CN104539909A - Video monitoring method and video monitoring server - Google Patents
Video monitoring method and video monitoring server Download PDFInfo
- Publication number
- CN104539909A CN104539909A CN201510021580.7A CN201510021580A CN104539909A CN 104539909 A CN104539909 A CN 104539909A CN 201510021580 A CN201510021580 A CN 201510021580A CN 104539909 A CN104539909 A CN 104539909A
- Authority
- CN
- China
- Prior art keywords
- target
- video monitoring
- monitoring server
- watch
- dog
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Closed-Circuit Television Systems (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention provides a video monitoring method and a video monitoring server. The video monitoring method and the video monitoring server are used for achieving target tracking among multiple monitoring devices, and the matching accuracy during target tracking is improved. The method in the embodiment includes the steps that when a first target leaves the scope of a first monitoring device, the video monitoring server obtains a preset motion track of the first target according to first target data transmitted by the first monitoring device; the video monitoring server obtains a smooth motion track of a second target according to received second target data transmitted by the second monitoring device; the video monitoring server judges whether any one of the smooth motion track of the second target and the preset motion track of the first target is matched or not; if any one of the smooth motion track of the second target and the preset motion track of the first target is matched, the video monitoring server determines the second target as the first target. The embodiment of the invention further provides the video monitoring server.
Description
Technical field
The present invention relates to technical field of video monitoring, particularly relate to a kind of video frequency monitoring method and video monitoring server.
Background technology
Along with the fast development of video technique, multiple-camera monitoring have also been obtained applies widely, because the region of monitoring constantly increases, consider economic factor and avoid unnecessary waste, there is provided saturated video camera obviously unlikely to guarded region, therefore the multiple-camera monitoring technique of Non-overlapping Domain have also been obtained increasing utilization.Video monitoring system is primarily of two part compositions: the tracking 1, in camera, and the current comparative maturity of technology of this respect, follows the tracks of moving target in simple terms exactly in the single camera ken, 2, tracking between camera, namely between multiple camera field, carry out the transmission of tracking target, tracking between camera can be divided into again the camera between the tracking of the camera between the overlapping ken and the non-overlapped ken to follow the tracks of, the research followed the tracks of in current camera mainly pays close attention to background modeling, prospect is monitored, shadow removal, block in the problems such as process, and achieve good achievement in research, these key issues are also the Research foundations followed the tracks of between camera, tracking technique between camera then difficulty is larger, different video cameras is due to specification difference, the difference of imaging angle and differing of image quality, same moving target there are differences at the external appearance characteristic of different cameras, not there is continuity, and, for the non-overlapped ken camera between tracking, ken blind area is there is between multiple video camera, the motion of moving target in these regions is not camera supervised, the situation of its motion is unknowable, considerably increase the difficulty of tracking, tracking problem between non-overlapped multiple-camera is different from the tracking problem between single camera and overlapping ken multiple-camera, in the non-overlapped ken, all discrete in the air during the appearing at of target, this makes tracking become very difficult.
In current non-overlapped ken multiple-camera monitor network, the research of Target Tracking Problem mainly concentrates on the coupling of target in different cameras, what existing matching process adopted mostly is CF feature, Motamed and Wallart achieves by the normalization color histogram of target and 3D dimension information the coupling that highway gets on the car jointly, need to demarcate video camera in advance, the people such as Arth propose the vehicle " fingerprint " of a kind of Based PC A-SIFT and words tree, meet the transfer of data of low bandwidth monitor network, a large amount of vehicle image data are needed to make reference when setting up " fingerprint ", the matched jamming of the pedestrian in Cheng and Madden Deng Ren research department in monitoring environment, they establish a kind of domain color spectrum histogram table representation model MCSHR to build presentation model, pedestrian is in the process of motion, subconsciousness can be subject to the impact of surrounding environment, the macroscopic view of these impacts is presented as the effect of some power, such as, in the process of walking, we automaticly can evade pedestrian, wall, barrier etc., these are presented as repulsive force to us, in the process of walking, we automaticly can be subject to the attraction of some thing equally, such as outlet, rest chair, shop etc., and these are presented as attraction to us, attraction and repulsive force are referred to as social force, Helbing and Molnar proposes social force model in nineteen ninety-five.
Existing a kind of video frequency monitoring method is: from the target data of camera surveillance, extract target external appearance characteristic and space-time characteristic, and the topological relation fully utilized between video camera sets up the space-time restriction that target shifts between video camera, suitable association algorithm is finally utilized to carry out characteristic matching to the target in different video cameras, obtain the corresponding relation of target between different cameras, and then complete the tracking of target between camera.
But the method is based on external appearance characteristic and space-time characteristic, and the quality of video camera and extraneous factor directly have influence on the extraction of the external appearance characteristic of target, thus make the matching accuracy rate of video camera to target low.
Summary of the invention
Embodiments provide a kind of video frequency monitoring method and video monitoring server, for judging the corresponding relation of first object and the second target according to the predicted motion track of first object and the smooth motion trajectories of the second target, the tracking of realize target between many watch-dogs, the accuracy rate of mating when improve target following.
In view of this, first aspect present invention improves a kind of video frequency monitoring method, comprising:
When being in the first object within the scope of the first watch-dog and leaving the scope of described first watch-dog, the predicted motion track of first object described in the data acquisition of the first object that video monitoring server sends according to described first watch-dog;
The smooth motion trajectories of data acquisition second target of the second target that described video monitoring server sends according to the second watch-dog received;
Described video monitoring server judges whether the arbitrary track in the smooth motion trajectories of described second target and the predicted motion track of described first object matches;
If so, then described video monitoring server determines that described second target is described first object.
In conjunction with first aspect present invention, in first aspect present invention first execution mode, described in the data acquisition of the first object that described video monitoring server sends according to described first watch-dog, the predicted motion track of first object comprises:
Described video monitoring server receives the data of the first object that described first watch-dog sends;
The movable information of first object described in the data acquisition that described video monitoring server resolves described first object;
Described video monitoring server uses the social force model based on point of interest, and obtains the predicted motion track of described first object in conjunction with the movable information of described first object.
In conjunction with first aspect present invention, in first aspect present invention second execution mode, described method also comprises:
The region MC feature of described video monitoring server first object according to the data acquisition of described first object.
In conjunction with first aspect present invention, in first aspect present invention the 3rd execution mode, described method also comprises:
If the smooth motion trajectories of described second target does not mate with the arbitrary track in the predicted motion track of described first object, then described video monitoring server receives the data of the 3rd target.
In conjunction with first aspect present invention second execution mode or first aspect present invention the 3rd execution mode, in first aspect present invention the 4th execution mode, described video monitoring server determines that described second target also comprises after being first object:
Described video monitoring server is the second order target area MC feature according to the data acquisition of described second target;
Described in the region MC characteristic sum that described video monitoring server judges described first object, whether the second order target area MC feature is identical;
If so, then described video monitoring server determines that described second target is described first object;
If not, then described video monitoring server receives the data of the 3rd target.
Second aspect present invention provides a kind of video monitoring server, comprising:
First acquisition module, for when being in the first object within the scope of the first watch-dog and leaving the scope of described first watch-dog, the predicted motion track of first object according to the data acquisition of the first object of described first watch-dog transmission;
Second acquisition module, for the smooth motion trajectories of data acquisition second target of the second target according to the second watch-dog transmission received;
Whether the first judge module, match for the arbitrary track in the predicted motion track of the smooth motion trajectories and described first object that judge described second target;
First determination module, for when a track in the smooth motion trajectories of described second target and the predicted motion track of described first object matches, determines that described second target is described first object.
In conjunction with second aspect present invention, in second aspect present invention first execution mode, described first acquisition module comprises:
Receiving element, for receiving the data of the first object that described first watch-dog sends;
Resolution unit, for resolve described first object data acquisition described in the movable information of first object;
Acquiring unit, for using the social force model based on point of interest, and obtains the predicted motion track of described first object in conjunction with the movable information of described first object.
In conjunction with second aspect present invention, in second aspect present invention second execution mode, also comprise:
3rd acquisition module, for the region MC feature of first object according to the data acquisition of described first object.
In conjunction with second aspect present invention, in second aspect present invention the 3rd execution mode, also comprise:
First receiver module, for when the arbitrary track in the smooth motion trajectories of described second target and the predicted motion track of described first object does not mate, receives the data of the 3rd target.
In conjunction with second aspect present invention second execution mode or second aspect present invention the 3rd execution mode, in second aspect present invention the 4th execution mode, also comprise:
4th acquisition module, for the second order target area MC feature according to the data acquisition of described second target;
Second judge module, for judge described first object region MC characteristic sum described in the second order target area MC feature whether identical;
Second determination module, time identical for the second order target area MC feature described in the region MC characteristic sum when described first object, determines that described second target is described first object;
Second receiver module, time not identical for the second order target area MC feature described in the region MC characteristic sum when described first object, receives the data of the 3rd target.
As can be seen from the above technical solutions, the embodiment of the present invention has the following advantages:
When being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, the predicted motion track of the data acquisition first object of the first object that video monitoring server sends according to the first watch-dog, the smooth motion trajectories of data acquisition second target of the second target that video monitoring server sends according to the second watch-dog received, video monitoring server judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, if, then video monitoring server determines that the second target is first object, the data of the second target that video monitoring server receives according to different monitoring equipment obtain the smooth motion trajectories of the second target, smooth motion trajectories is compared with needing the predicted motion track of the first object monitored the relation judging the second target and first object, thus the target following realized between many watch-dogs, the accuracy rate of mating when improve target following.
Accompanying drawing explanation
Fig. 1 is an embodiment schematic diagram of a kind of video frequency monitoring method in the embodiment of the present invention;
Fig. 2 is another embodiment schematic diagram of a kind of video frequency monitoring method in the embodiment of the present invention;
Fig. 3 is another embodiment schematic diagram of a kind of video frequency monitoring method in the embodiment of the present invention;
Fig. 4 is an embodiment schematic diagram of a kind of video monitoring server in the embodiment of the present invention;
Fig. 5 is another embodiment schematic diagram of a kind of video monitoring server in the embodiment of the present invention;
Fig. 6 is another embodiment schematic diagram of a kind of video monitoring server in the embodiment of the present invention;
Fig. 7 is the mutual schematic diagram in the specific embodiment of the invention in video monitoring server between each module.
Embodiment
Embodiments provide a kind of video frequency monitoring method and video monitoring server, for the tracking of realize target between many watch-dogs, the accuracy rate of mating when improving target following.
Refer to Fig. 1, in the embodiment of the present invention, an a kind of embodiment of video frequency monitoring method comprises:
101, the predicted motion track of the data acquisition first object of first object that sends according to the first watch-dog of video monitoring server;
In the embodiment of the present invention, when being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, video monitoring server receives the data of the first object that the first watch-dog sends, and video monitoring server is according to the predicted motion track of the data acquisition first object of first object.
102, video monitoring server is according to the smooth motion trajectories of data acquisition second target of the second target of the second watch-dog transmission received;
In the embodiment of the present invention, video monitoring server receives the data of the second target that the second watch-dog sends, and video monitoring server is according to the smooth motion trajectories of data acquisition second target of the second target.
103, video monitoring server judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, and if so, then performs step 104;
In the embodiment of the present invention, video monitoring server judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object obtained mates, if so, then execution step 104.
104, video monitoring server determines that the second target is first object.
In the embodiment of the present invention, when a track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, video monitoring server determines that the second target is first object.
In the embodiment of the present invention, when being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, the predicted motion track of the data acquisition first object of the first object that video monitoring server sends according to the first watch-dog, the smooth motion trajectories of data acquisition second target of the second target that video monitoring server sends according to the second watch-dog received, video monitoring server judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, if, then video monitoring server determines that the second target is first object, the data of the second target that video monitoring server receives according to different monitoring equipment obtain the smooth motion trajectories of the second target, smooth motion trajectories is compared with needing the predicted motion track of the first object monitored the relation judging the second target and first object, thus the target following realized between many watch-dogs, the accuracy rate of mating when improve target following.
Refer to Fig. 2, in the embodiment of the present invention, another embodiment of a kind of video frequency monitoring method comprises:
201, video monitoring server receives the data of the first object that the first watch-dog sends;
In the embodiment of the present invention, when being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, video monitoring server receives the data of the first object that the first watch-dog sends.
202, video monitoring server resolves the movable information of the data acquisition first object of first object;
In the embodiment of the present invention, video monitoring server resolves the movable information of the data acquisition first object of the first object received, as speed etc.
203, video monitoring server uses based on the social force model of point of interest, and obtains the predicted motion track of first object in conjunction with the movable information of first object;
In the embodiment of the present invention, video monitoring server uses the social force model based on point of interest, and the movable information acquisition first object combining the first object obtained leaves the predicted motion track after the first watch-dog scope.
204, video monitoring server is according to the smooth motion trajectories of data acquisition second target of the second target of the second watch-dog transmission received;
In the embodiment of the present invention, video monitoring server receives the data of the second target that the second watch-dog sends, and video monitoring server is according to the smooth motion trajectories of data acquisition second target of the second target.
205, video monitoring server judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, and if so, then performs step 206;
In the embodiment of the present invention, video monitoring server judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object obtained mates, if so, then execution step 206.
206, video monitoring server determines that the second target is first object.
In the embodiment of the present invention, when a track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, video monitoring server determines that the second target is first object.
In the embodiment of the present invention, when being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, the predicted motion track of the data acquisition first object of the first object that video monitoring server sends according to the first watch-dog, the smooth motion trajectories of data acquisition second target of the second target that video monitoring server sends according to the second watch-dog received, video monitoring server judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, if, then video monitoring server determines that the second target is first object, the data of the second target that video monitoring server receives according to different monitoring equipment obtain the smooth motion trajectories of the second target, smooth motion trajectories is compared with needing the predicted motion track of the first object monitored the relation judging the second target and first object, thus the target following realized between many watch-dogs, the accuracy rate of mating when improve target following.
Secondly, video monitoring server obtains the refinement of predicted motion track step, makes scheme more detailed.
Refer to Fig. 3, in the embodiment of the present invention, another embodiment of a kind of video frequency monitoring method comprises:
301, the predicted motion track of the data acquisition first object of first object that sends according to the first watch-dog of video monitoring server;
In the embodiment of the present invention, when being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, video monitoring server receives the data of the first object that the first watch-dog sends, and video monitoring server is according to the predicted motion track of the data acquisition first object of first object.
302, video monitoring server is according to the region MC feature of the data acquisition first object of first object;
In the embodiment of the present invention, video monitoring server goes out the region MC feature of first object from the extracting data of the first object got.
303, video monitoring server is according to the smooth motion trajectories of data acquisition second target of the second target of the second watch-dog transmission received;
In the embodiment of the present invention, video monitoring server receives the data of the second target that the second watch-dog sends, and video monitoring server is according to the smooth motion trajectories of data acquisition second target of the second target.
304, video monitoring server is according to the data acquisition second order target area MC feature of the second target;
In the embodiment of the present invention, video monitoring server goes out the second order target area MC feature from the extracting data of the second target got.
305, video monitoring server judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, and if so, then performs step 306; If not, then step 307 is performed;
In the embodiment of the present invention, video monitoring server judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object obtained mates, if so, then execution step 306; If not, then step 308 is performed.
306, video monitoring server determines that the second target is first object;
In the embodiment of the present invention, when a track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, video monitoring server determines that the second target is first object.
307, video monitoring server judges that whether MC characteristic sum second order target area, the region MC feature of first object is identical, if so, then performs step 306, if not, then performs step 308;
In the embodiment of the present invention, video monitoring server judges that whether the second order target area MC feature that the region MC characteristic sum of the first object got gets is identical, if so, then performs step 306, if not, then performs step 308.
308, video monitoring server receives the data of the 3rd target.
In the embodiment of the present invention, when the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object does not mate or MC characteristic sum second order target area, the region MC feature of first object is not identical, video monitoring server receives the data of the 3rd target.
In the embodiment of the present invention, when being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, the predicted motion track of the data acquisition first object of the first object that video monitoring server sends according to the first watch-dog, the smooth motion trajectories of data acquisition second target of the second target that video monitoring server sends according to the second watch-dog received, video monitoring server judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, if, then video monitoring server determines that the second target is first object, the data of the second target that video monitoring server receives according to different monitoring equipment obtain the smooth motion trajectories of the second target, smooth motion trajectories is compared with needing the predicted motion track of the first object monitored the relation judging the second target and first object, thus the target following realized between many watch-dogs, the accuracy rate of mating when improve target following.
Secondly, adding of the auxiliary judgment step of region MC feature, make scheme more rigorous.
For convenience of understanding, with an embodiment, a kind of video frequency monitoring method is described in detail below, concrete:
The present invention works in coordination with tracking technique based on the non-viewing visual field multiple-camera of many points of interest social force model, in the guarded region of the known non-overlapped multiple-camera of scene, the social force model based on many points of interest is used to complete coupling and the tracking of moving target, achieve the Continuous Tracking of humanbody moving object in the non-overlapped monitoring ken, in the present invention, what use is social force model based on many points of interest, need first to determine the spatial distribution of whole guarded region and the topological relation of video camera, determine the point of interest in pedestrian movement's process and barrier, and the space constraint of pedestrian in motion process, then the space structure figure of guarded region is drawn out, according to the space structure figure of guarded region, use the movement locus of social force model prediction pedestrian,
Utilize the target following of single camera can obtain the smooth motion trajectories of moving target in each video camera, in the present invention, the target following of single camera adopts the method for meanShift, and extract the linked character of target, what mainly use here is the region MC feature that the people such as Li Bo propose, concrete mode is: first in single camera, detect target, according to existing method, target is divided into head, upper body and lower body part.In RGB color space, object pixel cluster is carried out to latter two part, adopts the Euclidean distance of RGB color space
Wherein, C
1and C
2represent the position of two pixels in RGB color space; If target A obtains M MC class by cluster, for:
the dot frequency of each class is expressed as: p (A)={ p (A
1), p (A
2) ..., p (A
i) ..., p (A
m); Set up the coordinate system being the origin of coordinates with target block center, can estimate the spatial distribution of target MC class like this, by cluster result, can obtain the spatial distribution of MC class, the height of cluster areas is set to D
max, the MC space-like positional representation of target A is:
Coordinate system center is L to the longest distance at edge
max, use formula:
calculate each center to coordinate original place
distance
Calculate the position weight of each color class:
and utilized normalization to obtain the relative weight coefficient of each color class:
Spatial distribution relative weight coefficient
with domain color p (A
i) be multiplied after renormalization process, obtain the domain color feature of Fusion of Color space-like distributing position weight:
Domain color feature p
sp(A
i) merge the Color-spatial distribution information of target, the interference of edge pixel can be processed preferably, and the coupling contribution rate of the MC feature relevant to object matching can also be improved;
To use based on the social force model predicted motion target of many points of interest at the movement locus without camera supervised region, for object matching is prepared, use in the present invention
represent pedestrian P
iin the coordinate of moment t in video camera c, wherein
represent pedestrian P
ithe time interval occurred in camera c.When
represent pedestrian P
inot in the monitoring of video camera 1, namely can start to predict the movement locus of pedestrian in non-supervised region, be specially:
Pedestrian P
ithe attraction of concrete point of interest can be subject in the process of motion
with the repulsive force of wall or barrier
comprehensive function, the exercising result of this two types power can affect movement velocity and the direction of motion of pedestrian, is presented as with formula:
The present invention utilizes movement velocity and the direction of motion of pedestrian, coordinate at the end of also having pedestrian to monitor in a upper video camera, determination pedestrian is step by step at the coordinate in each moment in non-supervised region, consider the unexpected acceleration of pedestrian in motion process or deceleration, in the process determining pedestrian's coordinate, also add the process of rate smoothing, with formula:
Calculate the coordinate of pedestrian's subsequent time, wherein
Front T
pthe average speed of frame, is used for the motion of level and smooth pedestrian;
In time not hindering power effect, pedestrian will only receive the effect of point of interest to the attraction of pedestrian, can pass through formula:
Calculating is tried to achieve, wherein
pedestrian P
itowards point of interest direction
desired speed, the maximal rate of pedestrian in a upper CCTV camera can be thought, τ
ifor time relaxation coefficient;
Be subject to the size of the repulsive force of wall and barrier and pedestrian and wall in pedestrian movement's process, the distance of barrier is inversely proportional to, and, because pedestrian is only subject to the active force impact of the barrier before them in the process of motion, so the repulsive force sphere of action of barrier is [-90 °, 90 °], repulsive force can use formula:
wherein A
bfor weights, B
bfor the distance of effect;
The Continuous Tracking of target, in conjunction with topological relation and the movement locus of video camera, obtain the complete trajectory of moving target, pedestrian, first in first video camera a, obtains level and smooth movement locus, in conjunction with the space topological figure of observation area, find out the next camera b that pedestrian may occur, and obtain the smooth track of some moving targets, here is exactly find out the matching relationship of moving target, is specially:
Suppose
for one that occurs in video camera b meets the moving target (doing same following process for other moving targets meeting time window) of time window, for certain prediction locus
(prediction locus for other does same process), when
time, the direction of motion of this prediction locus is set to
use coordinate determination formula, then along pedestrian movement's track that can observe, make prediction locus extend T
projframe, wherein Δ
tit is a time interval, finally from all prediction locus
in find out one and track
nearest identifies pedestrian P again
r;
Further, the present invention extends prediction locus
towards
wherein
and use formula:
Calculate T
proj;
Further, the present invention's definition
for prediction locus
with the movement locus of video camera b tracking pedestrians r
between Euclidean distance, finally use step 2 extract target MC characteristic sum prediction locus and pursuit path between Euclidean distance weighted registration moving target;
Obtain the smooth motion trajectories of pedestrian between two zero lap video cameras, complete the tracking of pedestrian, multiple video camera is similar, is all the coupling between two video cameras eventually.
Refer to Fig. 4, in the embodiment of the present invention, an a kind of embodiment of video monitoring server comprises:
First acquisition module 401, for when being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, according to the predicted motion track of the data acquisition first object of the first object of the first watch-dog transmission;
Second acquisition module 402, for the smooth motion trajectories of data acquisition second target of the second target according to the second watch-dog transmission received;
First judge module 403, whether the arbitrary track in the predicted motion track of the first object that smooth motion trajectories and the first acquisition module 401 for judging the second target that the second acquisition module 402 gets get matches;
First determination module 404, when a track for judging in the smooth motion trajectories of the second target and the predicted motion track of first object when the first judge module 403 matches, determines that the second target is first object.
In the embodiment of the present invention, when being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, the predicted motion track of the data acquisition first object of the first object that the first acquisition module 401 sends according to the first watch-dog, the smooth motion trajectories of data acquisition second target of the second target that the second acquisition module 402 sends according to the second watch-dog received, first judge module 403 judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, if, first determination module 404 determines that the second target is first object, the data of the second target that the second acquisition module 402 receives according to different monitoring equipment obtain the smooth motion trajectories of the second target, smooth motion trajectories is compared with needing the predicted motion track of the first object monitored the relation judging the second target and first object by the first judge module 403, thus the target following realized between many watch-dogs, the accuracy rate of mating when improve target following.
Refer to Fig. 5, in the embodiment of the present invention, another embodiment of a kind of video monitoring server comprises:
First acquisition module 501, for when being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, according to the predicted motion track of the data acquisition first object of the first object of the first watch-dog transmission;
Second acquisition module 502, for the smooth motion trajectories of data acquisition second target of the second target according to the second watch-dog transmission received;
First judge module 503, whether the arbitrary track in the predicted motion track of the first object that smooth motion trajectories and the first acquisition module 501 for judging the second target that the second acquisition module 502 gets get matches;
First determination module 504, when a track for judging in the smooth motion trajectories of the second target and the predicted motion track of first object when the first judge module 503 matches, determines that the second target is first object.
In the embodiment of the present invention, the first acquisition module 501 comprises:
Receiving element 5011, for receiving the data of the first object that the first watch-dog sends;
Resolution unit 5012, for resolving the movable information of the data acquisition first object of the first object that receiving element 5011 receives;
Acquiring unit 5013, for using the social force model based on point of interest, and the movable information of the first object obtained in conjunction with resolution unit 5012 obtains the predicted motion track of first object.
In the embodiment of the present invention, when being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, the predicted motion track of the data acquisition first object of the first object that the first acquisition module 501 sends according to the first watch-dog, the smooth motion trajectories of data acquisition second target of the second target that the second acquisition module 502 sends according to the second watch-dog received, first judge module 503 judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, if, first determination module 504 determines that the second target is first object, the data of the second target that the second acquisition module 502 receives according to different monitoring equipment obtain the smooth motion trajectories of the second target, smooth motion trajectories is compared with needing the predicted motion track of the first object monitored the relation judging the second target and first object by the first judge module 503, thus the target following realized between many watch-dogs, the accuracy rate of mating when improve target following.
Secondly, the refinement of the first acquisition module 501 makes scheme more concrete.
Refer to Fig. 6, in the embodiment of the present invention, an a kind of embodiment of video monitoring server comprises:
First acquisition module 601, for when being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, according to the predicted motion track of the data acquisition first object of the first object of the first watch-dog transmission;
Second acquisition module 602, for the smooth motion trajectories of data acquisition second target of the second target according to the second watch-dog transmission received;
First judge module 603, whether the arbitrary track in the predicted motion track of the first object that smooth motion trajectories and the first acquisition module 601 for judging the second target that the second acquisition module 602 gets get matches;
First determination module 604, when a track for judging in the smooth motion trajectories of the second target and the predicted motion track of first object when the first judge module 603 matches, determines that the second target is first object.
In the embodiment of the present invention, video monitoring server also comprises:
3rd acquisition module 605, for the region MC feature of the data acquisition first object according to first object;
First receiver module 606, for when the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object does not mate, receives the data of the 3rd target;
4th acquisition module 607, for the data acquisition second order target area MC feature according to the second target;
Whether the second judge module 608 is identical for judging MC characteristic sum second order target area, the region MC feature of first object;
Second determination module 609, for when MC characteristic sum second order target area, the region MC feature of first object is identical, determines that the second target is first object;
Second receiver module 610, for when MC characteristic sum second order target area, the region MC feature of first object is not identical, receives the data of the 3rd target.
In the embodiment of the present invention, when being in the first object within the scope of the first watch-dog and leaving the scope of the first watch-dog, the predicted motion track of the data acquisition first object of the first object that the first acquisition module 601 sends according to the first watch-dog, the smooth motion trajectories of data acquisition second target of the second target that the second acquisition module 602 sends according to the second watch-dog received, first judge module 603 judges whether the arbitrary track in the smooth motion trajectories of the second target and the predicted motion track of first object matches, if, first determination module 604 determines that the second target is first object, the data of the second target that the second acquisition module 602 receives according to different monitoring equipment obtain the smooth motion trajectories of the second target, smooth motion trajectories is compared with needing the predicted motion track of the first object monitored the relation judging the second target and first object by the first judge module 603, thus the target following realized between many watch-dogs, the accuracy rate of mating when improve target following.
Secondly, the 3rd acquisition module 605 and make scheme more perfect adding of the second judge module 608.
Below with a concrete implementation to being described in detail alternately between module each in embodiment of the present invention video monitoring server:
The present invention works in coordination with tracking technique based on the non-viewing visual field multiple-camera of many points of interest social force model, in the guarded region of the known non-overlapped multiple-camera of scene, the social force model based on many points of interest is used to complete coupling and the tracking of moving target, achieve the Continuous Tracking of humanbody moving object in the non-overlapped monitoring ken, in the present invention, what use is social force model based on many points of interest, need first to determine the spatial distribution of whole guarded region and the topological relation of video camera, determine the point of interest in pedestrian movement's process and barrier, and the space constraint of pedestrian in motion process, then the space structure figure of guarded region is drawn out, according to the space structure figure of guarded region, use the movement locus of social force model prediction pedestrian,
First acquisition module 701 and the second acquisition module 702 utilize the target following of single camera can obtain the smooth motion trajectories of moving target in each video camera, in the present invention, the target following of single camera adopts the method for meanShift, 3rd acquisition module 705 and the 4th acquisition module 707 extract order target area MC feature, what mainly use here is the region MC feature that the people such as Li Bo propose, concrete mode is: first in single camera, detect target, according to existing method, target is divided into head, upper body and lower body part.In RGB color space, object pixel cluster is carried out to latter two part, adopts the Euclidean distance of RGB color space
Wherein, C
1and C
2represent the position of two pixels in RGB color space; If target A obtains M MC class by cluster, for:
the dot frequency of each class is expressed as: p (A)={ p (A
1), p (A
2) ..., p (A
i) ..., p (A
m); Set up the coordinate system being the origin of coordinates with target block center, can estimate the spatial distribution of target MC class like this, by cluster result, can obtain the spatial distribution of MC class, the height of cluster areas is set to D
max, the MC space-like positional representation of target A is:
Coordinate system center is L to the longest distance at edge
max, use formula:
calculate each center to coordinate original place
distance
calculate the position weight of each color class:
and utilized normalization to obtain the relative weight coefficient of each color class:
Spatial distribution relative weight coefficient
with domain color p (A
i) be multiplied after renormalization process, obtain the domain color feature of Fusion of Color space-like distributing position weight:
Domain color feature p
sp(A
i) merge the Color-spatial distribution information of target, the interference of edge pixel can be processed preferably, and the coupling contribution rate of the MC feature relevant to object matching can also be improved;
Acquiring unit 7011 in first acquisition module 701 uses social force model predicted motion target based on many points of interest at the movement locus without camera supervised region, for object matching is prepared, uses in the present invention
represent pedestrian P
iin the coordinate of moment t in video camera c, wherein
represent pedestrian P
ithe time interval occurred in camera c, when
represent pedestrian P
inot in the monitoring of video camera 1, namely can start to predict the movement locus of pedestrian in non-supervised region, be specially:
Pedestrian P
ithe attraction of concrete point of interest can be subject in the process of motion
with the repulsive force of wall or barrier
comprehensive function, the exercising result of this two types power can affect movement velocity and the direction of motion of pedestrian, is presented as with formula:
The present invention utilizes movement velocity and the direction of motion of pedestrian, coordinate at the end of also having pedestrian to monitor in a upper video camera, determination pedestrian is step by step at the coordinate in each moment in non-supervised region, consider the unexpected acceleration of pedestrian in motion process or deceleration, in the process determining pedestrian's coordinate, also add the process of rate smoothing, with formula:
Calculate the coordinate of pedestrian's subsequent time, wherein
Front T
pthe average speed of frame, is used for the motion of level and smooth pedestrian; In time not hindering power effect, pedestrian will only receive the effect of point of interest to the attraction of pedestrian, can pass through formula:
Calculating is tried to achieve, wherein
pedestrian P
itowards point of interest direction
desired speed, the maximal rate of pedestrian in a upper CCTV camera can be thought, τ
ifor time relaxation coefficient;
Be subject to the size of the repulsive force of wall and barrier and pedestrian and wall in pedestrian movement's process, the distance of barrier is inversely proportional to, and, because pedestrian is only subject to the active force impact of the barrier before them in the process of motion, so the repulsive force sphere of action of barrier is [-90 °, 90 °], repulsive force can use formula:
wherein A
bfor weights, B
bfor the distance of effect;
The Continuous Tracking of target, in conjunction with topological relation and the movement locus of video camera, obtain the complete trajectory of moving target, pedestrian, first in first video camera a, obtains level and smooth movement locus, in conjunction with the space topological figure of observation area, find out the next camera b that pedestrian may occur, and obtain the smooth track of some moving targets, the first judge module 703 finds out the matching relationship of moving target, is specially:
Suppose
for one that occurs in video camera b meets the moving target (doing same following process for other moving targets meeting time window) of time window, for certain prediction locus
(prediction locus for other does same process), when
time, the direction of motion of this prediction locus is set to
use coordinate determination formula, then along pedestrian movement's track that can observe, make prediction locus extend T
projframe, wherein Δ
tit is a time interval, finally from all prediction locus
in find out one and track
nearest identifies pedestrian P again
r;
Further, the present invention extends prediction locus
towards
wherein
and use formula:
Calculate T
proj;
Further, the present invention's definition
for prediction locus
with the movement locus of video camera b tracking pedestrians r
between Euclidean distance, the Euclidean distance weighted registration moving target between the target area MC characteristic sum prediction locus finally using the 3rd acquiring unit 705 and the 4th acquiring unit 706 to extract and pursuit path;
Obtain the smooth motion trajectories of pedestrian between two zero lap video cameras, complete the tracking of pedestrian, multiple video camera is similar, is all the coupling between two video cameras eventually.
Those skilled in the art can be well understood to, and for convenience and simplicity of description, the system of foregoing description, the specific works process of device and unit, with reference to the corresponding process in preceding method embodiment, can not repeat them here.
In several embodiments that the application provides, should be understood that, disclosed system, apparatus and method, can realize by another way.Such as, device embodiment described above is only schematic, such as, the division of described unit, be only a kind of logic function to divide, actual can have other dividing mode when realizing, such as multiple unit or assembly can in conjunction with or another system can be integrated into, or some features can be ignored, or do not perform.Another point, shown or discussed coupling each other or direct-coupling or communication connection can be by some interfaces, and the indirect coupling of device or unit or communication connection can be electrical, machinery or other form.
The described unit illustrated as separating component or can may not be and physically separates, and the parts as unit display can be or may not be physical location, namely can be positioned at a place, or also can be distributed in multiple network element.Some or all of unit wherein can be selected according to the actual needs to realize the object of the present embodiment scheme.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, also can be that the independent physics of unit exists, also can two or more unit in a unit integrated.Above-mentioned integrated unit both can adopt the form of hardware to realize, and the form of SFU software functional unit also can be adopted to realize.
If described integrated unit using the form of SFU software functional unit realize and as independently production marketing or use time, can be stored in a computer read/write memory medium.Based on such understanding, the part that technical scheme of the present invention contributes to prior art in essence in other words or all or part of of this technical scheme can embody with the form of software product, this computer software product is stored in a storage medium, comprising some instructions in order to make a computer equipment (can be personal computer, server, or the network equipment etc.) perform all or part of step of method described in each embodiment of the present invention.And aforesaid storage medium comprises: USB flash disk, portable hard drive, read-only memory (ROM, Read-OnlyMemory), random access memory (RAM, Random Access Memory), magnetic disc or CD etc. various can be program code stored medium.
The above, above embodiment only in order to technical scheme of the present invention to be described, is not intended to limit; Although with reference to previous embodiment to invention has been detailed description, those of ordinary skill in the art is to be understood that: it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein portion of techniques feature; And these amendments or replacement, do not make the essence of appropriate technical solution depart from the spirit and scope of various embodiments of the present invention technical scheme.
Claims (10)
1. a video frequency monitoring method, is characterized in that, comprising:
When being in the first object within the scope of the first watch-dog and leaving the scope of described first watch-dog, the predicted motion track of first object described in the data acquisition of the first object that video monitoring server sends according to described first watch-dog;
The smooth motion trajectories of data acquisition second target of the second target that described video monitoring server sends according to the second watch-dog received;
Described video monitoring server judges whether the arbitrary track in the smooth motion trajectories of described second target and the predicted motion track of described first object matches;
If so, then described video monitoring server determines that described second target is described first object.
2. video frequency monitoring method according to claim 1, is characterized in that, described in the data acquisition of the first object that described video monitoring server sends according to described first watch-dog, the predicted motion track of first object comprises:
Described video monitoring server receives the data of the first object that described first watch-dog sends;
The movable information of first object described in the data acquisition that described video monitoring server resolves described first object;
Described video monitoring server uses the social force model based on point of interest, and obtains the predicted motion track of described first object in conjunction with the movable information of described first object.
3. video frequency monitoring method according to claim 1, is characterized in that, described method also comprises:
The region MC feature of described video monitoring server first object according to the data acquisition of described first object.
4. video frequency monitoring method according to claim 1, is characterized in that, described method also comprises:
If the smooth motion trajectories of described second target does not mate with the arbitrary track in the predicted motion track of described first object, then described video monitoring server receives the data of the 3rd target.
5. the video frequency monitoring method according to claim 3 or 4, is characterized in that, described video monitoring server determines that described second target also comprises after being first object:
Described video monitoring server is the second order target area MC feature according to the data acquisition of described second target;
Described in the region MC characteristic sum that described video monitoring server judges described first object, whether the second order target area MC feature is identical;
If so, then described video monitoring server determines that described second target is described first object;
If not, then described video monitoring server receives the data of the 3rd target.
6. a video monitoring server, is characterized in that, comprising:
First acquisition module, for when being in the first object within the scope of the first watch-dog and leaving the scope of described first watch-dog, the predicted motion track of first object according to the data acquisition of the first object of described first watch-dog transmission;
Second acquisition module, for the smooth motion trajectories of data acquisition second target of the second target according to the second watch-dog transmission received;
Whether the first judge module, match for the arbitrary track in the predicted motion track of the smooth motion trajectories and described first object that judge described second target;
First determination module, for when a track in the smooth motion trajectories of described second target and the predicted motion track of described first object matches, determines that described second target is described first object.
7. video monitoring server according to claim 6, is characterized in that, described first acquisition module comprises:
Receiving element, for receiving the data of the first object that described first watch-dog sends;
Resolution unit, for resolve described first object data acquisition described in the movable information of first object;
Acquiring unit, for using the social force model based on point of interest, and obtains the predicted motion track of described first object in conjunction with the movable information of described first object.
8. video monitoring server according to claim 6, is characterized in that, also comprises:
3rd acquisition module, for the region MC feature of first object according to the data acquisition of described first object.
9. video monitoring server according to claim 6, is characterized in that, also comprises:
First receiver module, for when the arbitrary track in the smooth motion trajectories of described second target and the predicted motion track of described first object does not mate, receives the data of the 3rd target.
10. video frequency monitoring method according to claim 8 or claim 9, is characterized in that, also comprise:
4th acquisition module, for the second order target area MC feature according to the data acquisition of described second target;
Second judge module, for judge described first object region MC characteristic sum described in the second order target area MC feature whether identical;
Second determination module, time identical for the second order target area MC feature described in the region MC characteristic sum when described first object, determines that described second target is described first object;
Second receiver module, time not identical for the second order target area MC feature described in the region MC characteristic sum when described first object, receives the data of the 3rd target.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510021580.7A CN104539909A (en) | 2015-01-15 | 2015-01-15 | Video monitoring method and video monitoring server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510021580.7A CN104539909A (en) | 2015-01-15 | 2015-01-15 | Video monitoring method and video monitoring server |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104539909A true CN104539909A (en) | 2015-04-22 |
Family
ID=52855368
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510021580.7A Pending CN104539909A (en) | 2015-01-15 | 2015-01-15 | Video monitoring method and video monitoring server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104539909A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105975633A (en) * | 2016-06-21 | 2016-09-28 | 北京小米移动软件有限公司 | Motion track obtaining method and device |
| CN106250846A (en) * | 2016-07-29 | 2016-12-21 | 昆明理工大学 | A kind of public security image method for detecting based on video monitoring |
| CN106326240A (en) * | 2015-06-18 | 2017-01-11 | 中兴通讯股份有限公司 | An object moving path identifying method and system |
| CN109040709A (en) * | 2018-09-21 | 2018-12-18 | 深圳市九洲电器有限公司 | Video monitoring method and device, monitoring server and video monitoring system |
| CN109934176A (en) * | 2019-03-15 | 2019-06-25 | 艾特城信息科技有限公司 | Pedestrian's identifying system, recognition methods and computer readable storage medium |
| CN110719398A (en) * | 2018-07-12 | 2020-01-21 | 浙江宇视科技有限公司 | Face snapshot object determination method and device |
| CN110785765A (en) * | 2018-08-23 | 2020-02-11 | 深圳市大疆创新科技有限公司 | Target object motion trajectory prediction method and monitoring platform |
| CN110781732A (en) * | 2019-09-17 | 2020-02-11 | 腾讯科技（深圳）有限公司 | Control method and device for intelligent equipment switch, storage medium and electronic device |
| CN111340856A (en) * | 2018-12-19 | 2020-06-26 | 杭州海康威视系统技术有限公司 | Vehicle tracking method, device, equipment and storage medium |
| CN112434557A (en) * | 2020-10-20 | 2021-03-02 | 深圳市华橙数字科技有限公司 | Three-dimensional display method and device of motion trail, terminal and storage medium |
| CN114550449A (en) * | 2022-02-16 | 2022-05-27 | 腾讯云计算（北京）有限责任公司 | Vehicle track completion method and device, computer readable medium and electronic equipment |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040105006A1 (en) * | 2002-12-03 | 2004-06-03 | Lazo Philip A. | Event driven video tracking system |
| CN101179707A (en) * | 2007-09-21 | 2008-05-14 | 清华大学 | Wireless network video image multi-visual angle cooperation target tracking measurement method |
| CN101344965A (en) * | 2008-09-04 | 2009-01-14 | 上海交通大学 | Tracking system based on binocular camera shooting |
| CN101572804A (en) * | 2009-03-30 | 2009-11-04 | 浙江大学 | Multi-camera intelligent control method and device |
| CN101751677A (en) * | 2008-12-17 | 2010-06-23 | 中国科学院自动化研究所 | Target continuous tracking method based on multi-camera |
| CN102256065A (en) * | 2011-07-25 | 2011-11-23 | 中国科学院自动化研究所 | Automatic video condensing method based on video monitoring network |
| CN102325250A (en) * | 2011-10-13 | 2012-01-18 | 交通运输部公路科学研究所 | Social-force-model-based monitoring system |
| CN102436662A (en) * | 2011-11-29 | 2012-05-02 | 南京信息工程大学 | Human body target tracking method in nonoverlapping vision field multi-camera network |
| CN102970517A (en) * | 2012-11-28 | 2013-03-13 | 四川长虹电器股份有限公司 | Holder lens autonomous control method based on abnormal condition identification |
| CN103942811A (en) * | 2013-01-21 | 2014-07-23 | 中国电信股份有限公司 | Method and system for determining motion trajectory of characteristic object in distributed and parallel mode |
-
2015
- 2015-01-15 CN CN201510021580.7A patent/CN104539909A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040105006A1 (en) * | 2002-12-03 | 2004-06-03 | Lazo Philip A. | Event driven video tracking system |
| CN101179707A (en) * | 2007-09-21 | 2008-05-14 | 清华大学 | Wireless network video image multi-visual angle cooperation target tracking measurement method |
| CN101344965A (en) * | 2008-09-04 | 2009-01-14 | 上海交通大学 | Tracking system based on binocular camera shooting |
| CN101751677A (en) * | 2008-12-17 | 2010-06-23 | 中国科学院自动化研究所 | Target continuous tracking method based on multi-camera |
| CN101572804A (en) * | 2009-03-30 | 2009-11-04 | 浙江大学 | Multi-camera intelligent control method and device |
| CN102256065A (en) * | 2011-07-25 | 2011-11-23 | 中国科学院自动化研究所 | Automatic video condensing method based on video monitoring network |
| CN102325250A (en) * | 2011-10-13 | 2012-01-18 | 交通运输部公路科学研究所 | Social-force-model-based monitoring system |
| CN102436662A (en) * | 2011-11-29 | 2012-05-02 | 南京信息工程大学 | Human body target tracking method in nonoverlapping vision field multi-camera network |
| CN102970517A (en) * | 2012-11-28 | 2013-03-13 | 四川长虹电器股份有限公司 | Holder lens autonomous control method based on abnormal condition identification |
| CN103942811A (en) * | 2013-01-21 | 2014-07-23 | 中国电信股份有限公司 | Method and system for determining motion trajectory of characteristic object in distributed and parallel mode |
Non-Patent Citations (2)
| Title |
|---|
| TAO ZHAO等: "Real-time Wide Area Multi-Camera Stereo Tracking", 《IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION》 * |
| 李博等: "非重叠监控视频中基于颜色空间分布模型的行人匹配方法", 《计算机应用研究》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106326240A (en) * | 2015-06-18 | 2017-01-11 | 中兴通讯股份有限公司 | An object moving path identifying method and system |
| WO2017219506A1 (en) * | 2016-06-21 | 2017-12-28 | 北京小米移动软件有限公司 | Method and device for acquiring movement trajectory |
| CN105975633A (en) * | 2016-06-21 | 2016-09-28 | 北京小米移动软件有限公司 | Motion track obtaining method and device |
| CN106250846A (en) * | 2016-07-29 | 2016-12-21 | 昆明理工大学 | A kind of public security image method for detecting based on video monitoring |
| CN110719398B (en) * | 2018-07-12 | 2021-07-20 | 浙江宇视科技有限公司 | Face snapshot object determination method and device |
| CN110719398A (en) * | 2018-07-12 | 2020-01-21 | 浙江宇视科技有限公司 | Face snapshot object determination method and device |
| CN110785765A (en) * | 2018-08-23 | 2020-02-11 | 深圳市大疆创新科技有限公司 | Target object motion trajectory prediction method and monitoring platform |
| CN109040709A (en) * | 2018-09-21 | 2018-12-18 | 深圳市九洲电器有限公司 | Video monitoring method and device, monitoring server and video monitoring system |
| CN111340856B (en) * | 2018-12-19 | 2024-04-02 | 杭州海康威视系统技术有限公司 | Vehicle tracking method, device, equipment and storage medium |
| CN111340856A (en) * | 2018-12-19 | 2020-06-26 | 杭州海康威视系统技术有限公司 | Vehicle tracking method, device, equipment and storage medium |
| CN109934176B (en) * | 2019-03-15 | 2021-09-10 | 艾特城信息科技有限公司 | Pedestrian recognition system, recognition method, and computer-readable storage medium |
| CN109934176A (en) * | 2019-03-15 | 2019-06-25 | 艾特城信息科技有限公司 | Pedestrian's identifying system, recognition methods and computer readable storage medium |
| CN110781732A (en) * | 2019-09-17 | 2020-02-11 | 腾讯科技（深圳）有限公司 | Control method and device for intelligent equipment switch, storage medium and electronic device |
| CN110781732B (en) * | 2019-09-17 | 2023-08-29 | 腾讯科技（深圳）有限公司 | Control method and device for intelligent equipment switch, storage medium and electronic device |
| CN112434557A (en) * | 2020-10-20 | 2021-03-02 | 深圳市华橙数字科技有限公司 | Three-dimensional display method and device of motion trail, terminal and storage medium |
| CN114550449A (en) * | 2022-02-16 | 2022-05-27 | 腾讯云计算（北京）有限责任公司 | Vehicle track completion method and device, computer readable medium and electronic equipment |
| CN114550449B (en) * | 2022-02-16 | 2023-03-14 | 腾讯云计算（北京）有限责任公司 | Vehicle track completion method and device, computer readable medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104539909A (en) | Video monitoring method and video monitoring server | |
| Yang et al. | Real-time multiple objects tracking with occlusion handling in dynamic scenes | |
| US8300102B2 (en) | Apparatus and method for inferencing topology of multiple cameras network by tracking movement | |
| CN102622769B (en) | Multi-target tracking method by taking depth as leading clue under dynamic scene | |
| US11288820B2 (en) | System and method for transforming video data into directional object count | |
| JP2021504856A (en) | Forward collision control methods and devices, electronics, programs and media | |
| WO2016149938A1 (en) | Video monitoring method, video monitoring system and computer program product | |
| JP6474126B2 (en) | Object tracking method, apparatus and program | |
| CN103679742B (en) | Method for tracing object and device | |
| CN104008371A (en) | Regional suspicious target tracking and recognizing method based on multiple cameras | |
| CN105493502A (en) | Video monitoring method, video monitoring system, and computer program product | |
| CN103824070A (en) | Rapid pedestrian detection method based on computer vision | |
| CN102521842B (en) | Method and device for detecting fast movement | |
| Lee et al. | Real-time illegal parking detection in outdoor environments using 1-D transformation | |
| Rasheed et al. | Tracking and abnormal behavior detection in video surveillance using optical flow and neural networks | |
| CN106326866B (en) | Early warning method and device for vehicle collision | |
| Li et al. | Robust vehicle tracking for urban traffic videos at intersections | |
| CN104732236A (en) | Intelligent crowd abnormal behavior detection method based on hierarchical processing | |
| CN112633096A (en) | Passenger flow monitoring method and device, electronic equipment and storage medium | |
| Rahim et al. | An adapted point based tracking for vehicle speed estimation in linear spacing | |
| US20220138475A1 (en) | Methods and systems for crowd motion summarization via tracklet based human localization | |
| Ling et al. | Colour-based object tracking in surveillance application | |
| CN103679172A (en) | Method for detecting long-distance ground moving object via rotary infrared detector | |
| CN116798100A (en) | Face video detection method and device | |
| Bota et al. | Multi-feature walking pedestrians detection for driving assistance systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150422 |
|
| RJ01 | Rejection of invention patent application after publication |