognition technology to carry out the child and monitor and can be provided in the function of reporting to the police behind the predetermined hazardous act that detects the child.
In order to reach purpose of the present invention, the invention provides a kind of infant behavior supervisory system, described system comprises: image capture module, images acquired; Module of target detection detects moving target and tracking from described image; The behavioural analysis module judges whether described moving target enters presumptive area; The object height measurement module calculates the height of described moving target and judges that whether described height is less than reservation threshold; With the alarm module, the height that enters presumptive area and described moving target when described moving target sends prompting during less than reservation threshold and/or reports to the police.
Further, described object height measurement module is according to the calibration information of described image capture module
Calculate the height α Z of described moving target:
Wherein
Be the line vector that goes out of described image, v is the vanishing point vector on the vertical direction of described image, and t, b have represented top coordinate and the bottom coordinate of described moving target in image.
Further, when described moving target region and described presumptive area have common factor, and the motion vector direction of described moving target meets the predetermined direction scope, and then described behavioural analysis module determines that described moving target has entered presumptive area.
Further, described moving target region and described presumptive area have common factor to judge in the following way: preestablish an image template, described image template is an onesize bianry image or a gray level image of gathering with described image capture module of image, pixel value corresponding to the unintended areas in the image in the described image template all is 0, all is greater than 0 and smaller or equal to 255 same pixel value corresponding to the pixel value of the presumptive area in the image in the described image template; Zone and the described image template of described moving target in image multiplied each other; Whether all the pixel value additions in the resulting image are greater than reservation threshold; If then described moving target has entered presumptive area.
Further, described module of target detection utilizes two class sorters to carry out moving object detection, and described two class sorters are " number of people human body/inhuman tribal chief's body " the two class sorters that adopt the training of Adaboost training method.
According to a further aspect in the invention, the invention provides a kind of infant behavior method for supervising, described method comprises: images acquired; From described image, detect moving target; Calculate the height of described moving target; Judge that whether described height is less than reservation threshold; Height is followed the tracks of less than the moving target of reservation threshold; Judge whether described height enters presumptive area less than the moving target of reservation threshold; With when described moving target enters presumptive area, send prompting and/or warning.
Further, the formula of the height α Z of the described moving target of described calculating is:
Wherein
Be the line vector that goes out of described image, v is the vanishing point vector on the vertical direction of described image, and t, b have represented top coordinate and the bottom coordinate of described moving target in image.
Further, whether described decision height enters presumptive area less than the moving target of reservation threshold is meant: when described moving target region and described presumptive area have common factor, and when the motion vector direction of described moving target meets the predetermined direction scope, determine that described moving target has entered presumptive area, whether wherein said moving target region and described presumptive area have common factor to judge in the following way:
Preestablish an image template, described image template is an onesize bianry image or a gray level image of gathering with described image capture module of image, pixel value corresponding to the unintended areas in the image in the described image template all is 0, all is greater than 0 and smaller or equal to 255 same pixel value corresponding to the pixel value of the presumptive area in the image in the described image template; Zone and the described image template of described moving target in image multiplied each other; Whether all the pixel value additions in the resulting image are greater than reservation threshold; If then described moving target region and described presumptive area have common factor.
In accordance with a further aspect of the present invention, also a kind of infant behavior method for supervising of the present invention, described method comprises: images acquired; From described image, detect moving target; Described moving target is followed the tracks of; Judge whether described moving target enters presumptive area; When described moving target enters presumptive area, calculate the height of described moving target; Judge that whether described height is less than reservation threshold; With send prompting when the height of described moving target during less than reservation threshold and/or report to the police.
Further, the formula of the height α Z of the described moving target of described calculating is:
Wherein
Be the line vector that goes out of described image, v is the vanishing point vector on the vertical direction of described image, and t, b have represented top coordinate and the bottom coordinate of described moving target in image.
Further, describedly judge whether moving target enters presumptive area and be meant: when described moving target region and described presumptive area have common factor, and when the motion vector direction of described moving target meets the predetermined direction scope, determine that described moving target has entered presumptive area, whether wherein said moving target region and described presumptive area have common factor to judge in the following way:
Preestablish an image template, described image template is an onesize bianry image or a gray level image of gathering with described image capture module of image, pixel value corresponding to the unintended areas in the image in the described image template all is 0, all is greater than 0 and smaller or equal to 255 same pixel value corresponding to the pixel value of the presumptive area in the image in the described image template; Zone and the described image template of described moving target in image multiplied each other; Whether all the pixel value additions in the resulting image are greater than reservation threshold; If then described moving target region and described presumptive area have common factor.
Compared with prior art, the present invention utilizes image recognition technology to carry out child's monitoring, and send warning after the child makes predetermined hazardous act, it not only can only utilize a camera just can implement separately, also can be integrated in easily in the existing monitored video camera system.
[description of drawings]
In conjunction with reaching ensuing detailed description with reference to the accompanying drawings, the present invention will be more readily understood, the structure member that wherein same Reference numeral is corresponding same, wherein:
Fig. 1 is the block diagram of the infant behavior supervisory system in one embodiment of the present of invention;
Fig. 2 is the block diagram of the module of target detection in one embodiment of the present of invention;
Fig. 3 is the block diagram of the behavioural analysis module in one embodiment of the present of invention;
Fig. 4 is the block diagram of the object height measurement module in one embodiment of the present of invention;
Fig. 5 is the method flow diagram of the infant behavior method for supervising in one embodiment of the present of invention;
Fig. 6 is the method flow diagram of the infant behavior method for supervising in an alternative embodiment of the invention; With
Fig. 7 is the enforcement synoptic diagram of the infant behavior supervisory system in one embodiment of the present of invention.
[embodiment]
For above-mentioned purpose of the present invention, feature and advantage can be become apparent more, the present invention is further detailed explanation below in conjunction with the drawings and specific embodiments.
Infant behavior supervising device and method among the present invention can utilize software, hardware and both combinations thereof to implement to become the part of a module, a system or a system.Described infant behavior supervising device and method utilize image recognition technology to carry out child's detection and warning, make it not only can implement separately, also can be integrated in easily in the existing monitored video camera system.
Please refer to Fig. 1, it shows the block diagram of the infant behavior supervisory system 100 in one embodiment of the present of invention.Described infant behavior supervisory system 100 comprises image capture module 110, module of target detection 120, object height measurement module 130, behavioural analysis module 140 and alarm module 150.
Image capture module 110 comprises the image or the video acquisition device of photo-sensitive cell (as ccd image sensor), camera and so on, and it can be the rig camera in the guarded region.The image in described image capture module 110 continuous acquisition monitoring zones, described image is generally a continuous frame two field picture.The video camera of installing at primary school or each front door of kindergarten for example is the monitoring image of 320 pixel *, 240 pixels according to the speed acquisition resolution of per second 30 frames.
Module of target detection 120 detects moving target and follows the tracks of from the image that described image capture module 110 is gathered.The method of moving object detection has a lot, for example frame-to-frame differences point-score, background subtraction point-score and object module detection method or the like.Because the enforcement environment of video monitoring system 100 is kindergarten, primary school and hospital or the like public arena normally, described moving target mainly is the people, also to distinguish adult and children specifically, so in a more excellent embodiment, described target monitoring module 120 can adopt background subtraction point-score and person model detection method to improve the degree of accuracy of moving object detection simultaneously.Can also adopt motion tracking method based on fast coupling to described motion target tracking to detected moving target simultaneously, also can adopt other motion target tracking technology that described moving target is followed the tracks of.
Object height measurement module 130 calculates the height of described moving target and judges that whether described height is less than reservation threshold.For same guarded region, if described image capture module 110 is two or multiple cameras, the high computational for a moving object is more or less freely so.But because the cost of two or multiple cameras is higher, 110 of common described image capture modules are a video camera, will calculate the height of a moving object this moment, can be summed up as the three-dimensional information problem that obtains object in the scene in the single-view image.According to the projective geometry principle, on image, give a bit mutually after the parallel lines projection in the three-dimensional theorem in Euclid space.Simply, the subject image in the two field picture meets near big and far smaller little, nearly farsighted low rule.If know the vanishing point vector in the two field picture and the line vector that goes out, just can utilize formula to calculate the true altitude α Z of the moving target in this image:
Wherein
Be the line vector that goes out after the normalization, v is the vanishing point vector on the vertical direction, and t, b have represented top coordinate and the bottom coordinate of described moving target in image.Wherein
Can obtain by described image capture module 110 being carried out the single-view demarcation.After calculating the true altitude of described moving target, if whether the true altitude that described object height measurement module 130 is further judged described moving target less than reservation threshold, then thinks described moving target be children less than reservation threshold.
Alarm module 150 is used to send prompting and/or reports to the police.That is to say, judge described moving target when described behavioural analysis module 140 and entered presumptive area, and the height that described object height measurement module 130 is judged described moving target is during less than reservation threshold, and described alarm module 150 is sent prompting or reported to the police.Described alarm module 150 can be the physical equipment of display or sound equipment and so on.
In sum, the enforcement emphasis and the difficult point of the video monitoring system among the present invention comprise: 1) utilize the object height measurement module to obtain the true altitude of the moving target of guarded region, whether judge less than reservation threshold whether it is children according to the true altitude of described moving target then; 2) utilize the behavioural analysis module to judge whether moving target has entered presumptive area.Another angle says, these 2 also is the necessary condition of described infant behavior supervisory system when reporting to the police.In view of infant behavior supervisory system 100 shown in Figure 1 is comparatively simple embodiment among the present invention, hereinafter will do elaboration further to each module of described infant behavior supervisory system and the preferred implementation of each module.
Please refer to Fig. 2, it shows the block diagram of the module of target detection 120 in one embodiment of the present of invention.Described module of target detection 120 comprises foreground detection unit 222, morphologic filtering unit 224, moving object detection unit 226, disaggregated model detecting unit 228 and motion target tracking unit 230.
Described foreground detection unit 222 subtracts each other present image and background image that to obtain difference image also be foreground image, and described background image can be by obtaining based on medium filtering or based on the background model that gaussian filtering is set up.
Described morphologic filtering unit 224 is converted into bianry image with described foreground image, utilizes erosion algorithm in the mathematical morphology, expansion algorithm, opening operation or the like to carry out filtering to described bianry image then.
The 226 pairs of described filtered bianry images in described moving object detection unit can utilize connected domain analysis, corresponding pixel points diversity ratio etc. method carry out motion detection.Because the technology that these three unit adopt is comparatively ripe in the prior art, is content that those skilled in the art know, so be not repeated.But in actual applications, the overlapping problem of a plurality of moving targets in one two field picture can often take place, if described module of target detection 220 includes only foreground detection unit 222, morphologic filtering unit 224 and moving object detection unit 226, when then in to the overlapping image of a plurality of moving targets, carrying out moving object detection, being easy to the overlapping zone afterwards of a plurality of moving targets is detected is an independent target, and single moving object detection can not be come out the testing result that lead to errors this moment easily.Walk such as two adults and outside being discharged into the kindergarten, be easy in the image of this scene two adults' overlapping part is detected as a moving target, and the height of this moving target is usually far below adult's height, thought by mistake be children and send wrong report by system easily.In view of this kind situation, described module of target detection 120 can also comprise disaggregated model detecting unit 228.
Described disaggregated model detecting unit 228 carries out moving object detection by two class sorters, and described two class sorters are " number of people human body/inhuman tribal chief's body " the two class sorters that adopt the training of Adaboost training method.The Adaboost theory is a kind of comparatively mature theory in the existing human face detection tech, specifically, the AdaBoost theory is the algorithm that a kind of Weak Classifier that will be better than conjecture at random arbitrarily is combined into strong classifier, because, combination is based on the training method of Adaboost theory in the present embodiment, gather a large amount of number of people samples, human sample, non-number of people sample and non-human sample come training pattern, a plurality of Weak Classifiers based on single feature are formed a stronger sorter, then a plurality of strong classifiers are cascaded into complete " number of people human body/inhuman tribal chief's body " two class sorters.When specifically carrying out moving object detection, utilize " number of people human body/inhuman tribal chief's body " two class sorters of this cascade to carry out hierarchical search to described difference image.This method can obtain good effect, can detect people's head and foot comparatively exactly.
After described moving object detection is come out, can also carry out the tracking of moving target according to motion target tracking unit 230.The motion tracking method that described motion target tracking unit 230 can adopt based on fast coupling.
Obviously, described module of target detection 120 can also be other implementations in other embodiments, such as including only morphologic filtering unit 224 and moving object detection unit 226; Such as including only described foreground detection unit 222 and described disaggregated model detecting unit 228; Again or adopt other Detection for Moving Target all to be fine.Be as the criterion but should reach the head and the corresponding foot that accurately detect human body in the image as far as possible, because the height that described object height measurement module 130 need calculate moving target according to the top coordinate and the bottom coordinate of moving target.
Please refer to Fig. 3, it shows the block diagram of the behavioural analysis module 140 in one embodiment of the present of invention.Described behavioural analysis module 140 comprises presumptive area seek common ground unit 342 and moving target vector analysis unit 344.
The described presumptive area unit 342 that seeks common ground judges whether described moving target enters presumptive area, in view of described image capture module 110 common position after setting up installation can not change, the guarded region that is described image capture module 110 is constant usually, ask the seek common ground operation efficiency of unit 342 of described presumptive area in order to improve so, can preestablish an image template, described image template is an onesize bianry image or a gray level image of gathering with described image capture module 110 of monitoring image, pixel value corresponding to the unintended areas in the monitoring image in the described image template all is 0, all be greater than 0 and smaller or equal to 255 same pixel value corresponding to the pixel value of the presumptive area in the monitoring image in the described image template, such as 1 or 59 or the like.Calculating the zone of described moving target in image when whether having common factor with described presumptive area, only zone and the described image template of described moving target in image need be multiplied each other, all the pixel value additions in the resulting image just can think that greater than reservation threshold described moving target has entered presumptive area.
On the other hand, consider a kind of like this application scenarios: doorway in the kindergarten, existing moving target of going out from the kindergarten, the moving target (sending child such as the head of a family) that enters the kindergarten is also arranged, should not report to the police under this kind application scenarios to the moving target that enters the kindergarten, so described behavioural analysis module 140 also should detect the motion vector direction of the moving target that enters presumptive area, so among some embodiment, described behavioural analysis module 140 also comprises moving target vector analysis unit 344, described moving target vector analysis unit 344 judges whether the motion vector method of described moving target meets the predetermined direction scope, for example the inner outside, kindergarten of pointing to from the kindergarten.All be under the sure situation only in the seek common ground judged result of unit 342 and moving target vector analysis unit 344 of presumptive area, described behavioural analysis module 140 thinks that just described moving target has entered presumptive area, that is to say, have only moving target to enter presumptive area according to certain mode, described behavioural analysis module 140 can judge that just this moving target meets the necessary condition that described infant behavior supervisory system is reported to the police.
Please continue with reference to figure 4, it shows the block diagram of the object height measurement module 130 in one embodiment of the present of invention.Described object height measurement module 130 comprises object height computing unit 432 and object height judging unit 434.
Described object height computing unit 432 utilizes formula to come the true altitude α Z of the moving target in the computed image:
Wherein
Be the line vector that goes out after the normalization, v is the vanishing point vector on the vertical direction, and t, b have represented top coordinate and the bottom coordinate of described moving target in image.Wherein
Can obtain by described image capture module 110 being carried out the single-view demarcation.T, b can be obtained by the testing result of module of target detection 120.
Described object height judging unit 434 judges that whether described true altitude α Z is less than reservation threshold.If less than just thinking that described moving target is children.Another necessary condition when this moment, object height measurement module 130 result of determination represented that this moving target meets described infant behavior supervisory system and reports to the police.
In a part of embodiment, the calibration information of described image capture module 110 is to be stored in its inside in advance, in certain module of perhaps described infant behavior supervisory system.For the image capture module of not demarcating in advance 110, can take traditional demarcating at guarded region placement scaling reference, also can take following method to demarcate.
Because the embodiment of the invention also is based on the principle of go out line and vanishing point and demarcates.Obtain for convenience the to go out position of line, the present invention utilizes in the monitoring scene target information perpendicular to ground, such as the image-forming information of same person after diverse location is stood, electric pole information, the lane line information on the road etc.According to the projective geometry principle, the parallel lines in the three-dimensional theorem in Euclid space intersect at a vanishing point in the meeting after the projection on image, and two vanishing points just can determine the line that goes out.As long as, just can calculate the line position that goes out so known parallel to the ground in the three dimensions but the two pairs of parallel lines that direction is different.
Principle according to demarcating except the line that goes out, also needs to know the vanishing point position of direction perpendicular to the ground.This vanishing point information can be according to calculating perpendicular to the pair of parallel line on ground (such as according to the body of wall information of buildings or perpendicular to the electric pole on ground).In order to simplify calculating, the vanishing point that also can suppose vertical direction is positioned at the infinite distant place on the y direction.
Therefore, the embodiment of the invention is by two pairs of parallel lines parallel to the ground but that direction is inequality, obtains the line vector that goes out after the normalization
By pair of parallel line, obtain the vanishing point vector v on the vertical direction perpendicular to ground.Certainly, can obtain the line vector that goes out
Only be not limited to by above geometric element with the mode of vanishing point vector v, those skilled in the art also can obtain by other geometric elements.
After obtaining calibration information, just need carry out the calculating of moving target true altitude according to top coordinate and bottom coordinate and the described calibration information of moving target in image by above method.
Will be appreciated that the processing that described object height measurement module 130 and described behavioural analysis module 140 are carried out separately decision logic in other words can walk abreast, and also sequencing can be arranged.In view of this different processing sequence, the present invention also provides two kinds of infant behavior method for supervising simultaneously,
Please refer to Fig. 5, it shows the method flow diagram of the infant behavior method for supervising 500 in one embodiment of the present of invention.Described infant behavior method for supervising 500 comprises:
Step 502 detects moving target from described image;
Step 503 is calculated the height of described moving target and is judged that whether described height is less than reservation threshold;
Step 504 is to wherein highly following the tracks of less than the moving target of reservation threshold;
Step 505 judges whether described height enters presumptive area less than the moving target of reservation threshold; With
Step 506 when described moving target enters presumptive area, is sent prompting and/or warning.
Obviously, adopt in the present embodiment and earlier the height of detected moving target is judged, then to wherein highly following the tracks of less than the moving target (being child or children) of reservation threshold, in case described height has entered presumptive area less than the moving target of reservation threshold, then send the scheme of warning.An advantage of this kind scheme is: need the moving target followed the tracks of less in step 504.But a shortcoming of this kind scheme is: need the moving target of computed altitude more in step 503.
Please continue with reference to figure 6, it shows the method flow diagram of the infant behavior method for supervising 600 in an alternative embodiment of the invention.Described infant behavior method for supervising 600 comprises:
Step 602 detects moving target from described image;
Step 603 is followed the tracks of described moving target;
Step 604 judges whether described moving target enters presumptive area; With
Step 605 when described moving target enters presumptive area, is calculated the height of described moving target and is judged that whether described height is less than reservation threshold; With
Step 606 is when the height of described moving target sends prompting during less than reservation threshold and/or reports to the police.
Obviously, adopting elder generation whether detected moving target has been entered presumptive area in the present embodiment judges, moving target (no matter adult or children) computed altitude to having entered presumptive area then is in case the height of described moving target less than reservation threshold, then sends the scheme of warning.An advantage of this kind scheme is: need the moving target of computed altitude less in step 603.But a shortcoming of this kind scheme is: need the moving target followed the tracks of more in step 604.
For different embodiment, the operand of the height of described calculating moving target is different with computational complexity; The operand of described pursuit movement target is different with computational complexity.Can select a kind of mode in Fig. 5 and the infant behavior method for supervising shown in Figure 6 as embodiment according to both computational complexities.
Please refer to Fig. 7, it shows the enforcement synoptic diagram of the infant behavior supervisory system in one embodiment of the present of invention.The enforcement environment of described infant behavior supervisory system is a kindergarten, and this kindergarten includes the first gate D
AWith the second gate D
BBe provided with corresponding to the described first gate D in inside, kindergarten
AWith the second gate D
BTwo video camera C, described video camera C is used for images acquired and is transferred to other modules of described infant behavior supervisory system.At the close first gate D
AWith the second gate D
BThe place is respectively arranged with the first presumptive area R
AWith the second presumptive area R
BIn case, there is child inner when walking toward the outside, kindergarten from the kindergarten, entered the described first presumptive area R
AThe perhaps second presumptive area R
B, so described infant behavior supervisory system will be reported to the police.
Above-mentioned explanation has fully disclosed the specific embodiment of the present invention.It is pointed out that and be familiar with the scope that any change that the person skilled in art does the specific embodiment of the present invention does not all break away from claims of the present invention.Correspondingly, the scope of claim of the present invention also is not limited only to described embodiment.
Claims (11)
1. infant behavior supervisory system is characterized in that it comprises:
Image capture module, images acquired;
Module of target detection detects moving target and tracking from described image;
The behavioural analysis module judges whether described moving target enters presumptive area;
The object height measurement module calculates the height of described moving target and judges that whether described height is less than reservation threshold; With
Alarm module, the height that enters presumptive area and described moving target when described moving target are sent prompting during less than reservation threshold and/or are reported to the police.
2. system according to claim 1 is characterized in that, described object height measurement module is according to the calibration information of described image capture module
Calculate the height α Z of described moving target:
Wherein
Be the line vector that goes out of described image, v is the vanishing point vector on the vertical direction of described image, and t, b have represented top coordinate and the bottom coordinate of described moving target in image.
3. system according to claim 1, it is characterized in that, when described moving target region and described presumptive area have common factor, and the motion vector direction of described moving target meets the predetermined direction scope, and then described behavioural analysis module determines that described moving target has entered presumptive area.
4. system according to claim 1 is characterized in that, described moving target region and described presumptive area have common factor to judge in the following way:
Preestablish an image template, described image template is an onesize bianry image or a gray level image of gathering with described image capture module of image, pixel value corresponding to the unintended areas in the image in the described image template all is 0, all is greater than 0 and smaller or equal to 255 same pixel value corresponding to the pixel value of the presumptive area in the image in the described image template;
Zone and the described image template of described moving target in image multiplied each other;
Whether all the pixel value additions in the resulting image are greater than reservation threshold;
If then described moving target has entered presumptive area.
5. system according to claim 1, it is characterized in that, described module of target detection utilizes two class sorters to carry out moving object detection, and described two class sorters are " number of people human body/inhuman tribal chief's body " the two class sorters that adopt the training of Adaboost training method.
6. infant behavior method for supervising is characterized in that it comprises:
Images acquired;
From described image, detect moving target;
Calculate the height of described moving target;
Judge that whether described height is less than reservation threshold;
Height is followed the tracks of less than the moving target of reservation threshold;
Judge whether described height enters presumptive area less than the moving target of reservation threshold; With
When described moving target enters presumptive area, send prompting and/or warning.
7. method according to claim 6 is characterized in that, the formula of the height α Z of the described moving target of described calculating is:
Wherein
Be the line vector that goes out of described image, v is the vanishing point vector on the vertical direction of described image, and t, b have represented top coordinate and the bottom coordinate of described moving target in image.
8. method according to claim 6, it is characterized in that, whether described decision height enters presumptive area less than the moving target of reservation threshold is meant: when described moving target region and described presumptive area have common factor, and when the motion vector direction of described moving target meets the predetermined direction scope, determine that described moving target has entered presumptive area, whether wherein said moving target region and described presumptive area have common factor to judge in the following way:
Preestablish an image template, described image template is an onesize bianry image or a gray level image of gathering with described image capture module of image, pixel value corresponding to the unintended areas in the image in the described image template all is 0, all is greater than 0 and smaller or equal to 255 same pixel value corresponding to the pixel value of the presumptive area in the image in the described image template;
Zone and the described image template of described moving target in image multiplied each other;
Whether all the pixel value additions in the resulting image are greater than reservation threshold;
If then described moving target region and described presumptive area have common factor.
9. infant behavior method for supervising is characterized in that it comprises:
Images acquired;
From described image, detect moving target;
Described moving target is followed the tracks of;
Judge whether described moving target enters presumptive area;
When described moving target enters presumptive area, calculate the height of described moving target;
Judge that whether described height is less than reservation threshold; With
When the height of described moving target sends prompting during less than reservation threshold and/or reports to the police.
10. method according to claim 9 is characterized in that, the formula of the height α Z of the described moving target of described calculating is:
Wherein
Be the line vector that goes out of described image, v is the vanishing point vector on the vertical direction of described image, and t, b have represented top coordinate and the bottom coordinate of described moving target in image.
11. method according to claim 9, it is characterized in that, describedly judge whether moving target enters presumptive area and be meant: when described moving target region and described presumptive area have common factor, and when the motion vector direction of described moving target meets the predetermined direction scope, determine that described moving target has entered presumptive area, whether wherein said moving target region and described presumptive area have common factor to judge in the following way:
Preestablish an image template, described image template is an onesize bianry image or a gray level image of gathering with described image capture module of image, pixel value corresponding to the unintended areas in the image in the described image template all is 0, all is greater than 0 and smaller or equal to 255 same pixel value corresponding to the pixel value of the presumptive area in the image in the described image template;
Zone and the described image template of described moving target in image multiplied each other;
Whether all the pixel value additions in the resulting image are greater than reservation threshold;
If then described moving target region and described presumptive area have common factor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105365013A CN102074095B (en) | 2010-11-09 | 2010-11-09 | System and method for monitoring infant behaviors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105365013A CN102074095B (en) | 2010-11-09 | 2010-11-09 | System and method for monitoring infant behaviors |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102074095A true CN102074095A (en) | 2011-05-25 |
| CN102074095B CN102074095B (en) | 2013-04-24 |
Family
ID=44032621
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010105365013A Expired - Fee Related CN102074095B (en) | 2010-11-09 | 2010-11-09 | System and method for monitoring infant behaviors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102074095B (en) |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102256026A (en) * | 2011-07-29 | 2011-11-23 | 中国联合网络通信集团有限公司 | Method for acquiring guarding state information and user equipment |
| CN103686086A (en) * | 2013-12-13 | 2014-03-26 | 广州中国科学院先进技术研究所 | Method for carrying out video monitoring on specific area |
| CN103735269A (en) * | 2013-11-14 | 2014-04-23 | 大连民族学院 | Height measurement method based on video multi-target tracking |
| CN103810717A (en) * | 2012-11-09 | 2014-05-21 | 浙江大华技术股份有限公司 | Human behavior detection method and device |
| CN104270609A (en) * | 2014-10-09 | 2015-01-07 | 深圳市中控生物识别技术有限公司 | Method, system and device for remote monitoring |
| CN104299359A (en) * | 2014-10-29 | 2015-01-21 | 合肥指南针电子科技有限责任公司 | System and method for monitoring activity range of children in community |
| CN104700054A (en) * | 2013-12-05 | 2015-06-10 | 三菱电机株式会社 | Dynamic management system |
| CN104823218A (en) * | 2012-12-03 | 2015-08-05 | 哈曼国际工业有限公司 | System and method for detecting pedestrians using a single normal camera |
| CN105373774A (en) * | 2015-10-10 | 2016-03-02 | 安徽清新互联信息科技有限公司 | Method for detecting physical punishment behaviors of kindergarten teachers on children |
| CN105451235A (en) * | 2015-11-13 | 2016-03-30 | 大连理工大学 | Wireless sensor network intrusion detection method based on background updating |
| CN105575049A (en) * | 2015-06-26 | 2016-05-11 | 宇龙计算机通信科技(深圳)有限公司 | Early warning method, device and terminal |
| CN106056035A (en) * | 2016-04-06 | 2016-10-26 | 南京华捷艾米软件科技有限公司 | Motion-sensing technology based kindergarten intelligent monitoring method |
| CN106127814A (en) * | 2016-07-18 | 2016-11-16 | 四川君逸数码科技股份有限公司 | A kind of wisdom gold eyeball identification gathering of people is fought alarm method and device |
| CN106327787A (en) * | 2015-07-03 | 2017-01-11 | 中兴通讯股份有限公司 | Alarm processing method, device and terminal |
| CN106384479A (en) * | 2016-12-16 | 2017-02-08 | Tcl集团股份有限公司 | Monitoring method and device |
| CN107045623A (en) * | 2016-12-30 | 2017-08-15 | 厦门瑞为信息技术有限公司 | A kind of method of the indoor dangerous situation alarm based on human body attitude trace analysis |
| CN107578422A (en) * | 2017-08-01 | 2018-01-12 | 中国农业大学 | A kind of pedestrian's trace detection system and method |
| WO2018218563A1 (en) * | 2017-05-31 | 2018-12-06 | 深圳市永恒丰科技有限公司 | Smart gate opening and closing control method |
| CN109567815A (en) * | 2018-10-23 | 2019-04-05 | 电子科技大学 | Remote monitoring method for baby behavior state |
| CN110175512A (en) * | 2019-04-12 | 2019-08-27 | 天津华来科技有限公司 | A kind of scene detection alarm method and device |
| CN110428582A (en) * | 2019-06-28 | 2019-11-08 | 苏州浪潮智能科技有限公司 | A kind of method and system of infantal sleeping protection |
| CN111797776A (en) * | 2020-07-07 | 2020-10-20 | 云知声智能科技股份有限公司 | Infant monitoring method and device based on posture |
| CN112395921A (en) * | 2019-08-16 | 2021-02-23 | 杭州海康威视数字技术股份有限公司 | Abnormal behavior detection method, device and system |
| CN113251558A (en) * | 2021-05-13 | 2021-08-13 | 重庆海尔空调器有限公司 | Infant anti-falling monitoring control method and device, storage medium and air conditioner |
| CN113467325A (en) * | 2021-07-23 | 2021-10-01 | 俞小飞 | Environment monitoring method based on Internet of things |
| CN113657189A (en) * | 2021-07-26 | 2021-11-16 | 浙江大华技术股份有限公司 | Behavior detection method, electronic device, and computer-readable storage medium |
| CN114360209A (en) * | 2022-01-17 | 2022-04-15 | 常州信息职业技术学院 | Video behavior identification security system based on deep learning |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070127774A1 (en) * | 2005-06-24 | 2007-06-07 | Objectvideo, Inc. | Target detection and tracking from video streams |
| CN101106700A (en) * | 2007-08-01 | 2008-01-16 | 大连海事大学 | Intelligent target detail capturing device and method in video monitoring system |
| CN101494772A (en) * | 2009-02-13 | 2009-07-29 | 北京中星微电子有限公司 | Method and apparatus for detecting image |
| CN101587624A (en) * | 2008-05-21 | 2009-11-25 | 上海新联纬讯科技发展有限公司 | System of detecting and indentifying unexpected conditions of staff |
-
2010
- 2010-11-09 CN CN2010105365013A patent/CN102074095B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070127774A1 (en) * | 2005-06-24 | 2007-06-07 | Objectvideo, Inc. | Target detection and tracking from video streams |
| CN101106700A (en) * | 2007-08-01 | 2008-01-16 | 大连海事大学 | Intelligent target detail capturing device and method in video monitoring system |
| CN101587624A (en) * | 2008-05-21 | 2009-11-25 | 上海新联纬讯科技发展有限公司 | System of detecting and indentifying unexpected conditions of staff |
| CN101494772A (en) * | 2009-02-13 | 2009-07-29 | 北京中星微电子有限公司 | Method and apparatus for detecting image |
Cited By (38)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102256026B (en) * | 2011-07-29 | 2014-01-22 | 中国联合网络通信集团有限公司 | Method for acquiring guarding state information and user equipment |
| CN102256026A (en) * | 2011-07-29 | 2011-11-23 | 中国联合网络通信集团有限公司 | Method for acquiring guarding state information and user equipment |
| CN103810717A (en) * | 2012-11-09 | 2014-05-21 | 浙江大华技术股份有限公司 | Human behavior detection method and device |
| CN103810717B (en) * | 2012-11-09 | 2016-08-24 | 浙江大华技术股份有限公司 | A kind of human body behavioral value method and device |
| CN104823218A (en) * | 2012-12-03 | 2015-08-05 | 哈曼国际工业有限公司 | System and method for detecting pedestrians using a single normal camera |
| CN103735269A (en) * | 2013-11-14 | 2014-04-23 | 大连民族学院 | Height measurement method based on video multi-target tracking |
| CN103735269B (en) * | 2013-11-14 | 2015-10-28 | 大连民族学院 | A kind of height measurement method followed the tracks of based on video multi-target |
| CN104700054A (en) * | 2013-12-05 | 2015-06-10 | 三菱电机株式会社 | Dynamic management system |
| CN103686086B (en) * | 2013-12-13 | 2017-07-14 | 广州中国科学院先进技术研究所 | A kind of method that video monitoring is carried out for specific region |
| CN103686086A (en) * | 2013-12-13 | 2014-03-26 | 广州中国科学院先进技术研究所 | Method for carrying out video monitoring on specific area |
| CN104270609A (en) * | 2014-10-09 | 2015-01-07 | 深圳市中控生物识别技术有限公司 | Method, system and device for remote monitoring |
| CN104299359A (en) * | 2014-10-29 | 2015-01-21 | 合肥指南针电子科技有限责任公司 | System and method for monitoring activity range of children in community |
| CN105575049B (en) * | 2015-06-26 | 2018-05-15 | 宇龙计算机通信科技(深圳)有限公司 | A kind of method for early warning, device and terminal |
| CN105575049A (en) * | 2015-06-26 | 2016-05-11 | 宇龙计算机通信科技(深圳)有限公司 | Early warning method, device and terminal |
| CN106327787A (en) * | 2015-07-03 | 2017-01-11 | 中兴通讯股份有限公司 | Alarm processing method, device and terminal |
| CN105373774B (en) * | 2015-10-10 | 2018-12-28 | 安徽清新互联信息科技有限公司 | A kind of detection method of preschool teacher's school's corporal punishment child's behavior |
| CN105373774A (en) * | 2015-10-10 | 2016-03-02 | 安徽清新互联信息科技有限公司 | Method for detecting physical punishment behaviors of kindergarten teachers on children |
| CN105451235A (en) * | 2015-11-13 | 2016-03-30 | 大连理工大学 | Wireless sensor network intrusion detection method based on background updating |
| CN106056035A (en) * | 2016-04-06 | 2016-10-26 | 南京华捷艾米软件科技有限公司 | Motion-sensing technology based kindergarten intelligent monitoring method |
| CN106127814A (en) * | 2016-07-18 | 2016-11-16 | 四川君逸数码科技股份有限公司 | A kind of wisdom gold eyeball identification gathering of people is fought alarm method and device |
| CN106384479A (en) * | 2016-12-16 | 2017-02-08 | Tcl集团股份有限公司 | Monitoring method and device |
| CN107045623A (en) * | 2016-12-30 | 2017-08-15 | 厦门瑞为信息技术有限公司 | A kind of method of the indoor dangerous situation alarm based on human body attitude trace analysis |
| CN107045623B (en) * | 2016-12-30 | 2020-01-21 | 厦门瑞为信息技术有限公司 | Indoor dangerous condition warning method based on human body posture tracking analysis |
| WO2018218563A1 (en) * | 2017-05-31 | 2018-12-06 | 深圳市永恒丰科技有限公司 | Smart gate opening and closing control method |
| CN107578422B (en) * | 2017-08-01 | 2024-04-09 | 中国农业大学 | Pedestrian tracking detection system and method |
| CN107578422A (en) * | 2017-08-01 | 2018-01-12 | 中国农业大学 | A kind of pedestrian's trace detection system and method |
| CN109567815A (en) * | 2018-10-23 | 2019-04-05 | 电子科技大学 | Remote monitoring method for baby behavior state |
| CN110175512A (en) * | 2019-04-12 | 2019-08-27 | 天津华来科技有限公司 | A kind of scene detection alarm method and device |
| CN110428582A (en) * | 2019-06-28 | 2019-11-08 | 苏州浪潮智能科技有限公司 | A kind of method and system of infantal sleeping protection |
| CN112395921A (en) * | 2019-08-16 | 2021-02-23 | 杭州海康威视数字技术股份有限公司 | Abnormal behavior detection method, device and system |
| CN111797776A (en) * | 2020-07-07 | 2020-10-20 | 云知声智能科技股份有限公司 | Infant monitoring method and device based on posture |
| CN113251558A (en) * | 2021-05-13 | 2021-08-13 | 重庆海尔空调器有限公司 | Infant anti-falling monitoring control method and device, storage medium and air conditioner |
| CN113467325A (en) * | 2021-07-23 | 2021-10-01 | 俞小飞 | Environment monitoring method based on Internet of things |
| CN113467325B (en) * | 2021-07-23 | 2023-03-14 | 俞小飞 | Environment monitoring method based on Internet of things |
| CN113657189A (en) * | 2021-07-26 | 2021-11-16 | 浙江大华技术股份有限公司 | Behavior detection method, electronic device, and computer-readable storage medium |
| CN113657189B (en) * | 2021-07-26 | 2024-02-09 | 浙江大华技术股份有限公司 | Behavior detection method, electronic device, and computer-readable storage medium |
| CN114360209A (en) * | 2022-01-17 | 2022-04-15 | 常州信息职业技术学院 | Video behavior identification security system based on deep learning |
| CN114360209B (en) * | 2022-01-17 | 2023-06-23 | 常州信息职业技术学院 | Video behavior recognition security system based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102074095B (en) | 2013-04-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102074095B (en) | System and method for monitoring infant behaviors | |
| Liu et al. | A fall detection system using k-nearest neighbor classifier | |
| US10540552B2 (en) | Device and method surveilling abnormal behavior using 3D image information | |
| CN105141885B (en) | Carry out the method and device of video monitoring | |
| Munir et al. | Real-time fine grained occupancy estimation using depth sensors on arm embedded platforms | |
| CN101777114B (en) | Intelligent analysis system and intelligent analysis method for video monitoring, and system and method for detecting and tracking head and shoulder | |
| KR101788269B1 (en) | Method and apparatus for sensing innormal situation | |
| KR102260120B1 (en) | Apparatus for Performing Recognition of Activity Based on Deep Learning and Driving Method Thereof | |
| CN106778655B (en) | Human body skeleton-based entrance trailing entry detection method | |
| US20070098222A1 (en) | Scene analysis | |
| CN107662872A (en) | The monitoring system and its monitoring method of passenger conveyor | |
| CN103839373A (en) | Sudden abnormal event intelligent identification alarm device and system | |
| JP2009143722A (en) | Person tracking apparatus, person tracking method and person tracking program | |
| CN101258512A (en) | Method and image evaluation unit for scene analysis | |
| WO2021082112A1 (en) | Neural network training method, skeleton diagram construction method, and abnormal behavior monitoring method and system | |
| CN111241913A (en) | Method, device and system for detecting falling of personnel | |
| Alshbatat et al. | Automated vision-based surveillance system to detect drowning incidents in swimming pools | |
| KR102580434B1 (en) | Dangerous situation detection device and dangerous situation detection method | |
| KR101651410B1 (en) | Violence Detection System And Method Based On Multiple Time Differences Behavior Recognition | |
| Akamatsu et al. | Development of a person counting system using a 3D laser scanner | |
| JP5679760B2 (en) | Intruder detection device | |
| CN109977796A (en) | Trail current detection method and device | |
| Merrouche et al. | Fall detection using head tracking and centroid movement based on a depth camera | |
| KR20160118783A (en) | Method and Apparatus for counting the number of person | |
| CN111144260A (en) | Detection method, device and system of crossing gate |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C56 | Change in the name or address of the patentee | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 214028 Jiangsu New District of Wuxi, Taihu international science and Technology Park Jia Qing 530 building 10 layer Patentee after: Zgmicro Corporation Address before: 214028 Jiangsu New District of Wuxi, Taihu international science and Technology Park Jia Qing 530 building 10 layer Patentee before: WUXI VIMICRO Corp. |
|
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130424 |