CN102819847A - Method for extracting movement track based on PTZ mobile camera - Google Patents
Method for extracting movement track based on PTZ mobile camera Download PDFInfo
- Publication number
- CN102819847A CN102819847A CN2012102498968A CN201210249896A CN102819847A CN 102819847 A CN102819847 A CN 102819847A CN 2012102498968 A CN2012102498968 A CN 2012102498968A CN 201210249896 A CN201210249896 A CN 201210249896A CN 102819847 A CN102819847 A CN 102819847A
- Authority
- CN
- China
- Prior art keywords
- camera
- ptz
- destination object
- dollying head
- 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
Links
Images
Landscapes
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention provides a method for extracting a movement track based on a PTZ mobile camera. A target object is detected through the PTZ mobile camera and a pedestrian detection program; the mobile camera rotates at a corresponding angle, so that the target object is at the middle position of an image captured by the camera all the time, and keeps tracking the target object. The position of the target object in the current image in a current coordinate system can be calculated on the basis of size information of the target object and the information of angle change of the PTZ mobile camera, so as to determine the coordinate of the target object in the total coordinate system, namely the position in a global environment; finally, the unreliable data point is removed through a filter function, and correct position points are connected to form the movement track of the global environment. The method disclosed by the invention is high in precision and high in speed, and the extraction effect on the movement track is better than that of the existing method.
Description
Technical field
What the present invention relates to is a kind of method of image processing and analyzing technical field, specifically is a kind of movement locus method for distilling based on PTZ dollying head.
Background technology
In recent years; Along with developing rapidly of fields such as image analysis technology, psychology, investigation; And public safety becomes more and more important, and the application of abnormal behaviour recognition technology receives publicity, and in intelligent transportation system, motion analysis and public place supervisory system, application is arranged all.The movement locus of pedestrian's target is extracted is the important component part of abnormal behaviour identification, and movement locus method for distilling commonly used is based on single fixing camera or a plurality of camera arrays.
Extract for the movement locus in the scene on a large scale and usually to adopt camera array.Such as: people such as Emanuel E.Zelniker propose in " Global Abnormal Behaviour Detection Using a Network of CCTV Cameras " (utilizing the detection of closed-circuit television network global abnormal behavior) paper that " The Eighth International Workshop on Visual Surveillance-VS2008, Marseille:France " (the 8th international vision monitoring symposial) delivered passes through the global context movement locus that destination object was caught and synthesized to camera array.In order to obtain the global motion track; Need the track that each camera captures be connected; Its difficult point is to mate the same target object in the different cameras; Because the background of destination object is different in different cameras, the color histogram of correspondence image is widely different, may in different cameras, have the different objects color histogram more similar simultaneously.Through having made up a figure, each node is represented a track among the figure, and the beginning of track is known with finishing.So, if the time dependent feature similarity of different tracks in two cameras possibly these two tracks be same destination objects so.
Generally speaking, could arrive tracking results preferably based on the movement locus method for distilling of the camera array camera array of need arranging accurately.For the bigger and a limited number of situation of camera of global context, this method is also inapplicable, yet can use the dollying head to solve.In theory, use the just motion of ability captured target object in global context of single dollying head, extract the movement locus of global context, yet the mobile of camera can be introduced machine error, and improve the difficulty of camera calibration.
Summary of the invention
The present invention is directed to the above-mentioned deficiency that existing method exists, proposed a kind of movement locus method for distilling based on PTZ dollying head.The inventor considers that in tracing process the variation of the angle of the size of destination object to be detected and PTZ dollying head can be known, if can extract the movement locus of target in global context based on these multidate informations.
The present invention realizes through following technical scheme, the present invention includes following steps:
The first step, destination object is detected through PTZ dollying head; Based on the destination object that is detected,, PTZ dollying head make destination object be in the medium position that camera is caught image all the time thereby will turning over corresponding angle; In this way, make PTZ dollying head can keep tracking to destination object always;
The information that the size information of second step, based target object and the angle of PTZ dollying head change calculates the position of destination object in current coordinate system in current this two field picture;
The 3rd the step, by coordinate position in the current coordinate system and the transformation relation between different coordinates, obtain the coordinate of destination object in global coordinate system, just the position in the global context;
The 4th goes on foot, through filter function insecure data point is removed, and correct location point is connected to the movement locus of global context.
Preferably; In the said first step: the required angle that turns over of PTZ dollying head is confirmed according to the position of destination object in image; Wherein, come accurately to control the shooting head-turned angle through the velocity of rotation of fixing PTZ dollying head, the rotation time of control PTZ dollying head.
Preferably, in said second step:, obtain destination object two-dimensional coordinate and the transformation relation between the world coordinates in the global context in the image according to the principle of camera calibration:
Wherein, (i
x, i
y) be plane of delineation coordinate, (W
x, W
y, W
z) be the world coordinates in the global context, parameter f refers to the focusing length of PTZ dollying head, is unit with the pixel, and φ is the camera depression angle.Camera height h is known, and camera depression angle φ can know that focal distance f is unknown.
Preferably, destination object two-dimensional coordinate in the described image and the transformation relation between the world coordinates in the global context are in particular:
The coordinate (i of bottom centre with the destination object in the image
x, i
y) and top center coordinate (i
x+ △ i
x, i
y+ h
1), and the corresponding world coordinates (W of destination object in global context
x, W
y, 0) and (W
x, W
y, h
0) the substitution simultaneous solution, can find the solution and obtain the coordinate of destination object in current coordinate system:
Wherein, h
1Being the height of destination object in image, is unit with the pixel, in calculating, uses the destination object average height h of estimation
0
Preferably, in said the 3rd step:
Each rotation of PTZ dollying head will produce a new coordinate system k on the ground; The initial point of coordinate system k is at the intersection point place on camera sight line and ground, and the x direction of principal axis of coordinate system k is identical to the vertical projection direction on ground with the camera sight line, and the x direction of principal axis of coordinate system k is vertical each other in ground level with the y direction of principal axis of coordinate system k.
Ground to the pedestrian place is the situation of level, just the z of the world coordinates of destination object
kBe 0, world coordinates just can be reduced to two-dimensional coordinate so;
At k constantly, the intersection point of camera sight line and ground level is exactly the initial point of k coordinate system; The horizontal range of camera and k coordinate origin is exactly h/tan φ
kH is the height of camera, φ
kIt is the depression angle of camera; According to the position of destination object in image, calculate the world coordinates P of destination object
k(w
Xk, w
Yk); Wherein, tracking mode is that first Flame Image Process detects destination object, and camera rotates tracing object then, so repeatedly; Personnel control camera and rotate, and make camera realize becoming P with the intersection point of ground level
k(w
Xk, w
Yk), if k is not mobile to pedestrian's target between (k+1) moment constantly, pedestrian's pin will be in the central authorities of picture so at this moment; If k has moved to pedestrian's target between (k+1) moment constantly, this moment, the pedestrian can be in picture central authorities so, and what camera was aimed at is the position of former frame destination object pin all the time, and promptly the time is gone up hysteresis 1 frame;
(k+1) true origin in the moment is camera sight line and the intersection point of ground level, the just P after rotating
k(w
Xk, w
Yk); θ is the angle that the camera level turns over, φ
K+1It is the absolute depression angle after camera rotates; Camera is exactly h/tan φ with (k+1) horizontal range of coordinate origin
K+1Again after the Flame Image Process, with the new world coordinates P that obtains destination object
K+1(w
X (k+1), w
Y (k+1)), conversion process afterwards is identical with before.
Preferably,, can calculate the position of destination object, be in particular at global coordinate system through the transformation relation that the adjacent coordinates that extracts is:
Calculate P
k(w
Xk, w
Yk); The air line distance of the world coordinates of destination object and PTZ dollying head can be expressed as:
The angle △ θ that PTZ dollying head need rotate
iConfirm by following formula:
At last, world coordinates just can be reduced to two-dimensional coordinate (x
k, y
k), they can be obtained by computes:
Wherein
is the preliminary examination depression angle of camera constantly, and
is the angle of camera vertical moving between adjacent moment.
Preferably, in said the 4th step: be in particular, use Kalman filtering, if estimated value and observed reading have big difference, it is insecure then this data acknowledge being decided to be, and insecure world coordinates data are gone out, remaining data point is linked be movement locus.
Principle of the present invention is, owing to camera along with the motion of destination object is rotated, have corresponding transformation relation between the position of destination object in global context and angle that its size and camera in camera image rotates.Rotation information through utilizing camera and the target pedestrian information in the image; And according to the principle of camera calibration; Try to achieve the coordinate of destination object in current coordinate system; According to the transformational relation between adjacent coordinates system, obtain the position of pedestrian's target in global context again, thereby location point connection with a high credibility is extracted movement locus.
Compared with prior art, the motion of use dollying head captured target object of the present invention in global context extracts the movement locus of global context; Utilize the rotation information of camera and the target pedestrian information in the image; Additional complexity is low, and real-time is good, and dirigibility is high.
Description of drawings
Fig. 1 is the frame diagram according to the movement locus method for distilling based on PTZ dollying head provided by the invention.
Fig. 2 is the camera side view.
Fig. 3 is the pedestrian's target information in the camera image.
Fig. 4 is the coordinate system graph of a relation.
Fig. 5 is global context trajectory plane figure.
Fig. 6 is a global context track front elevation.
Embodiment
Elaborate in the face of embodiments of the invention down, present embodiment provided detailed embodiment and concrete operating process, but protection scope of the present invention is not limited to following embodiment being to implement under the prerequisite with technical scheme of the present invention.
Embodiment
Present embodiment may further comprise the steps:
The first step, destination object is detected through PTZ dollying head and pedestrian detection method.Based on the destination object that is detected,, PTZ dollying head make destination object be in the medium position that camera is caught image all the time thereby will turning over corresponding angle.In this way, PTZ dollying head can keep the tracking to destination object always.
In the present embodiment, preferably use based on the pedestrian detection method of profile and realize detection, use improved particle filter algorithm, in every frame picture, keep tracking destination object to pedestrian's target.
The required mobile angle of PTZ dollying head is confirmed according to the position of pedestrian's target in image, the velocity of rotation of fixing camera, and the rotation time of control camera just can be controlled the shooting head-turned angle more accurately.
The information that the angle of size information of second step, based target object (the general size differences of pedestrian in image is less to the result of calculation influence) and PTZ dollying head changes can calculate the position of destination object in current coordinate system in current this two field picture.
According to the principle of camera calibration, can obtain pedestrian's target two-dimensional coordinate and the transformation relation between the world coordinates in the global context in the image:
Wherein, (i
x, i
y) be plane of delineation coordinate, (W
x, W
y, W
z) be the world coordinates in the global context, parameter f refers to the focusing length of PTZ dollying head, is unit with pixel (pixel), and φ is the camera depression angle.Camera height h is known, and camera depression angle φ can know that focal distance f is unknown.
As shown in Figure 3, with the target pedestrian's in the image the coordinate (i of bottom centre
x, i
y) and the top center coordinate be (i
x+ Δ i
x, i
y+ h
1), and the corresponding world coordinates (w of destination object in global context
x, w
y, 0) and (w
x, w
y, h
0) substitution simultaneous solution (h
1Being the height of destination object in image, is unit with the pixel, in calculating, uses pedestrian's average height h of estimation
0, be 1.75m in this instance), can find the solution and obtain the coordinate of target pedestrian in current coordinate system:
The 3rd the step, by coordinate position in the current coordinate system and the transformation relation between different coordinates, obtain the coordinate of destination object in global coordinate system, just the position in the global context.
Each rotation of PTZ dollying head will produce a new coordinate system k on the ground.The initial point of coordinate system k is at the intersection point place on PTZ dollying head sight line and ground; The x direction of principal axis of coordinate system k is identical to the vertical projection direction on ground with PTZ dollying head sight line, and the x direction of principal axis of coordinate system k is vertical each other in ground level with the y direction of principal axis of coordinate system k.
If the inventor supposes that further the ground that the pedestrian belongs to is level, just the z of the world coordinates of destination object
kBe 0, world coordinates just can be reduced to two-dimensional coordinate so.
As shown in Figure 4, at k constantly, the intersection point of PTZ dollying head sight line and ground level is exactly the initial point of k coordinate system.The horizontal range of PTZ dollying head and k coordinate origin is exactly h/tan φ
kH is the height of camera, φ
kIt is the depression angle of camera.According to the position of destination object in image, the inventor can calculate the world coordinates P of destination object
k(w
Xk, w
Yk).Inventor's tracking mode is that first Flame Image Process detects destination object, and PTZ dollying head rotates tracing object then, so repeatedly.The inventor controls PTZ dollying head and rotates, and makes PTZ dollying head realize becoming P with the intersection point of ground level
k(w
Xk, w
Yk), if k is not mobile to pedestrian's target between (k+1) moment constantly, pedestrian's pin will be in the central authorities of picture so at this moment.If k has moved to pedestrian's target between (k+1) moment constantly, this moment, the pedestrian can be in picture central authorities so, and what camera was aimed at is the position of former frame destination object pin all the time, and promptly the time is gone up hysteresis 1 frame.
(k+1) true origin in the moment is PTZ dollying head sight line and the intersection point of ground level, the just P after rotating
k(w
Xk, w
Yk).θ among Fig. 4 is the angle that PTZ dollying head level turns over, φ
K+1It is the absolute depression angle after PTZ dollying head rotates.PTZ dollying head is exactly h/tan φ with (k+1) horizontal range of coordinate origin
K+1After the Flame Image Process, the inventor will obtain the new world coordinates P of destination object again
K+1(w
X (k+1), w
Y (k+1)), conversion process afterwards is identical with before.Following inventor introduces P
k(w
Xk, w
Yk) computing method.In this instance, h is 12m.
Result through second step can calculate P
k(w
Xk, w
Yk).The air line distance of the world coordinates of destination object and PTZ dollying head can be expressed as:
The angle delta θ that PTZ dollying head need rotate
iConfirm by following formula:
At last, world coordinates just can be reduced to two-dimensional coordinate (x
k, y
k), they can be obtained by computes:
In addition, depression angle φ
kObtain by following formula:
φ wherein
0Be the preliminary examination depression angle of PTZ dollying head constantly, △ φ
iIt is the angle of PTZ dollying head vertical moving between adjacent moment.
The 4th goes on foot, through filter function insecure data point is removed, and correct location point is connected to the movement locus of global context.
Use Kalman filtering in this example, these data are considered to insecure when estimated value and observed reading have big difference, and insecure world coordinates data are gone out, remaining data point are linked be movement locus.
Implementation result
According to above-mentioned steps, the inventor uses a model to be used for following the tracks of use as the PTZ dollying head of MG-TK 3518, has write down 400 groups simultaneously and has followed the tracks of videos.For each behavior sample, the inventor has manually demarcated the True Data (ground truth) of the global motion track of target pedestrian's object, is used to weigh the accuracy that method of the present invention is extracted the global motion track.All tests realize on the PC computing machine that all the major parameter of this PC computing machine is: central processing unit Intel (R) Core (TM) 2Duo CPU E75002.93GHz, internal memory 2GB.
The result shows, the rate of accuracy reached to 94.2% of present embodiment is 39.0ms to the average handling time of every two field picture.It is higher to compare the method for punctuate in advance precision comparatively commonly used, and speed is slightly fast.On processing speed, it is a lot of soon to mate this accuracy method than point of interest.
Claims (7)
1. the movement locus method for distilling based on PTZ dollying head is characterized in that, may further comprise the steps:
The first step, destination object is detected through PTZ dollying head; Based on the destination object that is detected,, PTZ dollying head make destination object be in the medium position that camera is caught image all the time thereby will turning over corresponding angle; In this way, make PTZ dollying head can keep tracking to destination object always;
The information that the size information of second step, based target object and the angle of PTZ dollying head change calculates the position of destination object in current coordinate system in current this two field picture;
The 3rd the step, by coordinate position in the current coordinate system and the transformation relation between different coordinates, obtain the coordinate of destination object in global coordinate system, just the position in the global context;
The 4th goes on foot, through filter function insecure data point is removed, and correct location point is connected to the movement locus of global context.
2. the movement locus method for distilling based on PTZ dollying head according to claim 1 is characterized in that, in the said first step:
The required angle that turns over of PTZ dollying head confirms according to the position of destination object in image, and wherein, the velocity of rotation through fixing PTZ dollying head, the rotation time of controlling PTZ dollying head come accurately to control the shooting head-turned angle.
3. the movement locus method for distilling based on PTZ dollying head according to claim 1 is characterized in that, in said second step:
According to the principle of camera calibration, obtain destination object two-dimensional coordinate and the transformation relation between the world coordinates in the global context in the image:
Wherein, (i
x, i
y) be plane of delineation coordinate, (W
x, W
y, W
z) be the world coordinates in the global context, parameter f refers to the focusing length of PTZ dollying head, is unit with the pixel, and φ is the camera depression angle; Camera height h is known, and camera depression angle φ can know that focal distance f is unknown.
4. the movement locus method for distilling based on PTZ dollying head according to claim 3 is characterized in that, destination object two-dimensional coordinate in the described image and the transformation relation between the world coordinates in the global context are in particular:
The coordinate (i of bottom centre with the destination object in the image
x, i
y) and top center coordinate (i
x+ △ i
x, i
y+ h
1), and the corresponding world coordinates (W of destination object in global context
x, W
y, 0) and (W
x, W
y, h
0) the substitution simultaneous solution, can find the solution and obtain the coordinate of destination object in current coordinate system:
Wherein, h
1Being the height of destination object in image, is unit with the pixel, in calculating, uses the destination object average height h of estimation
0
5. the movement locus method for distilling based on PTZ dollying head according to claim 1 is characterized in that, in said the 3rd step:
Each rotation of PTZ dollying head will produce a new coordinate system k on the ground; The initial point of coordinate system k is at the intersection point place on camera sight line and ground, and the x direction of principal axis of coordinate system k is identical to the vertical projection direction on ground with the camera sight line, and the x direction of principal axis of coordinate system k is vertical each other in ground level with the y direction of principal axis of coordinate system k;
Ground to the pedestrian place is the situation of level, just the z of the world coordinates of destination object
kBe 0, world coordinates just can be reduced to two-dimensional coordinate so;
At k constantly, the intersection point of camera sight line and ground level is exactly the initial point of k coordinate system; The horizontal range of camera and k coordinate origin is exactly h/tan φ
kH is the height of camera, φ
kIt is the depression angle of camera; According to the position of destination object in image, calculate the world coordinates P of destination object
k(w
Xk, w
Yk); Wherein, tracking mode is that first Flame Image Process detects destination object, and camera rotates tracing object then, so repeatedly; The control camera rotates, and makes camera realize becoming P with the intersection point of ground level
k(w
Xk, w
Yk), if k is not mobile to pedestrian's target between (k+1) moment constantly, pedestrian's pin will be in the central authorities of picture so at this moment; If k has moved to pedestrian's target between (k+1) moment constantly, this moment, the pedestrian can be in picture central authorities so, and what camera was aimed at is the position of former frame destination object pin all the time, and promptly the time is gone up hysteresis 1 frame;
(k+1) true origin in the moment is camera sight line and the intersection point of ground level, the just P after rotating
k(w
Xk, w
Yk); θ is the angle that the camera level turns over, φ
K+1It is the absolute depression angle after camera rotates; Camera is exactly h/tan φ with (k+1) horizontal range of coordinate origin
K+1Again after the Flame Image Process, with the new world coordinates P that obtains destination object
K+1(w
X (k+1), w
Y (k+1)), conversion process afterwards is identical with before.
6. the movement locus method for distilling based on PTZ dollying head according to claim 5 is characterized in that, through the transformation relation that the adjacent coordinates that extracts is, can calculate the position of destination object at global coordinate system, is in particular:
Calculate P
k(w
Xk, w
Yk); The air line distance of the world coordinates of destination object and PTZ dollying head can be expressed as:
The angle delta θ that PTZ dollying head need rotate
iConfirm by following formula:
At last, world coordinates just can be reduced to two-dimensional coordinate (x
k, y
k), they can be obtained by computes:
7. the movement locus method for distilling based on PTZ dollying head according to claim 1 is characterized in that, in said the 4th step:
Be in particular, use Kalman filtering, if estimated value and observed reading have big difference, it is insecure then this data acknowledge being decided to be, and insecure world coordinates data are gone out, remaining data point is linked be movement locus.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210249896.8A CN102819847B (en) | 2012-07-18 | 2012-07-18 | Based on the movement locus extracting method of PTZ dollying head |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210249896.8A CN102819847B (en) | 2012-07-18 | 2012-07-18 | Based on the movement locus extracting method of PTZ dollying head |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102819847A true CN102819847A (en) | 2012-12-12 |
CN102819847B CN102819847B (en) | 2016-01-13 |
Family
ID=47303950
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210249896.8A Active CN102819847B (en) | 2012-07-18 | 2012-07-18 | Based on the movement locus extracting method of PTZ dollying head |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102819847B (en) |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103414872A (en) * | 2013-07-16 | 2013-11-27 | 南京师范大学 | Method for driving PTZ cameras through target position |
CN103826103B (en) * | 2014-02-27 | 2017-03-22 | 浙江宇视科技有限公司 | Cruise control method for tripod head video camera |
CN106887014A (en) * | 2017-01-13 | 2017-06-23 | 中山大学 | A kind of pedestrian track matching process across camera |
CN107087141A (en) * | 2017-05-09 | 2017-08-22 | 浙江宇视科技有限公司 | A kind of security target method for real-time monitoring and control device |
CN107566787A (en) * | 2017-08-15 | 2018-01-09 | 武汉烽火众智数字技术有限责任公司 | A kind of video pictures relay player method, system and device |
CN107682660A (en) * | 2017-08-15 | 2018-02-09 | 武汉烽火众智数字技术有限责任公司 | Video pictures relay player method, system and device |
WO2018086360A1 (en) * | 2016-11-08 | 2018-05-17 | Zhejiang Dahua Technology Co., Ltd. | Methods and systems for data visualization |
CN108961313A (en) * | 2018-06-29 | 2018-12-07 | 大连民族大学 | Vertical view pedestrian's risk quantification method of two-dimensional world coordinate system |
CN109034104A (en) * | 2018-08-15 | 2018-12-18 | 罗普特(厦门)科技集团有限公司 | A kind of scene tag localization method and device |
CN109558831A (en) * | 2018-11-27 | 2019-04-02 | 成都索贝数码科技股份有限公司 | It is a kind of fusion space-time model across camera shooting head's localization method |
CN109831616A (en) * | 2017-11-23 | 2019-05-31 | 上海未来伙伴机器人有限公司 | A kind of face follower method and its device based on monocular cam |
CN111259825A (en) * | 2020-01-19 | 2020-06-09 | 成都依能科技股份有限公司 | PTZ scanning path generation method based on face recognition |
CN111479055A (en) * | 2020-04-10 | 2020-07-31 | Oppo广东移动通信有限公司 | Shooting method, shooting device, electronic equipment and storage medium |
CN112183170A (en) * | 2019-07-05 | 2021-01-05 | 浙江宇视科技有限公司 | Method, device and system for generating pedestrian road network |
CN113033441A (en) * | 2021-03-31 | 2021-06-25 | 广州敏视数码科技有限公司 | Pedestrian collision early warning method based on wide-angle imaging |
CN114938429A (en) * | 2022-05-20 | 2022-08-23 | 重庆紫光华山智安科技有限公司 | Target tracking method, system, equipment and computer readable medium |
CN115631465A (en) * | 2022-12-22 | 2023-01-20 | 中关村科学城城市大脑股份有限公司 | Key crowd risk perception method and device, electronic equipment and readable medium |
CN115731658A (en) * | 2021-08-31 | 2023-03-03 | 国家电网有限公司 | Security positioning device and method for power system equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101173987A (en) * | 2007-10-31 | 2008-05-07 | 北京航空航天大学 | Multi-module and multi-target accurate tracking apparatus and method thereof |
CN101344965A (en) * | 2008-09-04 | 2009-01-14 | 上海交通大学 | Tracking system based on binocular camera shooting |
CN101719986A (en) * | 2009-12-30 | 2010-06-02 | 湖北莲花山计算机视觉和信息科学研究院 | PTZ tracking method and system based on multi-layered full-view modeling |
CN101840507A (en) * | 2010-04-09 | 2010-09-22 | 江苏东大金智建筑智能化系统工程有限公司 | Target tracking method based on character feature invariant and graph theory clustering |
-
2012
- 2012-07-18 CN CN201210249896.8A patent/CN102819847B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101173987A (en) * | 2007-10-31 | 2008-05-07 | 北京航空航天大学 | Multi-module and multi-target accurate tracking apparatus and method thereof |
CN101344965A (en) * | 2008-09-04 | 2009-01-14 | 上海交通大学 | Tracking system based on binocular camera shooting |
CN101719986A (en) * | 2009-12-30 | 2010-06-02 | 湖北莲花山计算机视觉和信息科学研究院 | PTZ tracking method and system based on multi-layered full-view modeling |
CN101840507A (en) * | 2010-04-09 | 2010-09-22 | 江苏东大金智建筑智能化系统工程有限公司 | Target tracking method based on character feature invariant and graph theory clustering |
Non-Patent Citations (1)
Title |
---|
许社教: "三维图形系统中两种坐标系之间的坐标变换", 《西安电子科技大学学报》 * |
Cited By (29)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103414872B (en) * | 2013-07-16 | 2016-05-25 | 南京师范大学 | A kind of target location drives the method for Pan/Tilt/Zoom camera |
CN103414872A (en) * | 2013-07-16 | 2013-11-27 | 南京师范大学 | Method for driving PTZ cameras through target position |
CN103826103B (en) * | 2014-02-27 | 2017-03-22 | 浙江宇视科技有限公司 | Cruise control method for tripod head video camera |
US11568548B2 (en) | 2016-11-08 | 2023-01-31 | Zhejiang Dahua Technology Co., Ltd. | Methods and systems for data visualization |
US11049262B2 (en) | 2016-11-08 | 2021-06-29 | Zhejiang Dahua Technology Co., Ltd. | Methods and systems for data visualization |
WO2018086360A1 (en) * | 2016-11-08 | 2018-05-17 | Zhejiang Dahua Technology Co., Ltd. | Methods and systems for data visualization |
CN106887014A (en) * | 2017-01-13 | 2017-06-23 | 中山大学 | A kind of pedestrian track matching process across camera |
CN106887014B (en) * | 2017-01-13 | 2019-10-15 | 中山大学 | A kind of pedestrian track matching process across camera |
CN107087141A (en) * | 2017-05-09 | 2017-08-22 | 浙江宇视科技有限公司 | A kind of security target method for real-time monitoring and control device |
CN107682660A (en) * | 2017-08-15 | 2018-02-09 | 武汉烽火众智数字技术有限责任公司 | Video pictures relay player method, system and device |
CN107566787A (en) * | 2017-08-15 | 2018-01-09 | 武汉烽火众智数字技术有限责任公司 | A kind of video pictures relay player method, system and device |
CN109831616A (en) * | 2017-11-23 | 2019-05-31 | 上海未来伙伴机器人有限公司 | A kind of face follower method and its device based on monocular cam |
CN109831616B (en) * | 2017-11-23 | 2022-07-26 | 上海未来伙伴机器人有限公司 | Human face following method and device based on monocular camera |
CN108961313A (en) * | 2018-06-29 | 2018-12-07 | 大连民族大学 | Vertical view pedestrian's risk quantification method of two-dimensional world coordinate system |
CN108961313B (en) * | 2018-06-29 | 2021-06-29 | 大连民族大学 | Overlooking pedestrian risk quantification method of two-dimensional world coordinate system |
CN109034104A (en) * | 2018-08-15 | 2018-12-18 | 罗普特(厦门)科技集团有限公司 | A kind of scene tag localization method and device |
CN109558831A (en) * | 2018-11-27 | 2019-04-02 | 成都索贝数码科技股份有限公司 | It is a kind of fusion space-time model across camera shooting head's localization method |
CN109558831B (en) * | 2018-11-27 | 2023-04-07 | 成都索贝数码科技股份有限公司 | Cross-camera pedestrian positioning method fused with space-time model |
CN112183170B (en) * | 2019-07-05 | 2024-04-09 | 浙江宇视科技有限公司 | Pedestrian road network generation method, device and system |
CN112183170A (en) * | 2019-07-05 | 2021-01-05 | 浙江宇视科技有限公司 | Method, device and system for generating pedestrian road network |
CN111259825B (en) * | 2020-01-19 | 2023-06-02 | 成都依能科技股份有限公司 | PTZ scanning path generation method based on face recognition |
CN111259825A (en) * | 2020-01-19 | 2020-06-09 | 成都依能科技股份有限公司 | PTZ scanning path generation method based on face recognition |
CN111479055A (en) * | 2020-04-10 | 2020-07-31 | Oppo广东移动通信有限公司 | Shooting method, shooting device, electronic equipment and storage medium |
CN111479055B (en) * | 2020-04-10 | 2022-05-20 | Oppo广东移动通信有限公司 | Shooting method and device, electronic equipment and storage medium |
CN113033441A (en) * | 2021-03-31 | 2021-06-25 | 广州敏视数码科技有限公司 | Pedestrian collision early warning method based on wide-angle imaging |
CN115731658A (en) * | 2021-08-31 | 2023-03-03 | 国家电网有限公司 | Security positioning device and method for power system equipment |
CN114938429A (en) * | 2022-05-20 | 2022-08-23 | 重庆紫光华山智安科技有限公司 | Target tracking method, system, equipment and computer readable medium |
CN114938429B (en) * | 2022-05-20 | 2023-10-24 | 重庆紫光华山智安科技有限公司 | Target tracking method, system, equipment and computer readable medium |
CN115631465A (en) * | 2022-12-22 | 2023-01-20 | 中关村科学城城市大脑股份有限公司 | Key crowd risk perception method and device, electronic equipment and readable medium |
Also Published As
Publication number | Publication date |
---|---|
CN102819847B (en) | 2016-01-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102819847B (en) | Based on the movement locus extracting method of PTZ dollying head | |
CN109887040B (en) | Moving target active sensing method and system for video monitoring | |
Sidla et al. | Pedestrian detection and tracking for counting applications in crowded situations | |
CN105550670B (en) | A kind of target object dynamically track and measurement and positioning method | |
CN110400352B (en) | Camera calibration with feature recognition | |
Shih | A robust occupancy detection and tracking algorithm for the automatic monitoring and commissioning of a building | |
CN107357286A (en) | Vision positioning guider and its method | |
CN108731587A (en) | A kind of the unmanned plane dynamic target tracking and localization method of view-based access control model | |
CN103149939A (en) | Dynamic target tracking and positioning method of unmanned plane based on vision | |
CN101969548A (en) | Active video acquiring method and device based on binocular camera shooting | |
CN102542289A (en) | Pedestrian volume statistical method based on plurality of Gaussian counting models | |
KR20150021526A (en) | Self learning face recognition using depth based tracking for database generation and update | |
CN109445453A (en) | A kind of unmanned plane Real Time Compression tracking based on OpenCV | |
CN105160649A (en) | Multi-target tracking method and system based on kernel function unsupervised clustering | |
CN107509055A (en) | A kind of rotary panorama focus identification optronic tracker and its implementation | |
Choi et al. | Ptz camera assisted face acquisition, tracking & recognition | |
Neves et al. | Acquiring high-resolution face images in outdoor environments: A master-slave calibration algorithm | |
Tian et al. | Absolute head pose estimation from overhead wide-angle cameras | |
Xu et al. | A real-time, continuous pedestrian tracking and positioning method with multiple coordinated overhead-view cameras | |
CN112785564B (en) | Pedestrian detection tracking system and method based on mechanical arm | |
Fahn et al. | A high-definition human face tracking system using the fusion of omni-directional and PTZ cameras mounted on a mobile robot | |
Kang et al. | Detection and tracking of moving objects from overlapping EO and IR sensors | |
CN107563371B (en) | Method for dynamically searching interesting region based on line laser light strip | |
Zhou et al. | Fast road detection and tracking in aerial videos | |
JP5293429B2 (en) | Moving object detection apparatus and moving object detection method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |