CN107396037A - Video frequency monitoring method and device - Google Patents

Video frequency monitoring method and device Download PDF

Info

Publication number
CN107396037A
CN107396037A CN201610322001.7A CN201610322001A CN107396037A CN 107396037 A CN107396037 A CN 107396037A CN 201610322001 A CN201610322001 A CN 201610322001A CN 107396037 A CN107396037 A CN 107396037A
Authority
CN
China
Prior art keywords
world coordinates
class
scene
benchmark
coordinate
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
Application number
CN201610322001.7A
Other languages
Chinese (zh)
Other versions
CN107396037B (en
Inventor
孙杰
韩杰茜
全晓臣
呼志刚
浦世亮
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Hikvision Digital Technology Co Ltd
Original Assignee
Hangzhou Hikvision Digital Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hangzhou Hikvision Digital Technology Co Ltd filed Critical Hangzhou Hikvision Digital Technology Co Ltd
Priority to CN201610322001.7A priority Critical patent/CN107396037B/en
Publication of CN107396037A publication Critical patent/CN107396037A/en
Application granted granted Critical
Publication of CN107396037B publication Critical patent/CN107396037B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/183—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/06—Management of faults, events, alarms or notifications
    • H04L41/0604—Management of faults, events, alarms or notifications using filtering, e.g. reduction of information by using priority, element types, position or time
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30232—Surveillance

Abstract

The embodiments of the invention provide video frequency monitoring method and device.Wherein, this method includes：When detecting moving target during being directed to the video monitoring of target scene, target image coordinate of the moving target under the image three-dimensional coordinate system of place frame of video is obtained；The first transformational relation based on the world three dimensional coordinate system corresponding to default described image three-dimensional system of coordinate and the target scene, calculate the first kind world coordinates corresponding to the target image coordinate, wherein, the first kind world coordinates is coordinate of the moving target in the world three dimensional coordinate system；Obtain the second class world coordinates corresponding to default picture alert locations；Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates meets predetermined alarm rule, if it is, triggering is directed to the alert event of the target scene.The triggering accuracy of alert event can be improved by this programme.

Description

Video frequency monitoring method and device
Technical field
The present invention relates to technical field of video monitoring, more particularly to video frequency monitoring method and device.
Background technology
, it is necessary to guard against some positions in real scene in real life, to prevent people from invading, Pass through, into or leave, wherein it is desired to the position guarded against is referred to as scene alert locations, it is wrapped The type included includes but is not limited to warning line, warning face and warning body.
In the prior art, in order to guard against scene alert locations, the mesh generally where scene alert locations The picture alert locations that 2D is set in video monitoring picture corresponding to scene are marked, set-up mode is to utilize mouse 2D picture alert locations are directly drawn, so, when the moving target and picture that occur in video monitoring image frame When the position relationship change of face alert locations meets predetermined alarm rule, alert event is triggered, specifically, working as Image coordinate (coordinate under two-dimensional image coordinate system) and moving target institute corresponding to picture alert locations The position relationship change of corresponding image coordinate (coordinate under two-dimensional image coordinate system) meets predetermined alarm When regular, just trigger event is alarmed.
But the picture alert locations that prior art is set only are based on 2D planes, without true field Three-dimensional spatial information in scape, i.e., not apart from ground each object of height and real scene relative distance Etc. information, easily there is misjudgment phenomenon.For example, in Fig. 1, directly set on video monitoring picture Picture alert locations are put, the picture alert locations then produce to guard against flower bed region when someone enters flower bed Raw affair alarm, when pedestrian pass through in roadside should not trigger event alarm, still, in Fig. 1 Image to rights Pedestrian, due to perspective relation, judge to come into scene alert locations, so as to which trigger event is alarmed, this Belong to false alarm.
The content of the invention
The purpose of the embodiment of the present invention is to provide video frequency monitoring method and device, to improve touching for alert event Send out accuracy.Concrete technical scheme is as follows：
In a first aspect, the embodiments of the invention provide a kind of video frequency monitoring method, including：
When detecting moving target during being directed to the video monitoring of target scene, the motion mesh is obtained Target image coordinate under the image three-dimensional coordinate system of frame of video where being marked on；
Based on the world three dimensional coordinate corresponding to default described image three-dimensional system of coordinate and the target scene First transformational relation of system, calculates the first kind world coordinates corresponding to the target image coordinate, wherein, The first kind world coordinates is coordinate of the moving target in the world three dimensional coordinate system；
The second class world coordinates corresponding to default picture alert locations is obtained, wherein, the second class generation Boundary's coordinate is the world coordinates corresponding with the image coordinate of the picture alert locations, the second class world Coordinate is determined that the picture alert locations are based on first transformational relation：In advance in the target field Set and in the target scene scene is warned in the video monitoring picture that moving target not be present of scape The corresponding picture position in position is guarded against, wherein, the second class world coordinates is that the scene alert locations exist World coordinates under the world three dimensional coordinate system；
Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates accords with Predetermined alarm rule is closed, if it is, triggering is directed to the alert event of the target scene.
Second aspect, the embodiments of the invention provide a kind of video monitoring apparatus, including：
Target image coordinate determining module, for detecting being directed to during the video monitoring of target scene During moving target, target of the moving target under the image three-dimensional coordinate system on place frame of video is obtained Image coordinate；
First kind world coordinates determining module, for based on default described image three-dimensional system of coordinate and the mesh The first transformational relation of the world three dimensional coordinate system corresponding to scene is marked, it is right to calculate the target image coordinate institute The first kind world coordinates answered, wherein, the first kind world coordinates is the moving target in the world The coordinate of three-dimensional system of coordinate；
Second class world coordinates determining module, for obtaining the second class corresponding to default picture alert locations World coordinates, wherein, the second class world coordinates is relative with the image coordinate of the picture alert locations The world coordinates answered, the second class world coordinates determined based on first transformational relation, the picture Alert locations are：It is set in the video monitoring picture that moving target not be present of the target scene in advance , corresponding with the scene alert locations in target scene picture position, wherein, second class World coordinates is world coordinates of the scene alert locations under the world three dimensional coordinate system；
Position relationship determining module, for judging the first kind world coordinates and the second class world coordinates Position relationship change whether meet predetermined alarm rule, if it is, perform alert event generation module；
The alert event generation module, the alert event of the target scene is directed to for triggering.
In the embodiment of the present invention, when detecting moving target, determine the moving target in world's three-dimensional coordinate First kind world coordinates under system, and judge the image coordinate of the first kind world coordinates and picture alert locations Whether the position relationship change of the second corresponding class world coordinates meets predetermined alarm rule, if it is, touching Hairpin for the target scene alert event, wherein, picture alert locations in advance in target scene not Scene alert locations in set by existing in the video monitoring picture of the moving target and target scene Corresponding picture position.Compared with prior art, in this programme, based under real world three dimensional coordinate Coordinate information carry out position relationship matching, and 2D planes being not based under two-dimensional image coordinate system, reduce False alarm, improve the triggering accuracy rate of alert event.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to implementing The required accompanying drawing used is briefly described in example or description of the prior art, it should be apparent that, describe below In accompanying drawing be only some embodiments of the present invention, for those of ordinary skill in the art, do not paying On the premise of going out creative work, other accompanying drawings can also be obtained according to these accompanying drawings.
The first flow chart for the video frequency monitoring method that Fig. 1 is provided by the embodiment of the present invention；
Second of flow chart of the video frequency monitoring method that Fig. 2 is provided by the embodiment of the present invention；
The third flow chart for the video frequency monitoring method that Fig. 3 is provided by the embodiment of the present invention；
4th kind of flow chart of the video frequency monitoring method that Fig. 4 is provided by the embodiment of the present invention；
The structural representation for the video monitoring apparatus that Fig. 5 is provided by the embodiment of the present invention；
Fig. 6 for calculate binocular camera depth information institute foundation schematic diagram；
The drafting schematic diagram for the picture warning line that Fig. 7 is provided by the embodiment of the present invention；
The picture that Fig. 8 is provided by the embodiment of the present invention guards against the drafting schematic diagram in face；
The picture that Fig. 9 is provided by the embodiment of the present invention guards against the drafting schematic diagram in face；
The drafting schematic diagram for the picture warning body that Figure 10 positions embodiment of the present invention is provided；
Figure 11 is gathers the schematic flow sheet of the extraction process of the moving target in image on binocular camera.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear Chu, it is fully described by, it is clear that described embodiment is only part of the embodiment of the present invention, rather than Whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art are not making creation Property work under the premise of the every other embodiment that is obtained, belong to the scope of protection of the invention.
It is accurate to improve the triggering of alert event the embodiments of the invention provide video frequency monitoring method and device Property.
The video frequency monitoring method provided first below the embodiment of the present invention is introduced.
It should be noted that the video frequency monitoring method executive agent that the embodiment of the present invention is provided can be video Supervising device, also, the video monitoring apparatus can be run in the equipment being connected with image capture device, The image capture device can have moving object detection function, for example：The image capture device can be with For camera, the equipment that should be connected with image capture device can be data hard disc video recorder or terminal, Wherein, the data hard disc video recorder or terminal can obtain target scene from image capture device in real time Video flowing and export obtained video flowing, wherein, the Component units of video flowing are frame of video composition.When So, the video monitoring apparatus can also run on image capture device, wherein, image capture device can have There is moving object detection function.
As shown in figure 1, the video frequency monitoring method that the embodiment of the present invention is provided, may include steps of：
S101, when detecting moving target during being directed to the video monitoring of target scene, obtain the fortune Target image coordinate of the moving-target under the image three-dimensional coordinate system of place frame of video；
Wherein, when detecting moving target during being directed to the video monitoring of target scene, video prison Control device can obtain target image of the moving target under the image three-dimensional coordinate system of place frame of video and sit Mark, wherein it is possible to be realized by any moving object detection mode in the presence of prior art to motion The detection of target.It should be noted that target image coordinate can unique mark go out the moving target, The quantity of target image coordinate does not limit herein；Further, since the detection of moving target is needed to continuous Multiple frame of video analyses, therefore, the frame of video where the moving target are continuous multiple frame of video, accordingly , the target image coordinate is the coordinate information in continuous multiple frame of video.
It will be appreciated by persons skilled in the art that image three-dimensional coordinate system is for image, after Continuous world three dimensional coordinate system is the camera three-dimensional system of coordinate for the target scene for belonging to true environment It is for image capture device (such as camera).Specifically, image three-dimensional coordinate system is with picture Element is the coordinate system of unit demarcation, and the coordinate of its XY axle is the sign of pixel in image, and the coordinate of Z axis is The depth value of the depth information of image, i.e. pixel；Camera three-dimensional system of coordinate be using image capture device in The coordinate system of heart setting, the origin of the coordinate system is image capture device position, and it is objective generation The absolute coordinate on boundary；World three dimensional coordinate system is also referred to as true or real world three-dimensional system of coordinate, and it is objective generation The absolute coordinate on boundary.In addition, it is necessary to, it is emphasized that in the embodiment of the present invention, the original of world three dimensional coordinate system Point is image capture device position in the subpoint of horizontal plane, i.e., the throwing of origin in camera three-dimensional system of coordinate Shadow point, camera three-dimensional system of coordinate can be with identical with the setting means of three axles of world three dimensional coordinate system.
It should be noted that the mode of the acquisition of the depth information of image can be：Using binocular camera conduct Image capture device, the depth information of each pixel is obtained using prior art, i.e., each pixel phase To the actual range of binocular camera；Or the acquisition pattern of the depth information of image can be：Utilize physics The modes such as sensor TOF (Time of Flight), Structure Light obtain the depth map corresponding to image, So as to obtain the depth information of image.By taking binocular camera technology as an example, technique of binocular stereoscopic vision is to imitate The vision system of human eye, the target image that two video cameras by obtaining diverse location are shot, calculates target and (applies Q points shown in 6) parallax in the picture, so as to get the depth information of target；Wherein, binocular phase Machine principle is as shown in fig. 6, in the CCD image planes of left and right, it is known that xleftAnd xrightDifference, i.e. parallax, it is known that base Line b and focal length f, you can try to achieve the distance l of Q and binocular camera, i.e. depth value, calculation formula is as follows
According to above-mentioned principle, it can be deduced that in frame of video each pixel relative to binocular camera it is actual away from From the depth information of that is, each pixel can be acquired.
In addition, for image capture device is binocular camera, extraction gathers the moving target in picture Basic thought is：Using binocular ranging algorithm, the characteristic point of binocular image is extracted, depth map is calculated, goes forward side by side Row depth background modeling and depth foreground extraction, specifically, in flow as shown in Figure 11, characteristic point meter Calculate step to be used to extract the characteristic point in the image of binocular two (left figure, right figure), Feature Points Matching step is used for For the selected characteristic point in left figure and maximal correlation position of this feature point in right figure is matched, generates depth Figure step is used to be to calculate final depth map using match information, calculates three-dimensional data step and is used for using deep Spend figure information and combine binocular calibration parameter, calculate the three-dimensional coordinate data of target；Depth background modeling step Profit is used for the background in filter depth figure, generation depth prospect step for carrying out background modeling to depth map The prospect i.e. extraction of moving target is carried out to depth map with background model.It is understood that extracting fortune Target image of the moving target under the image three-dimensional coordinate system of place frame of video can be determined after moving-target Coordinate.
It will be appreciated by persons skilled in the art that for image capture device is binocular camera, obtain The specific implementation side of target image coordinate of the moving target under the image three-dimensional coordinate system of place frame of video Formula can use prior art；Be physical sensors TOF (Time of Flight) for image capture device, For Structure Light etc., mesh of the moving target under the image three-dimensional coordinate system of place frame of video is obtained The specific implementation of logo image coordinate can also use prior art；Image capture device is on moving target Detection mode can use any prior art realize, do not limit herein.
S102, based on the world three dimensional coordinate corresponding to the default image three-dimensional coordinate system and the target scene First transformational relation of system, calculates the first kind world coordinates corresponding to the target image coordinate；
Wherein, the first kind world coordinates is coordinate of the moving target in the world three dimensional coordinate system.
In order to improve the triggering accuracy of alert event, the embodiment of the present invention, which uses, judges moving target and scene Position relationship of the alert locations in real target scene changes to determine whether to trigger alert event, that is, sentences Disconnected coordinate of the moving target under world's three-dimensional system of coordinate is with scene alert locations under world's three-dimensional coordinate The position relationship of coordinate come determine whether trigger alert event.Based on above-mentioned thought, the moving target is being obtained After target image coordinate under the image three-dimensional coordinate system of place frame of video, default be somebody's turn to do can be primarily based on First transformational relation of image three-dimensional coordinate system and the world three dimensional coordinate system corresponding to the target scene, calculate First kind world coordinates corresponding to the target image coordinate, and then utilize the first kind world be calculated Coordinate performs follow-up processing.
Wherein, first transformational relation is public by the conversion of image three-dimensional coordinate system and camera three-dimensional system of coordinate Formula, and the conversion formula of the camera three-dimensional system of coordinate and world three dimensional coordinate system characterize；
Wherein, the conversion formula of the image three-dimensional coordinate system and camera three-dimensional system of coordinate is：
The conversion formula of the camera three-dimensional system of coordinate and world three dimensional coordinate system：
Wherein, (x, y, z) be the image three-dimensional coordinate system under image coordinate, (Xc, Yc, Zc) it is the phase Camera coordinates under machine three-dimensional system of coordinate, (X, Y, Z) are the world coordinates under the world three dimensional coordinate system, The constant value that x1, x2, x3, x4 and w are tried to achieve for the image capture device based on collection frame of video, θ is should The angle of pitch of image capture device, ψ are the inclination angle of the image capture device, and Hcam is that the IMAQ is set The standby height away from ground；The origin of the world three dimensional coordinate system is the projection of the origin of camera three-dimensional system of coordinate.
It is understood that after the first above-mentioned transformational relation determines, it is determined in the target image coordinate Afterwards, the first kind world coordinates corresponding to each target image coordinate can be calculated.
S103, obtain the second class world coordinates corresponding to default picture alert locations；
Wherein, the second class world coordinates is sat for the world corresponding with the image coordinate of the picture alert locations Mark, the second class world coordinates are determined that the picture alert locations are based on first transformational relation：In advance In set by the video monitoring picture that moving target not be present of the target scene and target scene The corresponding picture position of scene alert locations, wherein, the second class world coordinates is that the scene guards against position Put the world coordinates under the world three dimensional coordinate system.Due to needing to be guarded against with scene by detecting moving target Position relationship of the position in real target scene improves the triggering accuracy of alert event, therefore, can To obtain the second class world coordinates corresponding to default picture alert locations.
It is understood that be to determine due to default picture alert locations, therefore, default picture police Guarding against the second class world coordinates corresponding to the image coordinate of position can precalculate to obtain, with video monitoring During directly utilize；Certainly, the second class world corresponding to the image coordinate of default picture alert locations Coordinate can be calculated during first time video monitoring, can be direct during follow-up video monitoring Utilize, this is also rational.Wherein, because the image coordinate of the required picture alert locations utilized can The picture alert locations are uniquely determined out, therefore, the image of the required picture alert locations utilized is sat Target quantity does not limit herein；Also, because the second class world coordinates and required picture utilized are guarded against The image coordinate of position has one-to-one corresponding property, and therefore, the quantity of the second class world coordinates does not limit herein It is fixed.
Specifically, the picture alert locations can be：The figure determined based on the second class benchmark image coordinate Image position, the second class benchmark image coordinate are：Pixel conversion fortune is performed to first kind benchmark image coordinate The image coordinate of gained is calculated, the first kind benchmark image coordinate is the video prison that moving target be present Control the image coordinate included by virtual projection in picture；Wherein, the pixel translation operation is to utilize this first turn Change the conversion fortune of relation, the height relationships of the scene alert locations and real projection and the second transformational relation Calculate；Wherein, second transformational relation is transformational relation with the first transformational relation contrary, the real projection For projection of the scene alert locations on the predetermined datum plane in the target scene, the virtual projection is should Real projection is in the position in the video monitoring picture of moving target being present.It is understood that this is pre- It can be ground to determine datum plane, be not limited thereto certainly.
It is emphasized that the picture alert locations have shape pair with the scene alert locations in target scene Ying Xing.Specifically, for target scene, the scene alert locations set by it can be：Scene is warned Line is guarded against, accordingly, picture alert locations are picture warning line；Scene alert locations set by it can be： Scene guards against face, and accordingly, picture alert locations are picture warning line；Scene alert locations set by it Can be：Scene guards against body, and accordingly, picture alert locations are that picture guards against body.
It is clear in order to be laid out, the setting up procedure of all kinds of picture alert locations is subsequently introduced in conjunction with specific embodiments.
S104, judges whether the position relationship change of the first kind world coordinates and the second class world coordinates accords with Predetermined alarm rule is closed, if it is, performing S105；
After first kind world coordinates and the second class world coordinates is obtained, it can be determined that the first kind world is sat Whether the change of the position relationship of mark and the second class world coordinates meets predetermined alarm rule, i.e. the moving target With position relationship of the scene alert locations in target scene, and performed according to different judged result different Operation.Specifically, when the position relationship for judging the first kind world coordinates and the second class world coordinates becomes When change meets predetermined alarm rule, show that the trigger condition of alert event is satisfied, therefore, can perform S105；And when the position relationship change for judging the first kind world coordinates and the second class world coordinates is not inconsistent When closing predetermined alarm rule, show that the trigger condition of alert event is not satisfied, therefore, can not deal with.
S105, triggering are directed to the alert event of the target scene.
Meet in the position relationship change for judging the first kind world coordinates and the second class world coordinates pre- When determining alarm rule, the alert event for being directed to the target scene can be triggered.Wherein, alert event is defeated Go out the form embodiment of the present invention not limit, for example：Concrete form can be：Audible alarm, short message Alarm, dialog box alarm etc..
In the embodiment of the present invention, when detecting moving target, determine the moving target in world's three-dimensional coordinate First kind world coordinates under system, and judge the image coordinate of the first kind world coordinates and picture alert locations Whether the position relationship change of the second corresponding class world coordinates meets predetermined alarm rule, if it is, touching Hairpin for the target scene alert event, wherein, picture alert locations in advance in target scene not Scene alert locations in set by existing in the video monitoring picture of the moving target and target scene Corresponding picture position.Compared with prior art, in this programme, based under real world three dimensional coordinate Coordinate information carry out position relationship matching, and 2D planes being not based under two-dimensional image coordinate system, reduce False alarm, improve the triggering accuracy rate of alert event.
Below so that scene alert locations are scene warning line as an example, the video provided the embodiment of the present invention is supervised Prosecutor method is introduced.
It should be noted that the video frequency monitoring method executive agent that the embodiment of the present invention is provided can be video Supervising device, also, the video monitoring apparatus can be run in the equipment being connected with image capture device, The image capture device can have moving object detection function, for example：The image capture device can be with For camera, the equipment that should be connected with image capture device can be data hard disc video recorder or terminal, Wherein, the data hard disc video recorder or terminal can obtain target scene from image capture device in real time Video flowing and export obtained video flowing, wherein, the Component units of video flowing are frame of video composition.When So, the video monitoring apparatus can also run on image capture device, wherein, image capture device can have There is moving object detection function.
As shown in Fig. 2 the video frequency monitoring method that the embodiment of the present invention is provided, may include steps of：
S201, when detecting moving target during being directed to the video monitoring of target scene, obtain the fortune Target image coordinate of the moving-target under the image three-dimensional coordinate system of place frame of video；
S202, based on the world three dimensional coordinate corresponding to the default image three-dimensional coordinate system and the target scene First transformational relation of system, calculates the first kind world coordinates corresponding to the target image coordinate；
Wherein, the first kind world coordinates is coordinate of the moving target in the world three dimensional coordinate system.
In the present embodiment, S201-S202 is similar with the content of S101-S102 in above-described embodiment, does not do herein Repeat.
S203, obtain the second class world coordinates corresponding to default picture warning line；
Wherein, the second class world coordinates is sat for the world corresponding with the image coordinate of the picture warning line Mark, the second class world coordinates are determined that the picture warning line is based on first transformational relation：Exist in advance In set by the video monitoring picture that moving target not be present of the target scene and target scene The corresponding picture position of scene warning line, wherein, the second class world coordinates is the scene warning line at this World coordinates under world three dimensional coordinate system.
Due to needing to close by detecting position of the moving target with scene warning line in real target scene It is to improve the triggering accuracy of alert event, thus, it is possible to obtain corresponding to default picture warning line Second class world coordinates.
It is understood that be to determine due to default picture warning line, therefore, default picture warning Line image coordinate corresponding to the second class world coordinates can precalculate to obtain, with video monitoring mistake Directly utilized in journey；Certainly, the second class world coordinates corresponding to the image coordinate of default picture warning line It can be calculated during first time video monitoring, can be directly sharp during follow-up video monitoring With this is also rational.Wherein, because the image coordinate of the required picture warning line utilized can be unique The picture warning line is determined, therefore, the quantity of the image coordinate of the required picture warning line utilized Do not limit herein；Also, due to the second class world coordinates and the image of the required picture warning line utilized Coordinate has one-to-one corresponding property, and therefore, the quantity of the second class world coordinates does not limit herein.
Specifically, the picture warning line is specially：The image determined based on the second class benchmark image coordinate Position, the second class benchmark image coordinate are：Pixel translation operation institute is performed to first kind benchmark image coordinate The image coordinate obtained, the first kind benchmark image coordinate do not exist in the video monitoring picture of moving target for this Image coordinate included by virtual projection；Wherein, the pixel translation operation be using first transformational relation, The height relationships of the scene warning line and real projection and the translation operation of the second transformational relation；Wherein, Second transformational relation is the transformational relation with the first transformational relation contrary, and the real projection is warned for the scene Projection of the line on the predetermined datum plane in the target scene is guarded against, the virtual projection is the real projection at this The position in the video monitoring picture of moving target be present.It is understood that the predetermined datum plane can Think ground, be not limited thereto certainly.
In order to which solution layout understands, subsequently the setting up procedure of the picture warning line corresponding to scene warning line is entered Row is discussed in detail.
S204, judge whether the position relationship change of the first kind world coordinates and the second class world coordinates is full Foot：The first kind world coordinates is not included in the range of being formed from the second class world coordinates, to including this First kind world coordinates arrives again does not include the first kind world coordinates, if it is, performing S205；
Wherein, after first kind world coordinates and the second class world coordinates is obtained, it can be determined that the first kind Whether the position relationship change of world coordinates and the second class world coordinates meets：From the second class world coordinates Do not include the first kind world coordinates in the range of forming, do not wrap to being arrived again including the first kind world coordinates The first kind world coordinates is included, and different operations is performed according to different judged results.Sentence specifically, working as Break and the position relationship change satisfaction of the first kind world coordinates and the second class world coordinates：From second class World coordinates does not include the first kind world coordinates, to including the first kind world coordinates in the range of forming Arriving again does not include the first kind world coordinates, shows that moving target has passed through the scene in real target scene Warning line, that is, position relationship change meet it is predetermined pass through warning line alarm rule, at this point it is possible to perform S205；When the position relationship change for judging the first kind world coordinates and the second class world coordinates is discontented Foot：The first kind world coordinates is not included in the range of being formed from the second class world coordinates, to including this First kind world coordinates arrives again does not include the first kind world coordinates, shows to move mesh in real target scene Mark does not pass through the scene warning line, that is, position relationship change does not meet and makes a reservation for pass through warning line alarm rule Then, at this point it is possible to not process.
It is emphasized that the continuous multiple frame of video of frame of video where moving target, in order to ensure to judge knot The validity of fruit, first kind world coordinates and second can be judged according to acquisition time principle near after first remote Whether the position relationship change of class world coordinates meets to make a reservation for pass through warning line alarm rule.
S205, triggering are directed to the alert event of the target scene.
In the present embodiment, S205 is similar with the S105 of above-described embodiment, will not be described here.
Compared with prior art, in this programme, carried out based on the coordinate information under real world three dimensional coordinate Position relationship matches, and the 2D planes being not based under two-dimensional image coordinate system, reduces false alarm, improves The triggering accuracy rate of alert event.
Citing introduction is carried out to the setting up procedure of picture warning line corresponding to scene warning line below.
The setting up procedure of picture warning line can include corresponding to the scene warning line：
A1：Determine that user is set in the video monitoring picture that moving target not be present for be directed to target scene Virtual projection put, related to the scene warning line；
Wherein, scene warning line can be to be suspended in aerial line segment, or the line segment resting on the ground. Before picture warning line is drawn, user is needed using ground as predetermined datum plane, in video monitoring picture The virtual projection related to the scene warning line is drawn, wherein, it is somebody's turn to do the virtual projection related to scene warning line Relative position relation with other objects in video monitoring picture is equal to：It is true corresponding to the scene warning line The relative position relation with other corresponding objects in target scene is projected in fact.Specifically, such as Fig. 7 (a)- The drawing process of the virtual projection related to scene warning line given by Fig. 7 (c), wherein, user can be with In it the video monitoring picture of moving target not be present, using ground as predetermined datum plane, mouse is placed on In video monitoring picture, mouse is changed into cross foresight, and expression can start setting up related to the scene warning line Virtual projection starting point, mouse is moved to and wishes to set the virtual projection related to the scene warning line The position of starting point, left mouse button is clicked on, complete the setting of starting point, mobile mouse, the starting point and mouse There is the connecting line of dotted line type, the Dynamic Announce virtual projection related to the scene warning line between mark pointer Line segment position, by mouse be moved to wish set end point position, click on left mouse button, complete The setting of end point, so far user the virtual projection related to the scene warning line, such as Fig. 7 are set out manually (c) shown in.
It is emphasized that the video monitoring apparatus can have virtual projection to set function, so as to obtain user The virtual projection related to the scene warning line can be set manually on video monitoring picture.It is in addition, predetermined Datum plane selection ground, and the not unfixed object of metope or desktop etc structural form, are ensured with this The validity of set virtual projection.
B1：Based on first transformational relation, calculate included by the virtual projection related to the scene warning line First kind benchmark world coordinates corresponding to first kind benchmark image coordinate, wherein, the first kind benchmark image Coordinate uniquely determines out the virtual projection related to the scene warning line；
After the virtual projection related to the scene warning line is determined, in order to realize painting for picture warning line Coordinate points on the virtual projection related to the scene warning line can be transformed into world three dimensional coordinate system by system Under, so as to obtain the related coordinate information of the real projection of the scene warning line.Specifically, it can be based on being somebody's turn to do First transformational relation, calculate and sat on the first kind benchmark image included by the virtual projection of the picture warning line The corresponding first kind benchmark world coordinates of mark, wherein, the first kind benchmark world coordinates can uniquely determine Go out the real projection of the scene warning line.
It should be noted that the first kind benchmark image included by the virtual projection related to the scene warning line Coordinate can comprise at least：Two end points of the virtual projection related to the scene warning line, as long as ensureing logical The virtual projection related to the scene warning line can be uniquely determined out by crossing first kind benchmark image coordinate, Particular number does not limit herein.In addition, it is necessary to the, it is emphasized that virtual throwing related to the scene warning line Angle of the shadow in video monitoring picture relative to other objects and the angle difference in ground level direction It is equal to：The real projection of the scene warning line is in target scene relative to other corresponding things in target scene The angle of body and the angle in ground level direction.
C1：The scene warning line is obtained to the distance value of its real projection；
Wherein, distance value can be what user was manually entered.Wherein, the scene warning line can be parallel to Ground or the line segment of uneven ground, specifically, when the scene warning line is the line segment parallel to ground, should Scene warning line to its real projection distance value be one, that is to say, that on the scene warning line institute a little To the distance value all same of the respective point of real projection；And when the scene warning line is the line for being not parallel to ground Duan Shi, the distance value of the scene warning line to its real projection can be multiple, that is to say, that the scene is warned The distance value for guarding against the respective point for arriving real projection a little on line is incomplete same.
D1：Based on the distance value, the second class benchmark world corresponding to the first kind benchmark world coordinates is determined Coordinate, wherein, the difference in height of the second class benchmark world coordinates and the first kind benchmark world coordinates is should be away from From value；
After the first kind benchmark world coordinates corresponding to distance value and real projection determines, it may be determined that go out this The second class benchmark world coordinates corresponding to scene warning line, wherein, the first kind benchmark world coordinates institute is right The the second class benchmark world coordinates answered can uniquely determine out the scene warning line.
E1：Based on second transformational relation, the second class base corresponding to the second class benchmark world coordinates is calculated Quasi- image coordinate；
After the second class benchmark world coordinates is determined, second transformational relation can be based on, calculate this The second class benchmark image coordinate corresponding to two class benchmark world coordinates, that is, determine frame out warning line in image The image coordinate of home position under three-dimensional coordinate, wherein, the second class benchmark image coordinate can uniquely determine Go out the picture warning line corresponding to the scene warning line.Wherein, first kind benchmark world coordinates, the second class base The quantity identical of quasi- world coordinates, first kind benchmark image coordinate and the second class benchmark image coordinate, and have There is corresponding relation.
F1：Based on the second class benchmark image coordinate, painted in the video monitoring picture that moving target not be present Make the picture warning line corresponding to the scene warning line.
After the second class benchmark image coordinate is determined, painted in the video monitoring picture that moving target not be present The picture warning line corresponding to the scene warning line is made, so far completes the picture corresponding to the scene warning line The setting up procedure of warning line, set result can be as shown in 7 (d) (e).
It is emphasized that picture warning line corresponding to the scene warning line given by above-mentioned A1-F1 Setting up procedure is merely exemplary, should not form the restriction to the embodiment of the present invention.
Below so that scene alert locations are scene warning face as an example, the video provided the embodiment of the present invention is supervised Prosecutor method is introduced.
It should be noted that the video frequency monitoring method executive agent that the embodiment of the present invention is provided can be video Supervising device, also, the video monitoring apparatus can be run in the equipment being connected with image capture device, The image capture device can have moving object detection function, for example：The image capture device can be with For camera, the equipment that should be connected with image capture device can be data hard disc video recorder or terminal, Wherein, the data hard disc video recorder or terminal can obtain target scene from image capture device in real time Video flowing and export obtained video flowing, wherein, the Component units of video flowing are frame of video composition.When So, the video monitoring apparatus can also run on image capture device, wherein, image capture device can have There is moving object detection function.
As shown in figure 3, the video frequency monitoring method that the embodiment of the present invention is provided, may include steps of：
S301, when detecting moving target during being directed to the video monitoring of target scene, obtain the fortune Target image coordinate of the moving-target under the image three-dimensional coordinate system of place frame of video；
S302, based on the world three dimensional coordinate corresponding to the default image three-dimensional coordinate system and the target scene First transformational relation of system, calculates the first kind world coordinates corresponding to the target image coordinate；
In the present embodiment, S301-S302 is similar with the content of S101-S102 in above-described embodiment, does not do herein Repeat.
Wherein, the first kind world coordinates is coordinate of the moving target in the world three dimensional coordinate system.
S303, obtain the second class world coordinates corresponding to default picture warning face；
Wherein, the second class world coordinates is that the world corresponding with the image coordinate in picture warning face is sat Mark, the second class world coordinates are determined that picture warning face is based on first transformational relation：Exist in advance In set by the video monitoring picture that moving target not be present of the target scene and target scene The corresponding picture position in scene warning face, wherein, the second class world coordinates is that the scene guards against face at this World coordinates under world three dimensional coordinate system.
Due to needing to guard against position pass of the face in real target scene by detecting moving target and scene It is to improve the triggering accuracy of alert event, thus, it is possible to obtain corresponding to default picture warning face Second class world coordinates.
It is understood that be to determine due to default picture warning face, therefore, default picture warning The second class world coordinates corresponding to the image coordinate in face can precalculate to obtain, with video monitoring process In directly utilize；Certainly, the second class world coordinates corresponding to the image coordinate in default picture warning face can To be calculated during first time video monitoring, can directly be utilized during follow-up video monitoring, This is also rational.Wherein, because the image coordinate in the required picture warning face utilized can uniquely determine Go out the picture warning face, therefore, the quantity of the image coordinate in the required picture warning face utilized is herein Do not limit；Also, due to the second class world coordinates and the image coordinate in the required picture warning face utilized With one-to-one corresponding property, therefore, the quantity of the second class world coordinates does not limit herein.
Specifically, the picture warning mask body is：The figure determined based on the second class benchmark image coordinate Image position, the second class benchmark image coordinate are：Pixel conversion fortune is performed to first kind benchmark image coordinate The image coordinate of gained is calculated, the first kind benchmark image coordinate is：The video monitoring that moving target be present is drawn Image coordinate in face included by virtual projection；Wherein, the pixel translation operation is to utilize the described first conversion Relation, scene warning face and the height relationships of real projection and the translation operation of the second transformational relation； Wherein, second transformational relation is the transformational relation with the first transformational relation contrary, and the real projection is should Scene guards against projection of the face on the predetermined datum plane in the target scene, and the virtual projection is the true throwing Shadow is in the position in the video monitoring picture of moving target being present.It is understood that the predetermined benchmark Plane can be ground, be not limited thereto certainly.
In order to which solution layout understands, the setting up procedure for subsequently guarding against the picture corresponding to scene warning face face enters Row is discussed in detail.
S304, judge whether the position relationship change of the first kind world coordinates and the second class world coordinates is full Foot：Do not include in the range of being formed from the second class world coordinates the first kind world coordinates to including this A kind of world coordinates, if it is, performing S305；
Wherein, after first kind world coordinates and the second class world coordinates is obtained, it can be determined that the first kind Whether the position relationship change of world coordinates and the second class world coordinates meets：From the second class world coordinates Do not include the first kind world coordinates in the range of forming to including the first kind world coordinates, and according to not Same judged result performs different operations.Specifically, when judge the first kind world coordinates with this second The position relationship change of class world coordinates meets：Do not include in the range of being formed from the second class world coordinates The first kind world coordinates shows moving target in real target scene to the first kind world coordinates is included Scene warning face is entered, that is, position relationship change meets the warning face of being destined to go into alarm rule, this When, S305 can be performed；And when the position for judging the first kind world coordinates and the second class world coordinates Relationship change is unsatisfactory for：Do not include the first kind world in the range of forming from the second class world coordinates to sit Mark including the first kind world coordinates, show that moving target is introduced into scene police in real target scene Ring face, that is, position relationship change do not meet and are destined to go into warning line alarm rule, at this point it is possible to not do Processing.
For scene alert locations are scene warning face, in another implementation of the embodiment of the present invention In, whether the position relationship change for judging the first kind world coordinates and the second class world coordinates meets Predetermined alarm rule, it can include：
Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates meets：From The second class world coordinates does not include the first kind world coordinates, to including the first kind in the range of forming World coordinates arrives again does not include the first kind world coordinates, if it is, showing the first kind world coordinates and being somebody's turn to do The position relationship change of second class world coordinates, which meets, to be made a reservation for pass through warning face alarm rule, can be performed S305, otherwise, do not deal with；
Or
Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates meets：From The second class world coordinates includes the first kind world coordinates to not including the first kind in the range of forming World coordinates, if it is, showing that the position relationship of the first kind world coordinates and the second class world coordinates becomes Change, which meets, to be made a reservation for leave warning face alarm rule, at this point it is possible to perform S305, otherwise, is not dealt with；
Or
Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates meets：Should Second class world coordinates includes the first kind world coordinates in the range of forming, if it is, showing described The change of the position relationship of a kind of world coordinates and the second class world coordinates meets predetermined invasion warning face report Police regulations then, at this point it is possible to perform S305, otherwise, are not dealt with.
It is emphasized that the continuous multiple frame of video of frame of video where moving target, in order to ensure to judge knot The validity of fruit, first kind world coordinates and second can be judged according to acquisition time principle near after first remote Whether the position relationship change of class world coordinates meets to be destined to go into warning face alarm rule, makes a reservation for pass through warning Line face police regulations then, predetermined leave warning face alarm rule or predetermined invasion warning face alarm rule.
S305, triggering are directed to the alert event of the target scene.
In the present embodiment, S305 is similar with the S105 of above-described embodiment, will not be described here.
Compared with prior art, in this programme, carried out based on the coordinate information under real world three dimensional coordinate Position relationship matches, and the 2D planes being not based under two-dimensional image coordinate system, reduces false alarm, improves The triggering accuracy rate of alert event.
The setting up procedure for guarding against picture corresponding to scene warning face face below carries out citing introduction.
Specifically, in one embodiment of the invention, scene warning face is the scene parallel to ground Warning face, the setting up procedure in the picture warning face corresponding to the scene warning face are：
A2：Determine that user is set in the video monitoring picture that moving target not be present for be directed to target scene Virtual projection put, related to scene warning face；
Wherein, scene warning face can suspend in the air or rest on the ground.Before picture warning face is drawn, User is needed using ground as predetermined datum plane, is drawn in video monitoring picture related to scene warning face Virtual projection, wherein, to the related virtual projection in scene warning face in video monitoring picture with other things The relative position relation of body is equal to：Real projection corresponding to the scene warning face is in target scene and corresponding The relative position relation of other objects.Specifically, being guarded against with scene as given by Fig. 8 (a)-Fig. 8 (f) The drawing process of the related virtual projection in face, wherein, user can be in the video monitoring that moving target not be present In picture, using ground as predetermined datum plane, mouse is placed in frame of video picture, mouse is changed into cross Foresight, represent that the starting endpoint of the virtual projection related to scene warning face can be started setting up, by mouse The position for the starting endpoint for wishing to set the virtual projection related to scene warning face is moved to, it is left to click on mouse Key, complete the setting of starting endpoint；Mobile mouse, it is (empty to there is connecting line between starting endpoint and mouse pointer Line), the side position of the Dynamic Announce virtual projection related to scene warning face；By mouse be moved to Second end points of the related virtual projection in scene warning face, left mouse button is clicked on, second end points of completion Draw；The like, complete to guard against the drafting of the related all end points of virtual projection in face with scene, when with field When last end points of the related virtual projection in scape warning face overlaps with starting endpoint, plane closed loop is formed, Complete the drafting of the virtual projection related to scene warning face.
It is emphasized that the video monitoring apparatus can have virtual projection to set function, so as to obtain user The virtual projection related to scene warning face can be set manually on video monitoring picture.It is in addition, predetermined Datum plane selection ground, and the not unfixed object of metope or desktop etc structural form, are ensured with this The validity of set virtual projection.
B2：Based on first transformational relation, the included by guarding against the related virtual projection in face with scene is calculated First kind benchmark world coordinates corresponding to a kind of benchmark image coordinate, wherein, the first kind benchmark image is sat Mark uniquely determines out the virtual projection related to scene warning face；
After the virtual projection related to scene warning face is determined, in order to realize the drafting in picture warning face, Coordinate points on the virtual projection related to scene warning face can be transformed under world three dimensional coordinate system, So as to obtain the related coordinate information of the real projection in scene warning face.Specifically, can be based on this first Transformational relation, calculate the first kind benchmark image coordinate institute included by the virtual projection in picture warning face Corresponding first kind benchmark world coordinates, wherein, the first kind benchmark world coordinates can uniquely determine out should Scene guards against the real projection in face.
It should be noted that the quantity of first kind benchmark image coordinate does not limit herein, as long as ensureing to pass through First kind benchmark image coordinate can uniquely determine out the virtual projection related to scene warning face.Separately Outside, it is emphasized that, the virtual projection related to scene warning face is in video monitoring picture relative to it The angle of his object and angle in ground level direction are respectively equivalent to：Scene guards against the real projection in face The angle relative to other corresponding objects and the angle in ground level direction in target scene.
It is understood that scene warning face can be circular or square etc., this is all rational.
C2：Obtain the scene and guard against face to the distance value of its real projection；
Wherein, distance value can be what user was manually entered.Wherein, the scene warning face can be parallel to Ground or the plane of uneven ground, specifically, when the scene guards against face parallel to ground, scene warning Face to its real projection distance value be one, that is to say, that scene arrives a little real projection on warning face Respective point distance value all same；And when scene warning face is not parallel to ground, scene warning face Distance value to its real projection is multiple, that is to say, that scene arrives a little real projection on warning face The distance value of respective point is incomplete same.
D2：Based on the distance value, the second class benchmark world corresponding to the first kind benchmark world coordinates is determined Coordinate, wherein, the difference in height of the second class benchmark world coordinates and the first kind benchmark world coordinates is should be away from From value；
After the first kind benchmark world coordinates corresponding to distance value and real projection determines, it may be determined that go out this The second class benchmark world coordinates corresponding to scene warning face, wherein, the second class benchmark world coordinates can be only One determines the scene warning line.
E2：Based on second transformational relation, the second class base corresponding to the second class benchmark world coordinates is calculated Quasi- image coordinate；
After the second class benchmark world coordinates is determined, second transformational relation can be based on, calculate this The second class benchmark image coordinate corresponding to two class benchmark world coordinates, that is, determine that frame out guards against face in image The image coordinate of home position under three-dimensional coordinate.Wherein, the second class benchmark image coordinate can uniquely determine Go out the picture warning face corresponding to the scene warning face.Wherein, first kind benchmark world coordinates, the second class base The quantity identical of quasi- world coordinates, first kind benchmark image coordinate and the second class benchmark image coordinate, and have There is corresponding relation.
F2：Based on the second class benchmark image coordinate, in it the video monitoring picture of moving target be present in this Draw the picture warning face corresponding to the scene warning face.
After the second class benchmark image coordinate is determined, painted in the video monitoring picture that moving target not be present The picture warning face corresponding to the scene warning face is made, so far completes the setting up procedure in picture warning face, if Putting result can be as shown in 8 (g) (h).
Specifically, in another embodiment of the present invention, scene warning face is the scene perpendicular to ground Warning face, the setting up procedure in the picture warning face corresponding to the scene warning face can include：
A3：Determine that user is set in the video monitoring picture that moving target not be present for be directed to target scene Virtual projection put, related to scene warning face；
Wherein, scene warning face can suspend in the air or one side and ground.Drawing picture warning face Before, user is needed using ground as predetermined datum plane, is drawn in video monitoring picture and is guarded against with the scene The related virtual projection in face, wherein, the virtual projection related to scene warning face in video monitoring picture and The relative position relation of other objects is equal to：Real projection is in target scene corresponding to the scene warning face With the relative position relation of other corresponding objects.Specifically, as given by Fig. 9 (a)-Fig. 9 (c) with field The drawing process of the related virtual projection in scape warning face, wherein, user can not have regarding for moving target In frequency monitored picture, using ground as predetermined datum plane, mouse is placed in frame of video picture, mouse becomes For cross foresight, represent that the starting endpoint of the virtual projection related to scene warning face can be started setting up； Mouse is moved to the position for the starting endpoint for wishing to set the virtual projection related to scene warning face, clicked on Left mouse button, complete starting endpoint and set；Mobile mouse, is connected between starting endpoint and mouse pointer Line (dotted line), the line segment position of the Dynamic Announce virtual projection related to scene warning face；By mouse Second end points of the virtual projection related to scene warning face is moved to, clicks on left mouse button, completion and field The drafting of second end points of the related virtual projection in scape warning face, then complete to warn perpendicular to ground and scene The related virtual projection in ring face is drawn.
It is emphasized that the video monitoring apparatus can have virtual projection to set function, so as to obtain user The virtual projection related to scene warning face can be set manually on video monitoring picture.It is in addition, predetermined Datum plane selection ground, and the not unfixed object of metope or desktop etc structural form, are ensured with this The validity of set virtual projection.
B3：Based on first transformational relation, calculate included by the virtual projection related to scene warning face First kind benchmark world coordinates corresponding to first kind benchmark image coordinate, wherein, the first kind benchmark image Coordinate uniquely determines out the virtual projection related to scene warning face；
After the virtual projection related to scene warning face is determined, in order to realize the drafting in picture warning face, Coordinate points on the virtual projection related to scene warning face can be transformed under world three dimensional coordinate system, So as to obtain the related coordinate information of the real projection in scene warning face.Specifically, can be based on this first Transformational relation, calculate the first kind benchmark image coordinate institute included by the virtual projection in picture warning face Corresponding first kind benchmark world coordinates, wherein, the first kind benchmark world coordinates can uniquely determine out should Scene guards against the real projection in face.
It should be noted that the quantity of first kind benchmark image coordinate does not limit herein, as long as ensureing to pass through First kind benchmark image coordinate can uniquely determine out the virtual projection related to scene warning face.Separately Outside, it is emphasized that, the virtual projection related to scene warning face is in video monitoring picture relative to it The angle of his object and angle in ground level direction are respectively equivalent to：Scene guards against the real projection in face The angle relative to other corresponding objects and the angle in ground level direction in target scene.
Wherein, the left side in face should be guarded against perpendicular to the scene on ground and the right is each perpendicular to ground.
C3：The top in scene warning face is obtained to the first kind distance value of its real projection, and is arrived below Second class distance value of its real projection；
Wherein, first kind distance value and the second class distance value can be what user was manually entered.Wherein, this Scape warning face bottom and upper segment can parallel to ground can also not parallel ground, specifically, working as the scene The top in warning face or below parallel to ground when, corresponding distance value be one, that is to say, that top Or the distance value of the following respective point to real projection is identical；And when the scene warning face top or under While when being not parallel to ground, corresponding distance value is multiple, that is to say, that top is following to true throwing The distance value of the respective point of shadow is incomplete same.
D3：Determine that the second class benchmark world corresponding to top corresponding to the first kind benchmark world coordinates is sat It is marked with and corresponding to the second following class benchmark world coordinates, wherein, corresponding to the second class benchmark generation of top The difference in height of boundary's coordinate and the first kind benchmark world coordinates is the first kind distance value, corresponding to following the The difference in height of two class benchmark world coordinates and the first kind benchmark world coordinates is the second class distance value；
In the first kind benchmark world coordinates corresponding to first kind distance value, the second class distance value and real projection After it is determined that, it may be determined that the second class benchmark generation corresponding to top corresponding to the first kind benchmark world coordinates Boundary's coordinate and corresponding to the second following class benchmark world coordinates, wherein, for the second class base with top Quasi- world coordinates and it can be uniquely determined out perpendicular to ground corresponding to the second following class benchmark world coordinates The scene warning face in face.
E3：Based on second transformational relation, the corresponding to the second class benchmark world coordinates obtained by calculating Two class benchmark image coordinates；
After the second class benchmark world coordinates is determined, second transformational relation can be based on, calculates the second class The second class benchmark image coordinate corresponding to benchmark world coordinates, that is, determine that frame out guards against face in image three-dimensional The image coordinate of home position under coordinate.Wherein, the second class benchmark image coordinate, which can uniquely determine out, is somebody's turn to do Picture warning face corresponding to scene warning face.Wherein, first kind benchmark image coordinate and first kind benchmark generation Quantity corresponding to boundary's coordinate is identical, the quantity of the second class benchmark world coordinates and the second class benchmark image coordinate Identical, and there is corresponding relation.
F3：Based on the second class benchmark image coordinate, painted in the video monitoring picture that moving target not be present Make the picture warning face corresponding to the scene warning face.After the second class benchmark image coordinate is determined, not The picture drawn in the video monitoring picture of moving target corresponding to the scene warning face be present and guard against face, so far The setting up procedure in picture warning face is completed, sets result can be as shown in 9 (d) (e).
It is emphasized that the picture corresponding to the scene warning face given by above-mentioned A2-F2 and A3-F3 The setting up procedure in face warning face is merely exemplary, should not form the restriction to the embodiment of the present invention.
It is below that the one kind provided the embodiment of the present invention regards exemplified by scene guards against body by scene alert locations Frequency monitoring method is introduced.
It should be noted that the video frequency monitoring method executive agent that the embodiment of the present invention is provided can be video Supervising device, also, the video monitoring apparatus can be run in the equipment being connected with image capture device, The image capture device can have moving object detection function, for example：The image capture device can be with For camera, the equipment that should be connected with image capture device can be data hard disc video recorder or terminal, Wherein, the data hard disc video recorder or terminal can obtain target scene from image capture device in real time Video flowing and export obtained video flowing, wherein, the Component units of video flowing are frame of video composition.When So, the video monitoring apparatus can also run on image capture device, wherein, image capture device can have There is moving object detection function.
As shown in figure 4, the video frequency monitoring method that the embodiment of the present invention is provided, may include steps of：
S401, when detecting moving target during being directed to the video monitoring of target scene, obtain the fortune Target image coordinate of the moving-target under the image three-dimensional coordinate system of place frame of video；
S402, based on the world three dimensional coordinate corresponding to the default image three-dimensional coordinate system and the target scene First transformational relation of system, calculates the first kind world coordinates corresponding to the target image coordinate；
Wherein, the first kind world coordinates is coordinate of the moving target in the world three dimensional coordinate system.
In the present embodiment, S401-S402 is similar with the content of S101-S102 in above-described embodiment, does not do herein Repeat.
S403, obtain the second class world coordinates corresponding to default picture warning body；
Wherein, the second class world coordinates is that the world corresponding with the image coordinate of picture warning body is sat Mark, the second class world coordinates are determined that picture warning body is based on first transformational relation：Exist in advance In set by the video monitoring picture that moving target not be present of the target scene and target scene Picture position corresponding to scene warning body phase, wherein, the second class world coordinates is that the scene guards against body at this World coordinates under world three dimensional coordinate system.
Due to needing to guard against position pass of the body in real target scene by detecting moving target and scene It is to improve the triggering accuracy of alert event, thus, it is possible to obtain the image of default picture warning body is sat The second corresponding class world coordinates of mark.
It is understood that be to determine due to default picture warning body, therefore, default picture warning The second class world coordinates corresponding to the image coordinate of body can precalculate to obtain, with video monitoring process In directly utilize；Certainly, the second class world coordinates corresponding to the image coordinate of default picture warning body can To be calculated during first time video monitoring, can directly be utilized during follow-up video monitoring, This is also rational.Wherein, because the image coordinate of the required picture warning body utilized can uniquely determine Go out picture warning body, therefore, because the quantity of the image coordinate of the required picture warning body utilized Do not limit herein；Also, due to the image of the second class world coordinates and the required picture warning body utilized Coordinate has one-to-one corresponding property, and therefore, the quantity of the second class world coordinates does not limit herein.
Specifically, picture warning body is specially：The image determined based on the second class benchmark image coordinate Position, the second class benchmark image coordinate are：Pixel translation operation institute is performed to first kind benchmark image coordinate The image coordinate obtained, the first kind benchmark image coordinate are in the video monitoring picture for do not exist moving target Image coordinate included by virtual projection；Wherein, the pixel translation operation be using first transformational relation, The scene guards against body and the height relationships of real projection and the translation operation of the second transformational relation；Wherein, should Second transformational relation is the transformational relation with the first transformational relation contrary, and the real projection is guarded against for the scene Projection of the body on the predetermined datum plane in the target scene, the virtual projection be the real projection at this not The position in the video monitoring picture of moving target be present.It is understood that the predetermined datum plane can be with For ground, it is not limited thereto certainly.
In order to which solution layout understands, the setting up procedure for subsequently guarding against the picture corresponding to scene warning face face enters Row is discussed in detail.
S404, judge whether the position relationship change of the first kind world coordinates and the second class world coordinates is full Foot：Include in the range of being formed from the second class world coordinates the first kind world coordinates to do not include this A kind of world coordinates, if it is, performing S405；
Wherein, after first kind world coordinates and the second class world coordinates is obtained, it can be determined that the first kind Whether the position relationship change of world coordinates and the second class world coordinates meets：From the second class world coordinates Include the first kind world coordinates in the range of forming to not including the first kind world coordinates, and according to not Same judged result performs different operations.Specifically, when judge the first kind world coordinates with this second The position relationship change of class world coordinates meets：Including in the range of being formed from the second class world coordinates should First kind world coordinates shows moving target in real target scene to the first kind world coordinates is not included Scene warning body is have left, that is, position relationship change meets predetermined leave and guards against body alarm rule, this When, S405 can be performed；And when the position for judging the first kind world coordinates and the second class world coordinates Relationship change is unsatisfactory for：Include the first kind world coordinates in the range of being formed from the second class world coordinates To the first kind world coordinates is not included, show that moving target does not leave scene police in real target scene Ring face, that is, position relationship change do not meet predetermined leave and guard against body alarm rule, at this point it is possible to not do Processing.
For scene alert locations are scene warning body, in another implementation of the embodiment of the present invention In, whether the position relationship change for judging the first kind world coordinates and the second class world coordinates meets Predetermined alarm rule, it can include：
Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates meets：From The second class world coordinates does not include the first kind world coordinates, to including the first kind in the range of forming World coordinates arrives again does not include the first kind world coordinates, if it is, showing the first kind world coordinates and being somebody's turn to do The position relationship change of second class world coordinates meets predetermined pass through and guards against body alarm rule, can perform S405, otherwise, do not deal with；
Or
Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates meets：From The second class world coordinates does not include the first kind world coordinates to including the first kind in the range of forming World coordinates, if it is, showing that the position relationship of the first kind world coordinates and the second class world coordinates becomes Change, which meets, is destined to go into warning body alarm rule, at this point it is possible to perform S405, otherwise, does not deal with；
Or
Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates meets：Should Second class world coordinates includes the first kind world coordinates in the range of forming, if it is, show this first The change of the position relationship of class world coordinates and the second class world coordinates meets predetermined invasion warning body alarm rule Then, at this point it is possible to perform S405；Otherwise, do not deal with.
It is emphasized that the continuous multiple frame of video of frame of video where moving target, in order to ensure to judge knot The validity of fruit, first kind world coordinates and second can be judged according to acquisition time principle near after first remote Whether the position relationship change of class world coordinates meets to be destined to go into warning body alarm rule, makes a reservation for pass through warning Wire body police regulations then, predetermined leave warning body alarm rule or predetermined invasion warning body alarm rule.S405, touch Alert event of the hairpin for the target scene.
In the present embodiment, S405 is similar with the S105 of above-described embodiment, will not be described here.
Compared with prior art, in this programme, carried out based on the coordinate information under real world three dimensional coordinate Position relationship matches, and the 2D planes being not based under two-dimensional image coordinate system, reduces false alarm, improves The triggering accuracy rate of alert event.
The setting up procedure for guarding against picture corresponding to scene warning body body below carries out citing introduction.Specifically, The scene warning body corresponding to picture warning body setting up procedure be：
A4：Determine that user is set in the video monitoring picture that moving target not be present for be directed to target scene Virtual projection putting, being closed with scene warning body phase；
Wherein, scene warning body can be suspended in the air or rest on the ground.Before picture warning body is drawn, User is needed using ground as predetermined datum plane, is drawn in video monitoring picture and is closed with scene warning body phase Virtual projection, wherein, with scene warning body phase close virtual projection in video monitoring picture with other things The relative position relation of body is equal to：Real projection corresponding to scene warning body is in target scene and corresponding The relative position relation of other objects.Specifically, being warned with scene as given by Figure 10 (a)-Figure 10 (f) The drawing process for the virtual projection that body phase is closed is guarded against, wherein, user can be in the video prison that moving target not be present Control in picture, using ground as predetermined datum plane, mouse is placed in frame of video picture, mouse is changed into ten Word foresight, represent that the starting endpoint of the virtual projection closed with scene warning body phase can be started setting up, by mouse Mark is moved to the position for the starting endpoint for wishing to set the virtual projection closed with scene warning body phase, clicks on mouse Left button, complete the setting of starting endpoint；, there is connecting line in mobile mouse between starting endpoint and mouse pointer (dotted line), the side position for the virtual projection that Dynamic Announce closes with scene warning body phase；Mouse is moved To second end points of the virtual projection closed with scene warning body phase, left mouse button is clicked on, completes second end The drafting of point；The like, the drafting of all end points of virtual projection closed with scene warning body phase is completed, when When last end points of the virtual projection closed with scene warning body phase overlaps with starting endpoint, planar closed is formed Ring, complete the drafting of virtual projection closed with scene warning body phase.
It is emphasized that the video monitoring apparatus can have virtual projection to set function, so as to obtain user The virtual projection closed with scene warning body phase can be set manually on video monitoring picture.It is in addition, predetermined Datum plane selection ground, and the not unfixed object of metope or desktop etc structural form, are ensured with this The validity of set virtual projection.
B4：Based on first transformational relation, the included by the virtual projection closed with scene warning body phase is calculated First kind benchmark world coordinates corresponding to a kind of benchmark image coordinate, wherein, the first kind benchmark image is sat Mark uniquely determines out the virtual projection closed with scene warning body phase；
After the virtual projection for determining to close with scene warning body phase, in order to realize the drafting of picture warning body, The coordinate points on virtual projection that will can be closed with scene warning body phase are transformed under world three dimensional coordinate system, So as to obtain the related coordinate information of the real projection of scene warning body.Specifically, can be based on this first Transformational relation, calculate the first kind benchmark image coordinate institute included by the virtual projection of picture warning body Corresponding first kind benchmark world coordinates, wherein, the first kind benchmark world coordinates can uniquely determine out should Scene guards against the real projection of body.
It should be noted that the quantity of first kind benchmark image coordinate does not limit herein, as long as ensureing to pass through First kind benchmark image coordinate can uniquely determine out the virtual projection closed with scene warning body phase.Separately Outside, it is emphasized that, the virtual projection closed with scene warning body phase is in video monitoring picture relative to it The angle of his object and angle in ground level direction are respectively equivalent to：Scene guards against the real projection of body The angle relative to other corresponding objects and the angle in ground level direction in target scene.Wherein, Scene warning body can be cube or cylinder etc. these it is parallel above and below and except above and below with For other faces it is vertical with ground holding.
C4：Obtain and arrive the first kind distance value of its real projection above scene warning body, and arrive below Second class distance value of its real projection；
Wherein, first kind distance value and the second class distance value can be what user was manually entered.Wherein, this Above scape warning body ground can also be not parallel to following parallel to ground；Specifically, work as scene Guard against above body or below parallel on ground, corresponding distance value is one, that is to say, that this Arrive above scape warning body or a little its real projection all same below；When scene warning body above or under Face is not parallel on ground, and corresponding distance value is multiple, that is to say, that above scene warning body Or to arrive its real projection a little incomplete same below.
D4：Determine that the second class benchmark world above that corresponds to corresponding to the first kind benchmark world coordinates is sat It is marked with and corresponding to the second following class benchmark world coordinates, wherein, corresponding to the second class benchmark generation above Boundary's coordinate is the corresponding first kind distance value with the difference in height of the first kind benchmark world coordinates, under corresponding to The second class benchmark world coordinates in face and the difference in height of the first kind benchmark world coordinates for it is corresponding this second Class distance value；
In the first kind benchmark world coordinates corresponding to first kind distance value, the second class distance value and real projection After it is determined that, determine that the second class benchmark world above that corresponds to corresponding to the first kind benchmark world coordinates is sat Be marked with and corresponding to the second following class benchmark world coordinates, wherein, above corresponding the second class benchmark generation Boundary's coordinate, which combines following the second corresponding class reference coordinate, can uniquely determine out scene warning body.
E4：Based on second transformational relation, the corresponding to the second class benchmark world coordinates obtained by calculating Two class benchmark image coordinates；
After the second class benchmark world coordinates is determined, second transformational relation can be based on, calculate this The second class benchmark image coordinate corresponding to two class benchmark world coordinates, that is, determine that frame out guards against body in image The image coordinate of home position under three-dimensional coordinate.Wherein, the second class benchmark image coordinate can uniquely determine Go out the picture warning face corresponding to scene warning body.Wherein, first kind benchmark image coordinate and first kind base The quantity of quasi- world coordinates is identical, the quantity phase of the second class benchmark world coordinates and the second class benchmark image coordinate With, and there is corresponding relation.
F4：Based on the second class benchmark image coordinate, in it the video monitoring picture of moving target be present in this Draw the picture warning body corresponding to scene warning body.
After the second class benchmark image coordinate is determined, painted in the video monitoring picture that moving target not be present The picture warning face corresponding to scene warning body is made, so far completes the setting up procedure of picture warning body, if Putting result can be as shown in 10 (g) (h).
It is emphasized that the scene given by above-mentioned A4-F4 guards against the picture warning body corresponding to body Setting up procedure is merely exemplary, should not form the restriction to the embodiment of the present invention.
Corresponding to above method embodiment, the embodiments of the invention provide video monitoring apparatus.As shown in figure 5, The device can include：
Target image coordinate determining module 510, for examining being directed to during the video monitoring of target scene When measuring moving target, the moving target is obtained under the image three-dimensional coordinate system on place frame of video Target image coordinate；
First kind world coordinates determining module 520, for based on default described image three-dimensional system of coordinate and institute The first transformational relation of the world three dimensional coordinate system corresponding to target scene is stated, calculates the target image coordinate Corresponding first kind world coordinates, wherein, the first kind world coordinates is the moving target described The coordinate of world three dimensional coordinate system；
Second class world coordinates determining module 530, for obtaining corresponding to default picture alert locations Two class world coordinates, wherein, the second class world coordinates is the image coordinate phase of the picture alert locations Corresponding world coordinates, the second class world coordinates determined based on first transformational relation, the picture Face alert locations are：It is set in the video monitoring picture that moving target not be present of the target scene in advance Picture position put, corresponding with the scene alert locations in the target scene, wherein, described second Class world coordinates is world coordinates of the scene alert locations under the world three dimensional coordinate system；
Position relationship determining module 540, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets predetermined alarm rule, if it is, performing alert event generation module；
The alert event generation module 550, the alert event of the target scene is directed to for triggering.
Compared with prior art, in this programme, carried out based on the coordinate information under real world three dimensional coordinate Position relationship matches, and the 2D planes being not based under two-dimensional image coordinate system, reduces false alarm, improves The triggering accuracy rate of alert event.
Specifically, the picture alert locations are specially：Determined based on the second class benchmark image coordinate Picture position, the second class benchmark image coordinate are：Pixel conversion is performed to first kind benchmark image coordinate Image coordinate obtained by computing, the first kind benchmark image coordinate are the video that moving target be present Image coordinate in monitored picture included by virtual projection；
Wherein, the pixel translation operation be using first transformational relation, the scene alert locations with The translation operation of the height relationships of real projection and the second transformational relation；
Wherein, second transformational relation is the transformational relation with the first transformational relation contrary, described true Projection of the scene alert locations on the predetermined datum plane in the target scene is projected as in fact, it is described Virtual projection is position of the real projection in the video monitoring picture that moving target be present.
Specifically, in a kind of implementation of the present invention, the scene alert locations are：Scene warning line；
The position relationship determining module 540, can include：
First position relation determination unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Do not include in the range of being formed from the second class world coordinates The first kind world coordinates, do not include the first kind world to being arrived again including the first kind world coordinates Coordinate, if it is, showing that the position relationship of the first kind world coordinates and the second class world coordinates becomes Change, which meets, to be made a reservation for pass through warning line alarm rule.
Specifically, in a kind of implementation of the present invention, the scene alert locations are：Scene guards against face；
The position relationship determining module 540, can include：
Second place relation determination unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Do not include in the range of being formed from the second class world coordinates The first kind world coordinates, do not include the first kind world to being arrived again including the first kind world coordinates Coordinate, if it is, showing that the position relationship of the first kind world coordinates and the second class world coordinates becomes Change, which meets, to be made a reservation for pass through warning face alarm rule；
Or
3rd position relationship determining unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Do not include in the range of being formed from the second class world coordinates The first kind world coordinates is to the first kind world coordinates is included, if it is, showing the first kind generation The change of the position relationship of boundary's coordinate and the second class world coordinates meets the warning face of being destined to go into alarm rule；
Or
4th position relationship determining unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Include institute in the range of being formed from the second class world coordinates First kind world coordinates is stated to the first kind world coordinates is not included, if it is, showing the first kind generation The change of the position relationship of boundary's coordinate and the second class world coordinates, which meets, to be made a reservation for leave warning face alarm rule；
Or
5th position relationship determining unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：The second class world coordinates includes described in the range of forming First kind world coordinates, if it is, showing the first kind world coordinates and the second class world coordinates Position relationship change meets predetermined invasion warning face alarm rule.
Specifically, in one implementation, the scene alert locations are：Scene guards against body；
The position relationship determining module 540, can include：
6th position relationship determining unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Do not include in the range of being formed from the second class world coordinates The first kind world coordinates, do not include the first kind world to being arrived again including the first kind world coordinates Coordinate, if it is, showing that the position relationship of the first kind world coordinates and the second class world coordinates becomes Change meets predetermined pass through and guards against body alarm rule；
Or
7th position relationship determining unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Do not include in the range of being formed from the second class world coordinates The first kind world coordinates is to the first kind world coordinates is included, if it is, showing the first kind generation The change of the position relationship of boundary's coordinate and the second class world coordinates, which meets, is destined to go into warning body alarm rule；
Or
8 positions relation determination unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Include institute in the range of being formed from the second class world coordinates First kind world coordinates is stated to the first kind world coordinates is not included, if it is, showing described first The change of the position relationship of class world coordinates and the second class world coordinates meets the predetermined warning body that leaves and alarmed Rule；
Or
9th position relationship determining unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：The second class world coordinates includes described in the range of forming First kind world coordinates, if it is, showing the first kind world coordinates and the second class world coordinates Position relationship change meets predetermined invasion warning body alarm rule.
Specifically, picture warning line is as set by picture warning line setup module corresponding to the scene warning line Put；
The picture warning line setup module can include：
First virtual projection determining unit, for determining that user is being directed to the not presence motion mesh of target scene Set, related to scene warning line virtual projection in target video monitoring picture；
First benchmark world coordinates determining unit, for calculating the virtual projection related to the scene warning line First kind benchmark world coordinates corresponding to included first kind benchmark image coordinate, wherein, described first Class benchmark image coordinate uniquely determines out the virtual projection related to the scene warning line；
First distance value determining unit, for obtaining the scene warning line to the distance value of its real projection；
Second benchmark world coordinates determining unit, for based on the distance value, determining the first kind benchmark The second class benchmark world coordinates corresponding to world coordinates, wherein, the second class benchmark world coordinates and institute The difference in height for stating first kind benchmark world coordinates is the distance value；
First benchmark image coordinate determining unit, for based on second transformational relation, calculating described second The second class benchmark image coordinate corresponding to class benchmark world coordinates；
Picture warning line drawing unit, for based on the second class benchmark image coordinate, not existing described The picture warning line corresponding to the scene warning line is drawn in the video monitoring picture of moving target.
Specifically, the scene warning face is the scene warning face parallel to ground, the scene guards against face institute Corresponding picture warning face is as set by the first picture warning face setup module；
The first picture warning face setup module can include：
Second virtual projection determining unit, for determining that user is being directed to the not presence motion mesh of target scene Set, related to scene warning face virtual projection in target video monitoring picture
3rd benchmark world coordinates determining unit, for based on first transformational relation, calculating and the field The first kind benchmark corresponding to first kind benchmark image coordinate included by the related virtual projection in scape warning face World coordinates, wherein, the first kind benchmark image coordinate uniquely determines out related to scene warning face Virtual projection；
Second distance value determining unit, face is guarded against to the distance value of its real projection for obtaining the scene；
4th benchmark world coordinates determining unit, for based on the distance value, determining the first kind benchmark The second class benchmark world coordinates corresponding to world coordinates, wherein, the second class benchmark world coordinates and institute The difference in height for stating first kind benchmark world coordinates is the distance value；
Second benchmark image coordinate determining unit, for based on second transformational relation, calculating described second The second class benchmark image coordinate corresponding to class benchmark world coordinates；
First picture guards against iso-surface patch unit, for based on the second class benchmark image coordinate, it is described not The picture drawn in the video monitoring picture of moving target corresponding to the scene warning face be present and guard against face.
Specifically, the scene warning face is to guard against face institute perpendicular to the scene warning face on ground, the scene Corresponding picture warning face is as set by the second picture warning face setup module；
The second picture warning face setup module can include：
3rd virtual projection determining unit, for determining that user is being directed to the not presence motion mesh of target scene Set, related to scene warning face virtual projection in target video monitoring picture；
5th benchmark world coordinates determining unit, for based on first transformational relation, calculating and the field The first kind benchmark corresponding to first kind benchmark image coordinate included by the related virtual projection in scape warning face World coordinates, wherein, the first kind benchmark image coordinate uniquely determines out related to scene warning face Virtual projection；
3rd distance value determining unit, for obtaining the top in scene warning face to the of its real projection A kind of distance value, and the second class distance value of its real projection is arrived below；
6th benchmark world coordinates determining unit, for determining corresponding to the first kind benchmark world coordinates The second class benchmark world coordinates corresponding to top and corresponding to the second following class benchmark world coordinates, its In, the difference in height corresponding to the second class benchmark world coordinates and the first kind benchmark world coordinates of top is The first kind distance value, corresponding to the second following class benchmark world coordinates and the first kind benchmark world The difference in height of coordinate is the second class distance value；
3rd benchmark image coordinate determining unit, for obtained by based on second transformational relation, calculating The second class benchmark image coordinate corresponding to second class benchmark world coordinates；
Second picture guards against iso-surface patch unit, for based on the second class benchmark image coordinate, it is described not The picture drawn in the video monitoring picture of moving target corresponding to the scene warning face be present and guard against face.
Specifically, the picture warning body corresponding to the scene warning body is as set by picture warning body setup module Put；
The picture warning body setup module can include：
4th virtual projection determining unit, for determining that user is being directed to the not presence motion mesh of target scene Virtual projection that is set in target frame of video, being closed with scene warning body phase；
7th benchmark world coordinates determining unit, for based on first transformational relation, calculating and the field The first kind benchmark corresponding to the first kind benchmark image coordinate included by virtual projection that scape warning body phase is closed World coordinates, wherein, the first kind benchmark image coordinate uniquely determines out to close with scene warning body phase Virtual projection；
4th distance value determining unit, for obtaining above scene warning body to the of its real projection A kind of distance value, and the second class distance value of its real projection is arrived below；
8th benchmark world coordinates determining unit, for determining corresponding to the first kind benchmark world coordinates Corresponding to the second class benchmark world coordinates above and corresponding to the second following class benchmark world coordinates, its In, the difference in height corresponding to the second class benchmark world coordinates and the first kind benchmark world coordinates above is The corresponding first kind distance value, corresponding to the second following class benchmark world coordinates and the first kind base The difference in height of quasi- world coordinates is the corresponding second class distance value；
4th benchmark image coordinate determining unit, for obtained by based on second transformational relation, calculating The second class benchmark image coordinate corresponding to second class benchmark world coordinates；
Picture guards against volume drawing unit, for based on the second class benchmark image coordinate, not existing described The picture warning body corresponding to the scene warning body is drawn in the video monitoring picture of moving target.
Specifically, conversion of first transformational relation by image three-dimensional coordinate system and camera three-dimensional system of coordinate Formula, and the conversion formula of the camera three-dimensional system of coordinate and world three dimensional coordinate system characterize；
Wherein, the conversion formula of described image three-dimensional system of coordinate and camera three-dimensional system of coordinate is：
The conversion formula of the camera three-dimensional system of coordinate and world three dimensional coordinate system：
Wherein, (x, y, z) be described image three-dimensional system of coordinate under image coordinate, (Xc, Yc, Zc) for institute The camera coordinates under camera three-dimensional system of coordinate are stated, (X, Y, Z) is the generation under the world three dimensional coordinate system Boundary's coordinate, x1, x2, x3, x4 and w are tried to achieve normal for the image capture device based on collection frame of video Numerical value, θ be described image collecting device the angle of pitch, ψ be described image collecting device inclination angle, Hcam For height of the described image collecting device away from ground；The origin of the world three dimensional coordinate system, which is that camera is three-dimensional, to be sat Mark the projection of the origin of system.
It should be noted that herein, such as first and second or the like relational terms be used merely to by One entity or operation make a distinction with another entity or operation, and not necessarily require or imply these Any this actual relation or order be present between entity or operation.Moreover, term " comprising ", " bag Containing " or any other variant thereof is intended to cover non-exclusive inclusion, so that including a series of elements Process, method, article or equipment not only include those key elements, but also including being not expressly set out Other element, or also include for this process, method, article or the intrinsic key element of equipment. In the case of more limiting, the key element that is limited by sentence "including a ...", it is not excluded that including Other identical element also be present in the process of the key element, method, article or equipment.
Each embodiment in this specification is described by the way of related, identical phase between each embodiment As part mutually referring to, what each embodiment stressed be it is different from other embodiment it Place.For system embodiment, because it is substantially similar to embodiment of the method, so description Fairly simple, the relevent part can refer to the partial explaination of embodiments of method.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the protection model of the present invention Enclose.Any modification, equivalent substitution and improvements made within the spirit and principles of the invention etc., are wrapped Containing within the scope of the present invention.

Claims (20)

  1. A kind of 1. video frequency monitoring method, it is characterised in that including：
    When detecting moving target during being directed to the video monitoring of target scene, the motion mesh is obtained Target image coordinate under the image three-dimensional coordinate system of frame of video where being marked on；
    Based on the world three dimensional coordinate corresponding to default described image three-dimensional system of coordinate and the target scene First transformational relation of system, calculates the first kind world coordinates corresponding to the target image coordinate, wherein, The first kind world coordinates is coordinate of the moving target in the world three dimensional coordinate system；
    The second class world coordinates corresponding to default picture alert locations is obtained, wherein, the second class generation Boundary's coordinate is the world coordinates corresponding with the image coordinate of the picture alert locations, the second class world Coordinate is determined that the picture alert locations are based on first transformational relation：In advance in the target field Set and in the target scene scene is warned in the video monitoring picture that moving target not be present of scape The corresponding picture position in position is guarded against, wherein, the second class world coordinates is that the scene alert locations exist World coordinates under the world three dimensional coordinate system；
    Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates accords with Predetermined alarm rule is closed, if it is, triggering is directed to the alert event of the target scene.
  2. 2. according to the method for claim 1, it is characterised in that the picture alert locations are specially： The picture position determined based on the second class benchmark image coordinate, the second class benchmark image coordinate are： Image coordinate obtained by pixel translation operation, the first kind reference map are performed to first kind benchmark image coordinate As coordinate is the image coordinate included by virtual projection in the video monitoring picture that moving target be present；
    Wherein, the pixel translation operation be using first transformational relation, the scene alert locations with The translation operation of the height relationships of real projection and the second transformational relation；
    Wherein, second transformational relation is the transformational relation with the first transformational relation contrary, described true Projection of the scene alert locations on the predetermined datum plane in the target scene is projected as in fact, it is described Virtual projection is position of the real projection in the video monitoring picture that moving target be present.
  3. 3. according to the method for claim 2, it is characterised in that the scene alert locations are：Scene Warning line；
    The position relationship of the judgement first kind world coordinates and the second class world coordinates changes It is no to meet predetermined alarm rule, including：
    Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates is full Foot：Do not include the first kind world coordinates in the range of being formed from the second class world coordinates, to bag Including the first kind world coordinates and arriving again does not include the first kind world coordinates, if it is, showing described The change of the position relationship of a kind of world coordinates and the second class world coordinates, which meets, to be made a reservation for pass through warning report from a liner Police regulations are then.
  4. 4. according to the method for claim 2, it is characterised in that the scene alert locations are：Scene Warning face；
    The position relationship of the judgement first kind world coordinates and the second class world coordinates changes It is no to meet predetermined alarm rule, including：
    Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates is full Foot：Do not include the first kind world coordinates in the range of being formed from the second class world coordinates, to bag Including the first kind world coordinates and arriving again does not include the first kind world coordinates, if it is, showing described The change of the position relationship of a kind of world coordinates and the second class world coordinates meets predetermined warning face of passing through and reported Police regulations are then；
    Or
    Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates is full Foot：From the second class world coordinates form in the range of include the first kind world coordinates to including The first kind world coordinates, if it is, showing that the first kind world coordinates and the second class world are sat The change of target position relationship meets the warning face of being destined to go into alarm rule；
    Or
    Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates is full Foot：Include the first kind world coordinates in the range of being formed from the second class world coordinates to not including The first kind world coordinates, if it is, showing that the first kind world coordinates and the second class world are sat The change of target position relationship, which meets, to be made a reservation for leave warning face alarm rule；
    Or
    Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates is full Foot：The second class world coordinates includes the first kind world coordinates in the range of forming, if it is, Show that the position relationship change of the first kind world coordinates and the second class world coordinates meets pre- be incorporated into The warning face of invading alarm rule.
  5. 5. according to the method for claim 2, it is characterised in that the scene alert locations are：Scene Guard against body；
    The position relationship of the judgement first kind world coordinates and the second class world coordinates changes It is no to meet predetermined alarm rule, including：
    Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates is full Foot：Do not include the first kind world coordinates in the range of being formed from the second class world coordinates, to bag Including the first kind world coordinates and arriving again does not include the first kind world coordinates, if it is, showing described The change of the position relationship of a kind of world coordinates and the second class world coordinates meets predetermined pass through and guards against body report Police regulations are then；
    Or
    Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates is full Foot：From the second class world coordinates form in the range of include the first kind world coordinates to including The first kind world coordinates, if it is, showing that the first kind world coordinates and the second class world are sat The change of target position relationship, which meets, is destined to go into warning body alarm rule；
    Or
    Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates is full Foot：Include the first kind world coordinates in the range of being formed from the second class world coordinates to not including The first kind world coordinates, if it is, showing the first kind world coordinates and the second class generation The position relationship change of boundary's coordinate meets predetermined leave and guards against body alarm rule；
    Or
    Judge whether the position relationship change of the first kind world coordinates and the second class world coordinates is full Foot：The second class world coordinates includes the first kind world coordinates in the range of forming, if it is, Show that the position relationship change of the first kind world coordinates and the second class world coordinates meets pre- be incorporated into Invade warning body alarm rule.
  6. 6. according to the method for claim 3, it is characterised in that picture corresponding to the scene warning line The setting up procedure of warning line is：
    Determine that user is set in the video monitoring picture that moving target not be present for be directed to target scene , related to scene warning line virtual projection；
    Based on first transformational relation, calculate included by the virtual projection related to the scene warning line First kind benchmark world coordinates corresponding to first kind benchmark image coordinate, wherein, the first kind reference map As coordinate uniquely determines out the virtual projection related to the scene warning line；
    The scene warning line is obtained to the distance value of its real projection；
    Based on the distance value, the second class benchmark world corresponding to the first kind benchmark world coordinates is determined Coordinate, wherein, the difference in height of the second class benchmark world coordinates and the first kind benchmark world coordinates is The distance value；
    Based on second transformational relation, the second class base corresponding to the second class benchmark world coordinates is calculated Quasi- image coordinate；
    Based on the second class benchmark image coordinate, in the video monitoring picture that moving target be present Draw the picture warning line corresponding to the scene warning line.
  7. 7. according to the method for claim 4, it is characterised in that the scene warning face is parallel to ground The scene warning face in face, the setting up procedure in the picture warning face corresponding to the scene warning face are：
    Determine that user is set in the video monitoring picture that moving target not be present for be directed to target scene , to the related virtual projection in scene warning face；
    Based on first transformational relation, calculate included by the virtual projection related to scene warning face First kind benchmark world coordinates corresponding to first kind benchmark image coordinate, wherein, the first kind reference map As coordinate uniquely determines out the virtual projection related to scene warning face；
    Obtain the scene and guard against face to the distance value of its real projection；
    Based on the distance value, the second class benchmark world corresponding to the first kind benchmark world coordinates is determined Coordinate, wherein, the difference in height of the second class benchmark world coordinates and the first kind benchmark world coordinates is The distance value；
    Based on second transformational relation, the second class base corresponding to the second class benchmark world coordinates is calculated Quasi- image coordinate；
    Based on the second class benchmark image coordinate, in the video monitoring picture that moving target be present Draw the picture warning face corresponding to the scene warning face.
  8. 8. according to the method for claim 4, it is characterised in that the scene warning face is perpendicular to ground The scene warning face in face, the setting up procedure in the picture warning face corresponding to the scene warning face are：
    Determine that user is set in the video monitoring picture that moving target not be present for be directed to target scene , to the related virtual projection in scene warning face；
    Based on first transformational relation, calculate included by the virtual projection related to scene warning face First kind benchmark world coordinates corresponding to first kind benchmark image coordinate, wherein, the first kind reference map As coordinate uniquely determines out the virtual projection related to scene warning face；
    The top in the scene warning face is obtained to the first kind distance value of its real projection, and arrives it below Second class distance value of real projection；
    Determine that the second class benchmark world corresponding to top corresponding to the first kind benchmark world coordinates is sat It is marked with and corresponding to the second following class benchmark world coordinates, wherein, corresponding to the second class benchmark generation of top The difference in height of boundary's coordinate and the first kind benchmark world coordinates is the first kind distance value, corresponding to following The second class benchmark world coordinates and the first kind benchmark world coordinates difference in height for second class away from From value；
    Based on second transformational relation, second corresponding to the second class benchmark world coordinates obtained by calculating Class benchmark image coordinate；
    Based on the second class benchmark image coordinate, in the video monitoring picture that moving target be present Draw the picture warning face corresponding to the scene warning face.
  9. 9. according to the method for claim 5, it is characterised in that the picture corresponding to the scene warning body Face warning body setting up procedure be：
    Set by determining user in the frame of video that moving target not be present for be directed to target scene and institute State the virtual projection that scene warning body phase is closed；
    Based on first transformational relation, calculate included by the virtual projection closed with scene warning body phase First kind benchmark world coordinates corresponding to first kind benchmark image coordinate, wherein, the first kind reference map As coordinate uniquely determines out the virtual projection closed with scene warning body phase；
    Obtain and arrive the first kind distance value of its real projection above the scene warning body, and arrive it below Second class distance value of real projection；
    Determine that the second class benchmark world above that corresponds to corresponding to the first kind benchmark world coordinates is sat It is marked with and corresponding to the second following class benchmark world coordinates, wherein, corresponding to the second class benchmark generation above Boundary's coordinate is the corresponding first kind distance value with the difference in height of the first kind benchmark world coordinates, corresponding In the difference in height of the second following class benchmark world coordinates and the first kind benchmark world coordinates be corresponding The second class distance value；
    Based on second transformational relation, second corresponding to the second class benchmark world coordinates obtained by calculating Class benchmark image coordinate；
    Based on the second class benchmark image coordinate, in the video monitoring picture that moving target be present Draw the picture warning body corresponding to the scene warning body.
  10. 10. according to the method described in claim any one of 1-9, it is characterised in that first conversion is closed System passes through image three-dimensional coordinate system and the conversion formula of camera three-dimensional system of coordinate, and the camera three-dimensional coordinate It is to characterize with the conversion formula of world three dimensional coordinate system；
    Wherein, the conversion formula of described image three-dimensional system of coordinate and camera three-dimensional system of coordinate is：
    <mrow> <mfenced open = "(" close = ")"> <mtable> <mtr> <mtd> <mi>x</mi> </mtd> </mtr> <mtr> <mtd> <mi>y</mi> </mtd> </mtr> <mtr> <mtd> <mi>z</mi> </mtd> </mtr> <mtr> <mtd> <mn>1</mn> </mtd> </mtr> </mtable> </mfenced> <mo>=</mo> <msup> <mfenced open = "(" close = ")"> <mtable> <mtr> <mtd> <mn>1</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mrow> <mi>x</mi> <mn>1</mn> </mrow> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mrow> <mo>-</mo> <mi>x</mi> <mn>2</mn> </mrow> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mrow> <mi>x</mi> <mn>3</mn> </mrow> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mrow> <mi>x</mi> <mn>4</mn> </mrow> </mtd> <mtd> <mn>0</mn> </mtd> </mtr> </mtable> </mfenced> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <msup> <mi>w</mi> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <mfenced open = "(" close = ")"> <mtable> <mtr> <mtd> <msub> <mi>X</mi> <mi>c</mi> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>Y</mi> <mi>c</mi> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>Z</mi> <mi>c</mi> </msub> </mtd> </mtr> <mtr> <mtd> <mn>1</mn> </mtd> </mtr> </mtable> </mfenced> </mrow>
    The conversion formula of the camera three-dimensional system of coordinate and world three dimensional coordinate system：
    Wherein, (x, y, z) be described image three-dimensional system of coordinate under image coordinate, (Xc, Yc, Zc) for institute The camera coordinates under camera three-dimensional system of coordinate are stated, (X, Y, Z) is the generation under the world three dimensional coordinate system Boundary's coordinate, the constant that x1, x2, x3, x4 and w are tried to achieve for the image capture device based on collection frame of video Value, θ be described image collecting device the angle of pitch, ψ be described image collecting device inclination angle, Hcam For height of the described image collecting device away from ground；The origin of the world three dimensional coordinate system, which is that camera is three-dimensional, to be sat Mark the projection of the origin of system.
  11. A kind of 11. video monitoring apparatus, it is characterised in that including：
    Target image coordinate determining module, for detecting being directed to during the video monitoring of target scene During moving target, target of the moving target under the image three-dimensional coordinate system on place frame of video is obtained Image coordinate；
    First kind world coordinates determining module, for based on default described image three-dimensional system of coordinate and the mesh The first transformational relation of the world three dimensional coordinate system corresponding to scene is marked, it is right to calculate the target image coordinate institute The first kind world coordinates answered, wherein, the first kind world coordinates is the moving target in the world The coordinate of three-dimensional system of coordinate；
    Second class world coordinates determining module, for obtaining the second class corresponding to default picture alert locations World coordinates, wherein, the second class world coordinates is corresponding for the image coordinate of the picture alert locations World coordinates, the second class world coordinates determined based on first transformational relation, the picture police Guarding against position is：It is set in the video monitoring picture that moving target not be present of the target scene in advance , corresponding with the scene alert locations in target scene picture position, wherein, second class World coordinates is world coordinates of the scene alert locations under the world three dimensional coordinate system；
    Position relationship determining module, for judging the first kind world coordinates and the second class world coordinates Position relationship change whether meet predetermined alarm rule, if it is, perform alert event generation module；
    The alert event generation module, the alert event of the target scene is directed to for triggering.
  12. 12. device according to claim 11, it is characterised in that the picture alert locations are specially： The picture position determined based on the second class benchmark image coordinate, the second class benchmark image coordinate are： Image coordinate obtained by pixel translation operation, the first kind reference map are performed to first kind benchmark image coordinate As coordinate is the image coordinate included by virtual projection in the video monitoring picture that moving target be present；
    Wherein, the pixel translation operation be using first transformational relation, the scene alert locations with The translation operation of the height relationships of real projection and the second transformational relation；
    Wherein, second transformational relation is the transformational relation with the first transformational relation contrary, described true Projection of the scene alert locations on the predetermined datum plane in the target scene is projected as in fact, it is described Virtual projection is position of the real projection in the video monitoring picture that moving target be present.
  13. 13. device according to claim 12, it is characterised in that the scene alert locations are： Scape warning line；
    The position relationship determining module, including：
    First position relation determination unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Do not include in the range of being formed from the second class world coordinates The first kind world coordinates, do not include the first kind world to being arrived again including the first kind world coordinates Coordinate, if it is, showing that the position relationship of the first kind world coordinates and the second class world coordinates becomes Change, which meets, to be made a reservation for pass through warning line alarm rule.
  14. 14. device according to claim 12, it is characterised in that the scene alert locations are： Scape guards against face；
    The position relationship determining module, including：
    Second place relation determination unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Do not include in the range of being formed from the second class world coordinates The first kind world coordinates, do not include the first kind world to being arrived again including the first kind world coordinates Coordinate, if it is, showing that the position relationship of the first kind world coordinates and the second class world coordinates becomes Change, which meets, to be made a reservation for pass through warning face alarm rule；
    Or
    3rd position relationship determining unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Do not include in the range of being formed from the second class world coordinates The first kind world coordinates is to the first kind world coordinates is included, if it is, showing the first kind generation The change of the position relationship of boundary's coordinate and the second class world coordinates meets the warning face of being destined to go into alarm rule；
    Or
    4th position relationship determining unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Include institute in the range of being formed from the second class world coordinates First kind world coordinates is stated to the first kind world coordinates is not included, if it is, showing the first kind generation The change of the position relationship of boundary's coordinate and the second class world coordinates, which meets, to be made a reservation for leave warning face alarm rule；
    Or
    5th position relationship determining unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：The second class world coordinates includes described in the range of forming First kind world coordinates, if it is, showing the first kind world coordinates and the second class world coordinates Position relationship change meets predetermined invasion warning face alarm rule.
  15. 15. device according to claim 12, it is characterised in that the scene alert locations are： Scape guards against body；
    The position relationship determining module, including：
    6th position relationship determining unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Do not include in the range of being formed from the second class world coordinates The first kind world coordinates, do not include the first kind world to being arrived again including the first kind world coordinates Coordinate, if it is, showing that the position relationship of the first kind world coordinates and the second class world coordinates becomes Change meets predetermined pass through and guards against body alarm rule；
    Or
    7th position relationship determining unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Do not include in the range of being formed from the second class world coordinates The first kind world coordinates is to the first kind world coordinates is included, if it is, showing the first kind generation The change of the position relationship of boundary's coordinate and the second class world coordinates, which meets, is destined to go into warning body alarm rule；
    Or
    8 positions relation determination unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：Include institute in the range of being formed from the second class world coordinates First kind world coordinates is stated to the first kind world coordinates is not included, if it is, showing described first The change of the position relationship of class world coordinates and the second class world coordinates meets the predetermined warning body that leaves and alarmed Rule；
    Or
    9th position relationship determining unit, for judging the first kind world coordinates and the second class world Whether the position relationship change of coordinate meets：The second class world coordinates includes described in the range of forming First kind world coordinates, if it is, showing the first kind world coordinates and the second class world coordinates Position relationship change meets predetermined invasion warning body alarm rule.
  16. 16. device according to claim 13, it is characterised in that the scene warning line is corresponding to be drawn Face warning line is as set by picture warning line setup module；
    The picture warning line setup module includes：
    First virtual projection determining unit, for determining that user is being directed to the not presence motion mesh of target scene Set, related to scene warning line virtual projection in target video monitoring picture；
    First benchmark world coordinates determining unit, for calculating the virtual projection related to the scene warning line First kind benchmark world coordinates corresponding to included first kind benchmark image coordinate, wherein, described first Class benchmark image coordinate uniquely determines out the virtual projection related to the scene warning line；
    First distance value determining unit, for obtaining the scene warning line to the distance value of its real projection；
    Second benchmark world coordinates determining unit, for based on the distance value, determining the first kind benchmark The second class benchmark world coordinates corresponding to world coordinates, wherein, the second class benchmark world coordinates and institute The difference in height for stating first kind benchmark world coordinates is the distance value；
    First benchmark image coordinate determining unit, for based on second transformational relation, calculating described second The second class benchmark image coordinate corresponding to class benchmark world coordinates；
    Picture warning line drawing unit, for based on the second class benchmark image coordinate, not existing described The picture warning line corresponding to the scene warning line is drawn in the video monitoring picture of moving target.
  17. 17. device according to claim 14, it is characterised in that scene warning face be parallel to The scene warning face on ground, the picture warning face corresponding to the scene warning face are set by the first picture warning face Put set by module；
    The first picture warning face setup module includes：
    Second virtual projection determining unit, for determining that user is being directed to the not presence motion mesh of target scene Set, related to scene warning face virtual projection in target video monitoring picture
    3rd benchmark world coordinates determining unit, for based on first transformational relation, calculating and the field The first kind benchmark corresponding to first kind benchmark image coordinate included by the related virtual projection in scape warning face World coordinates, wherein, the first kind benchmark image coordinate uniquely determines out related to scene warning face Virtual projection；
    Second distance value determining unit, face is guarded against to the distance value of its real projection for obtaining the scene；
    4th benchmark world coordinates determining unit, for based on the distance value, determining the first kind benchmark The second class benchmark world coordinates corresponding to world coordinates, wherein, the second class benchmark world coordinates and institute The difference in height for stating first kind benchmark world coordinates is the distance value；
    Second benchmark image coordinate determining unit, for based on second transformational relation, calculating described second The second class benchmark image coordinate corresponding to class benchmark world coordinates；
    First picture guards against iso-surface patch unit, for based on the second class benchmark image coordinate, it is described not The picture drawn in the video monitoring picture of moving target corresponding to the scene warning face be present and guard against face.
  18. 18. device according to claim 14, it is characterised in that scene warning face be perpendicular to The scene warning face on ground, the picture warning face corresponding to the scene warning face are set by the second picture warning face Put set by module；
    The second picture warning face setup module includes：
    3rd virtual projection determining unit, for determining that user is being directed to the not presence motion mesh of target scene Set, related to scene warning face virtual projection in target video monitoring picture；
    5th benchmark world coordinates determining unit, for based on first transformational relation, calculating and the field The first kind benchmark corresponding to first kind benchmark image coordinate included by the related virtual projection in scape warning face World coordinates, wherein, the first kind benchmark image coordinate uniquely determines out related to scene warning face Virtual projection；
    3rd distance value determining unit, for obtaining the top in scene warning face to the of its real projection A kind of distance value, and the second class distance value of its real projection is arrived below；
    6th benchmark world coordinates determining unit, for determining corresponding to the first kind benchmark world coordinates The second class benchmark world coordinates corresponding to top and corresponding to the second following class benchmark world coordinates, its In, the difference in height corresponding to the second class benchmark world coordinates and the first kind benchmark world coordinates of top is The first kind distance value, corresponding to the second following class benchmark world coordinates and the first kind benchmark world The difference in height of coordinate is the second class distance value；
    3rd benchmark image coordinate determining unit, for obtained by based on second transformational relation, calculating The second class benchmark image coordinate corresponding to second class benchmark world coordinates；
    Second picture guards against iso-surface patch unit, for based on the second class benchmark image coordinate, it is described not The picture drawn in the video monitoring picture of moving target corresponding to the scene warning face be present and guard against face.
  19. 19. device according to claim 15, it is characterised in that corresponding to the scene warning body Picture guards against body as set by picture warning body setup module；
    The picture warning body setup module includes：
    4th virtual projection determining unit, for determining that user is being directed to the not presence motion mesh of target scene Virtual projection that is set in target frame of video, being closed with scene warning body phase；
    7th benchmark world coordinates determining unit, for based on first transformational relation, calculating and the field The first kind benchmark corresponding to the first kind benchmark image coordinate included by virtual projection that scape warning body phase is closed World coordinates, wherein, the first kind benchmark image coordinate uniquely determines out to close with scene warning body phase Virtual projection；
    4th distance value determining unit, for obtaining above scene warning body to the of its real projection A kind of distance value, and the second class distance value of its real projection is arrived below；
    8th benchmark world coordinates determining unit, for determining corresponding to the first kind benchmark world coordinates Corresponding to the second class benchmark world coordinates above and corresponding to the second following class benchmark world coordinates, its In, the difference in height corresponding to the second class benchmark world coordinates and the first kind benchmark world coordinates above is The corresponding first kind distance value, corresponding to the second following class benchmark world coordinates and the first kind base The difference in height of quasi- world coordinates is the corresponding second class distance value；
    4th benchmark image coordinate determining unit, for obtained by based on second transformational relation, calculating The second class benchmark image coordinate corresponding to second class benchmark world coordinates；
    Picture guards against volume drawing unit, for based on the second class benchmark image coordinate, not existing described The picture warning body corresponding to the scene warning body is drawn in the video monitoring picture of moving target.
  20. 20. according to the device described in claim any one of 11-19, it is characterised in that first conversion is closed System passes through image three-dimensional coordinate system and the conversion formula of camera three-dimensional system of coordinate, and the camera three-dimensional coordinate It is to characterize with the conversion formula of world three dimensional coordinate system；
    Wherein, the conversion formula of described image three-dimensional system of coordinate and camera three-dimensional system of coordinate is：
    <mrow> <mfenced open = "(" close = ")"> <mtable> <mtr> <mtd> <mi>x</mi> </mtd> </mtr> <mtr> <mtd> <mi>y</mi> </mtd> </mtr> <mtr> <mtd> <mi>z</mi> </mtd> </mtr> <mtr> <mtd> <mn>1</mn> </mtd> </mtr> </mtable> </mfenced> <mo>=</mo> <msup> <mfenced open = "(" close = ")"> <mtable> <mtr> <mtd> <mn>1</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mrow> <mi>x</mi> <mn>1</mn> </mrow> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mrow> <mo>-</mo> <mi>x</mi> <mn>2</mn> </mrow> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mrow> <mi>x</mi> <mn>3</mn> </mrow> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mrow> <mi>x</mi> <mn>4</mn> </mrow> </mtd> <mtd> <mn>0</mn> </mtd> </mtr> </mtable> </mfenced> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <msup> <mi>w</mi> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <mfenced open = "(" close = ")"> <mtable> <mtr> <mtd> <msub> <mi>X</mi> <mi>c</mi> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>Y</mi> <mi>c</mi> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>Z</mi> <mi>c</mi> </msub> </mtd> </mtr> <mtr> <mtd> <mn>1</mn> </mtd> </mtr> </mtable> </mfenced> </mrow>
    The conversion formula of the camera three-dimensional system of coordinate and world three dimensional coordinate system：
    Wherein, (x, y, z) be described image three-dimensional system of coordinate under image coordinate, (Xc, Yc, Zc) for institute The camera coordinates under camera three-dimensional system of coordinate are stated, (X, Y, Z) is the generation under the world three dimensional coordinate system Boundary's coordinate, the constant that x1, x2, x3, x4 and w are tried to achieve for the image capture device based on collection frame of video Value, θ be described image collecting device the angle of pitch, ψ be described image collecting device inclination angle, Hcam For height of the described image collecting device away from ground；The origin of the world three dimensional coordinate system, which is that camera is three-dimensional, to be sat Mark the projection of the origin of system.
CN201610322001.7A 2016-05-16 2016-05-16 Video monitoring method and device Active CN107396037B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610322001.7A CN107396037B (en) 2016-05-16 2016-05-16 Video monitoring method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610322001.7A CN107396037B (en) 2016-05-16 2016-05-16 Video monitoring method and device

Publications (2)

Publication Number Publication Date
CN107396037A true CN107396037A (en) 2017-11-24
CN107396037B CN107396037B (en) 2020-04-03

Family

ID=60338476

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610322001.7A Active CN107396037B (en) 2016-05-16 2016-05-16 Video monitoring method and device

Country Status (1)

Country Link
CN (1) CN107396037B (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108566534A (en) * 2018-04-23 2018-09-21 Oppo广东移动通信有限公司 Alarm method, device, terminal based on video monitoring and storage medium
CN111508027A (en) * 2019-01-31 2020-08-07 杭州海康威视数字技术股份有限公司 Method and device for calibrating external parameters of camera
CN111538009A (en) * 2019-01-21 2020-08-14 杭州海康威视数字技术股份有限公司 Radar point marking method and device
CN113068000A (en) * 2019-12-16 2021-07-02 杭州海康威视数字技术股份有限公司 Method, device, equipment and system for monitoring video target and storage medium
CN113106839A (en) * 2021-03-29 2021-07-13 杭州海康威视数字技术股份有限公司 Control method, device, equipment and system of lifting bridge
CN113452954A (en) * 2020-03-26 2021-09-28 浙江宇视科技有限公司 Behavior analysis method, apparatus, device, and medium
CN113476835A (en) * 2020-10-22 2021-10-08 青岛海信电子产业控股股份有限公司 Picture display method and device

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2209091B1 (en) * 2009-01-16 2012-08-08 Honda Research Institute Europe GmbH System and method for object motion detection based on multiple 3D warping and vehicle equipped with such system
CN102650514A (en) * 2012-05-03 2012-08-29 秦毅 Stereoscopic vision system and application thereof to real time monitoring of three-dimensional safety warning area
CN103745484A (en) * 2013-12-31 2014-04-23 国家电网公司 Worker target safety early-warning method for hot-line work on electric power facility
CN104680557A (en) * 2015-03-10 2015-06-03 重庆邮电大学 Intelligent detection method for abnormal behavior in video sequence image
US20150242691A1 (en) * 2012-09-07 2015-08-27 Siemens Schweiz AG a corporation Methods and apparatus for establishing exit/entry criteria for a secure location
CN104954747A (en) * 2015-06-17 2015-09-30 浙江大华技术股份有限公司 Video monitoring method and device
CN105141885A (en) * 2014-05-26 2015-12-09 杭州海康威视数字技术股份有限公司 Method for video monitoring and device

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2209091B1 (en) * 2009-01-16 2012-08-08 Honda Research Institute Europe GmbH System and method for object motion detection based on multiple 3D warping and vehicle equipped with such system
CN102650514A (en) * 2012-05-03 2012-08-29 秦毅 Stereoscopic vision system and application thereof to real time monitoring of three-dimensional safety warning area
US20150242691A1 (en) * 2012-09-07 2015-08-27 Siemens Schweiz AG a corporation Methods and apparatus for establishing exit/entry criteria for a secure location
CN103745484A (en) * 2013-12-31 2014-04-23 国家电网公司 Worker target safety early-warning method for hot-line work on electric power facility
CN105141885A (en) * 2014-05-26 2015-12-09 杭州海康威视数字技术股份有限公司 Method for video monitoring and device
CN104680557A (en) * 2015-03-10 2015-06-03 重庆邮电大学 Intelligent detection method for abnormal behavior in video sequence image
CN104954747A (en) * 2015-06-17 2015-09-30 浙江大华技术股份有限公司 Video monitoring method and device

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108566534A (en) * 2018-04-23 2018-09-21 Oppo广东移动通信有限公司 Alarm method, device, terminal based on video monitoring and storage medium
CN111538009A (en) * 2019-01-21 2020-08-14 杭州海康威视数字技术股份有限公司 Radar point marking method and device
CN111508027A (en) * 2019-01-31 2020-08-07 杭州海康威视数字技术股份有限公司 Method and device for calibrating external parameters of camera
CN111508027B (en) * 2019-01-31 2023-10-20 杭州海康威视数字技术股份有限公司 Method and device for calibrating external parameters of camera
CN113068000A (en) * 2019-12-16 2021-07-02 杭州海康威视数字技术股份有限公司 Method, device, equipment and system for monitoring video target and storage medium
CN113452954A (en) * 2020-03-26 2021-09-28 浙江宇视科技有限公司 Behavior analysis method, apparatus, device, and medium
CN113452954B (en) * 2020-03-26 2023-02-28 浙江宇视科技有限公司 Behavior analysis method, apparatus, device and medium
CN113476835A (en) * 2020-10-22 2021-10-08 青岛海信电子产业控股股份有限公司 Picture display method and device
CN113106839A (en) * 2021-03-29 2021-07-13 杭州海康威视数字技术股份有限公司 Control method, device, equipment and system of lifting bridge
CN113106839B (en) * 2021-03-29 2023-06-06 杭州海康威视数字技术股份有限公司 Control method, device, equipment and system of lifting bridge

Also Published As

Publication number Publication date
CN107396037B (en) 2020-04-03

Similar Documents

Publication Publication Date Title
CN107396037A (en) Video frequency monitoring method and device
CN104680555B (en) Cross the border detection method and out-of-range monitoring system based on video monitoring
CN104867174B (en) A kind of three-dimensional map rendering indication method and system
CN104618688B (en) A kind of visual control means of defence
US20170103581A1 (en) Visualization of physical interactions in augmented reality
CN103530881B (en) Be applicable to the Outdoor Augmented Reality no marks point Tracing Registration method of mobile terminal
CN102496232B (en) Transmission facility monitoring method and system
CN106463032A (en) Intrusion detection with directional sensing
CN110148196A (en) A kind of image processing method, device and relevant device
CN104349136B (en) Method for Detecting Image by Three-dimensional Detection Device
CN106295581A (en) Obstacle detection method, device and virtual reality device
CN104902246A (en) Video monitoring method and device
CN112798811B (en) Speed measurement method, device and equipment
CN104935893A (en) Monitoring method and device
CN104007240A (en) Fusion positioning technology based on binocular recognition and electronic nose network gas detection
CN106327461B (en) A kind of image processing method and device for monitoring
JP6673641B2 (en) Ground collapse detection system
CN104954747A (en) Video monitoring method and device
CN104966062A (en) Video monitoring method and device
CN108957504A (en) The method and system of indoor and outdoor consecutive tracking
CN106264537B (en) System and method for measuring human body posture height in image
CN110067274A (en) Apparatus control method and excavator
CN109740659A (en) A kind of image matching method and device, electronic equipment, storage medium
CN108830933A (en) A kind of electric force pole tower tower body method for reconstructing, system, medium and equipment
CN108510528A (en) A kind of method and device of visible light and infrared image registration fusion

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant