CN102737384A - Automatic spherical camera tracing method - Google Patents
Automatic spherical camera tracing method Download PDFInfo
- Publication number
- CN102737384A CN102737384A CN2011100876619A CN201110087661A CN102737384A CN 102737384 A CN102737384 A CN 102737384A CN 2011100876619 A CN2011100876619 A CN 2011100876619A CN 201110087661 A CN201110087661 A CN 201110087661A CN 102737384 A CN102737384 A CN 102737384A
- Authority
- CN
- China
- Prior art keywords
- trail
- video camera
- objective
- ball
- type video
- 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
Images
Abstract
The invention provides an automatic spherical camera tracing method, comprising the steps of, acquiring an image; acquiring a motion vector of an image; determining a motion state of the spherical camera; performing binaryzation of a scene image; segmenting connected objects in the motion vector; analyzing and tracing the segmented connected objects; and tracing a target object. The method of the invention uses the motion vector to assist in segmenting foreground objects. With respect to a macroblock of N*N pixels, pixels which need to be treated are 1/N*N of pixels of an original image. And for application of automatic tracing, only one object needs to be traced at a time, and the method can satisfy the requirement already, so hardware load can be reduced largely. The method is suitable for an embedded platform with poor hardware efficiency, and the spherical camera can run more fluently without an extra signal processing integrated circuit. Even if the target object suddenly accelerates, decelerates, is covered by barriers or overlaps with another object, a good tracing effect can be guaranteed.
Description
Technical field
The present invention relates to the method that a kind of ball-type video camera is followed the trail of automatically; Especially refer to a kind of cutting mode that utilizes motion vector (motion vector) to assist to do preceding scenery; And then can reduce the burden of hardware; Make the operation of ball-type video camera more smooth, and have excellent tracking effect, and be applicable to monitor, video camera, ball-type video camera or similar device.
Background technology
Quite variation of supervisory system at present; The image monitoring of many indoor or outdoors can't be contained the surrounding photograph machines that use under the restriction of all monitoring ranges more at existing fixed formula video camera; Though and the scope of its monitoring of surrounding photograph machine is big more many than conventional fixed formula video camera; But along with the change of monitoring range is big, the difficulty in image quality control also with increase.The camera lens that its difficulty comes from the surrounding photograph machine can't amplify or dwindle, for the mobile object in the monitored picture often can't be correct acquisition, cause the image maybe be too small or smudgy.
Thereby further develop out and have rotation zoom formula PTZ (Pan-Tilt-Zoom) video camera; So-called PTZ be the camera lens of expression video camera can carry out left-right rotation (Pan), on have a down dip (Tilt) with change focal length, amplification difference in functionalitys such as (Zoom); Not only can change the angle of shooting and the scope that is contained at any time through the PTZ video camera; More can adjust the size and the readability of image through the change of focal length, can obtain keeping of better monitoring effect and quality compared to the surrounding photograph machine.
And in existing target object tracer technique; That tradition and research and development course are also the longest is automatic target object features tracer technique (automatic object feature tracking technique); It mainly utilizes feature extraction (feature extraction) and aspect ratio to the method for (feature matching), from target object, captures image feature, again according to the image feature locking and the object that follows the trail of the objective; Yet in the moving of target object; The background of monitor area also and then changes usually, and target object is when rotating, and image feature also can change; Add that the PTZ video camera continues to scan action, all above-mentioned factors all can cause the degree of difficulty of characteristic comparison method in practical operation.
And another kind of existing target object tracer technique; Use two video cameras, utilize outer utmost point correspondence rule (epipolar rule), according to the unique point (feature points) of corresponding imaging surface to obtain projection fundamental matrix (projection fundamental matrix); In order to the locking and the object that follows the trail of the objective; But carry out scanning and when rotating, corresponding imaging surface also changes when video camera wherein, the corresponding relation that sets up of two video cameras changes; Utilize outer utmost point correspondence rule to calculate unique point correspondence accurately, and be not easy.
Moreover, from about following the trail of the algorithm developing history of (tracking), also there are many effective methods to be suggested; And CamShift algorithm wherein is to follow the trail of with the color probability distribution, after calculating color probability distribution and continuous iteration convergence, with center and the size of finding out this target; And then find out the object that desire is followed the trail of, if but its shortcoming for running into the object of similar color, the personage in for example following the trail of; Color is red, and the color of chair just is red, near the people goes to chair; Then the CamShift algorithm may skip on the close chair of color; Cause the mistake of tracking, and the CamShift algorithm is because of need calculating color probability distribution, so need a large amount of Data Processing and computing; Therefore signal processing integrated circuit (DSP, Digital Signal Processing digital signal processing) that need be extra comes auxiliary or comes computation assistance with dynamical computer.
Therefore; The inventor is because above-mentioned shortcoming; Phase can propose the method for following the trail of automatically through the ball-type video camera of the cutting mode of scenery before utilizing motion vector (motionvector) to assist to do, and to reduce the burden of hardware, makes the video camera operation more smooth; Be think of, the design team's system of grinding with great concentration, consumption is popular to be used to provide.
Summary of the invention
Fundamental purpose of the present invention is in the method that provides a kind of ball-type video camera to follow the trail of automatically; Cutting mode through scenery before utilizing motion vector (motion vector) to assist to do; Enable to find fast the object of desire tracking; And then direction, speed and the focal length that can control the ball-type video camera are next automatically with the bat object, make to have the automatic effect of following the trail of of excellent object, and then increase the practicality and the superiority of integral body.
An of the present invention purpose is in the method that provides a kind of ball-type video camera to follow the trail of automatically; Through the scene image binaryzation; From the embryo of scene image binaryzation, be partitioned into connection object (connected components) again with motion vector; And rely on image analysing computer to exclude noise,, and then increase whole convenience and rapidity with the object of formation desire tracking.
For reaching above-mentioned purpose, its key step of the present invention comprises: obtain an image: capture a static image in the picture of being kept watch on through the PTZ camera lens; Obtain the motion vector of image: image is cut into a plurality of huge collection blocks (macroblock), and obtain a motion vector (motion vector) by each huge collection block (macroblock) in the image; Judge the motion state of present ball-type video camera: when the ball-type video camera is mobile status, can cooperate information such as present direction of motion of ball-type video camera and speed, with the motion vector reduction of moving scene; Scene image carries out binaryzation: after the motion vector reduction, promptly carry out the scene image binaryzation, can find out the embryo of object in the scene; Cut apart the connection object in the motion vector: from the embryo of scene image binaryzation, be partitioned into connection object (connected components) with motion vector; Analysis is also followed the trail of the connection object that is partitioned into: will be communicated with object (connected components) again and exclude noise with image analysing computer, to form the object that desire is followed the trail of; And object is followed the trail of: when object become follow the trail of the objective after, promptly direction, speed and the focal length through control ball-type video camera comes automatically with clapping object, in order to the tracking of carrying out object.
Of the present invention have a following advantage:
Even 1, the object followed the trail of of this desire quickens suddenly, slow down, covered or staggered with another object by barrier, very excellent tracking effect can be arranged.
2, method of the present invention is fit to be implemented in the relatively poor embedded platform of hardware usefulness, and does not need extra signal processing integrated circuit (DSP) auxiliary, the more smoothness that just can let the ball-type video camera move.
Other characteristics of the present invention and specific embodiment can further be understood in the detailed description with following conjunction with figs..
Description of drawings
Fig. 1 is a key step schematic flow sheet of the present invention;
Fig. 2 is the steps flow chart synoptic diagram that follows the trail of the objective of the present invention;
Fig. 3 is an enforcement view of the present invention.
Description of reference numerals: 10-ball-type video camera; The 11-PTZ camera lens; The 20-object; The 30-scene image;
Step S100: obtain an image;
Step S110: the motion vector of obtaining image;
Step S120: the motion state of judging present ball-type video camera;
Step S130: scene image carries out binaryzation;
Step S140: cut apart the connection object in the motion vector;
Step S150: analyze and follow the trail of the connection object that is partitioned into;
Step S160: object is followed the trail of;
Step S161: judge that whether object is for following the trail of the objective;
Step S1611: judge whether lose the time of following the trail of the objective surpasses critical value;
Step S162: judge the size in picture that follows the trail of the objective;
Step S163: judge to follow the trail of the objective whether near the picture edge.
Embodiment
Please refer to Fig. 1 to shown in Figure 3, be the schematic flow sheet and the enforcement view of the automatic method of following the trail of of ball-type video camera of the present invention.The method key step that the present invention follows the trail of comprises:
Step S100 obtains an image: capture a static image in the picture of being kept watch on through PTZ camera lens 11;
Step S110 obtains the motion vector of image: image is cut into a plurality of huge collection blocks (macroblock), and obtain a motion vector (motion vector) by each huge collection block (macroblock) in the image;
Step S120 judges the motion state of present ball-type video camera: when ball-type video camera 10 during for mobile status, can cooperate information such as present direction of motion of ball-type video camera 10 and speed, with the motion vector reduction of moving scene;
Step S130 scene image carries out binaryzation: after the motion vector reduction, promptly carry out the scene image binaryzation, can find out the embryo of object in the scene;
Step S140 is cut apart the connection object in the motion vector: from the embryo of scene image binaryzation, be partitioned into the connection object (connected components) with motion vector;
Step S150 analyzes and follows the trail of the connection object that is partitioned into: will be communicated with object (connectedcomponents) again and exclude noise with image analysing computer, to form the object 20 that desire is followed the trail of;
Step S160 object is followed the trail of: when object 20 become follow the trail of the objective after, the direction, speed and the focal length that promptly see through control ball-type video camera 10 come automatically with clapping object 20, in order to the tracking of carrying out object 20.
Wherein this step S160 object is followed the trail of and further to be contained the following step: step S161 and judge that whether object is for following the trail of the objective: affirmation object 20 follows the trail of the objective for desiring, and begins to follow the trail of; Step S162 judges the size in picture that follows the trail of the objective: when following the trail of the objective in picture too greatly or too hour, the multiplying power of promptly adjusting ball-type video camera 10 is can meet the form of picture; Whether step S163 judges and to follow the trail of the objective near the picture edge: when following the trail of the objective too near the picture edge, promptly adjust ball-type video camera 10 speed and follow and follow the trail of the objective; Not when following the trail of the objective wherein when step S161 judges object; Promptly carry out next step: step S1611 and judge whether lose the time of following the trail of the objective surpasses critical value: when surpassing critical value, follow the trail of the objective and lose and finish to follow the trail of; Otherwise when not surpassing critical value, reduce when occurring when following the trail of the objective ball-type video camera 10 speed follow follow the trail of the objective mobile; And this image analysing computer further comprises position (position), size (size), direct of travel (direction), color (color) or the texture analyses such as (texture) of this object 20, can more promote the accurate of object 20 judgements; In addition, this each huge collection block (macroblock) is the NxN pixel, and N is 8,16,32 or other numerals; Moreover this motion vector further includes position vector, motion vector, velocity vector and vector acceleration; In addition, this step S120 judges that the scene image that promptly carries out next step S130 carried out binaryzation when the motion state of present ball-type video camera further was stationary state.
Please refer to Fig. 1 to shown in Figure 3, be the schematic flow sheet and the enforcement view of the automatic method of following the trail of of ball-type video camera of the present invention.Optimum operation principle of the present invention is followed the trail of on the mobile object (object 20 of the present invention is automobile) for being applied in ball-type video camera 10 automatically; Via PTZ camera lens 11 fast moving to shoot with video-corder scene image 30; Open static image and analyze and follow the trail of and in one second, can shoot with video-corder Y through the PTZ camera lens 11 of method of the present invention; So static image only needs just can accomplish all steps flow charts (as shown in Figure 1) less than 1/Y in second (Y can be 15 or other numerals), so in step S100, obtain an image: capture a static image in the picture of being kept watch on through PTZ camera lens 10; The scene image of shooting with video-corder 30 is being incident upon on the image sensor; And this image sensor is charge coupled cell inductor (CCD; Charge-coupled Device) or CMOS image sensor (CMOS; ComplementaryMetal Oxide Semiconductor) wherein a kind of; From image sensor, capturing many pieces of static image carrying out the analysis and the tracking of next step, and obtain the motion vector of image: image is cut into a plurality of huge collection blocks (macroblock), and obtain a motion vector (motion vector) by each huge collection block (macroblock) in the image at step S110; After from image sensor, capturing one static image, be about to image and cut into a plurality of huge collection blocks (macroblock); Wherein this each huge collection block (macroblock) is the NxN pixel; And N is 8,16,32 or other numerals; From huge collection block (macroblock), obtain motion vector again; This motion vector comprises X component and Y component; And each component all has positive and negative branch; And motion vector includes position vector, motion vector, velocity vector and vector acceleration; Wherein position vector is pointed to particle postition (x by true origin; Y) vector;
size
direction for x axle clamp angle
motion vector be the variable quantity of position vector; Point to the vector of last position by first position;
moreover; This velocity vector divides again has average velocity:
and instantaneous velocity:
and vector acceleration divides average acceleration is arranged:
and instantaneous acceleration:
therefore can learn that through the motion vector in the huge collection block (macroblock) which side has object moving in the scene image 30; And estimate its travel direction and speed; Whether there is object moving in the scene image 30 to enable to judge; And can carry out next step; And step S120 judges the motion state of present ball-type video camera: when ball-type video camera 10 is mobile status; Can cooperate information such as present direction of motion of ball-type video camera 10 and speed, with the motion vector reduction of moving scene; Because ball-type video camera 10 frequent fast moving PTZ camera lenses 11; Can let the motion vector of image become useless information; So cooperate ball-type video camera 10 present motion (moving) direction and speed information such as (rotating speeds), to enable that the motion vector in the scene image 30 was also judged the real motion vector of object originally, wherein this step S120 motion state of judging present ball-type video camera is when being stationary state; The object that can judge in the scene image 30 is in moving; And can calculate its travel direction and speed, therefore after the motion state of present ball-type video camera 10 is judged, can carry out next step, step S130 scene image carries out binaryzation: after the motion vector reduction; Promptly carry out the scene image binaryzation, can find out the embryo of object in the scene; Which side motion vector when huge collection block (macroblock) from each judges in the scene image 30 has object in motion; And after estimating its travel direction and speed; Promptly utilize grey-tone image binaryzation (B&W) with forming manifesting of object (Object) image and background (Background) image in the scene image; Be made as gray scale maximum value to pixel grey scale during its scene image binaryzation greater than certain critical gray-scale value; Be made as the gray scale minimal value to pixel grey scale less than this value; Thereby the realization binaryzation, and the main task of this work is correctly to find out the edge and the lines of object (Object), so that find out the embryo of object in the scene; Therefore when the embryo of object manifests, can carry out next step, step S140 is cut apart the connection object in the motion vector: from the embryo of scene image binaryzation, be partitioned into the connection object (connectedcomponents) with motion vector; After each huge collection block (macroblock) is via the scene image binaryzation; The object embryo that will have motion vector displays; And then can be partitioned into connection object (connected components) with motion vector; To obtain appearance and the kenel that is communicated with object; Be convenient to judge whether be communicated with object is the object 20 that desire is followed the trail of; Therefore after connection object (connectedcomponents) branch with motion vector cuts off, can carry out next step, step S150 analyzes and also follows the trail of the connection object that is partitioned into: will be communicated with object (connected components) again and exclude noise with image analysing computer, to form the object 20 that desire is followed the trail of; Be communicated with object because of including a large amount of noises yet be partitioned into; Make the connection object that splits reach really inadequately reliably; And cause the phenomenon of object 20 trace errors that desire follows the trail of easily; So need to exclude noise through image analysing computer again, wherein this image analysing computer includes position (position), size (size), direct of travel (direction), color (color) or texture (texture) analysis of etc.ing of this object 20, makes noise is filtered and can demonstrate than the really and reliable connection object of tool; And then can more promote the accurate of object 20 judgements; Whether so that judge is the object 20 that desire is followed the trail of, and therefore after image analysing computer is accomplished, promptly carries out next step, the tracking of step S160 object: when object 20 become follow the trail of the objective after; Promptly direction, speed and the focal length through control ball-type video camera 10 comes automatically with clapping object 20, in order to the tracking of carrying out object 20; And when be communicated with object through noise is filtered become an object 20 after; Whether promptly analyze is the object 20 that desire is followed the trail of; Therefore the object 20 that its desire is followed the trail of after analyzing is confirmed as when following the trail of the objective; Can follow the trail of the objective with clapping this automatically through direction, speed and the focal length of control ball-type video camera 10 are next immediately, in order to the automatic tracking of carrying out object 20.
And the object of following the trail of when desire 20 is confirmed as and can be carried out the following step (as shown in Figure 2) when following the trail of the objective, and step S161 judges that whether object is for following the trail of the objective: affirmation object 20 follows the trail of the objective for desiring, and begins to follow the trail of; Through confirming object 20 for after desiring to follow the trail of the objective, can be immediately direction, speed and focal length through control ball-type video camera 10 come follow the trail of the objective with clapping this automatically, and when step S161 judgement object 20 be not when following the trail of the objective; Promptly carry out another step; Step S1611 judges whether lose the time of following the trail of the objective surpasses critical value: when surpassing critical value, follow the trail of the objective and lose and finish to follow the trail of, so the time of general critical value approximately all be made as 3 to 5 seconds, so in case this follows the trail of the objective very likely not in 10 scopes that can shoot with video-corder of ball-type video camera above time of critical value; Or covered by barrier; Or follow the trail of the objective with staggered etc. promptly forfeiture of object, and this tracking that follows the trail of the objective that is through with, so that the tracking that newly follows the trail of the objective again; Otherwise; Surpass and set critical value during the time, it follows the trail of the objective when coming across in 10 scopes that can shoot with video-corder of ball-type video camera, immediately with the speed that reduces ball-type video camera 10 follow follow the trail of the objective mobile; Making ball-type video camera 10 can shoot with video-corder this follows the trail of the objective; And (as shown in Figure 3) promptly carries out next step in the PTZ camera lens 11 that comes across ball-type video camera 10 when following the trail of the objective, and step S162 judges the size in picture that follows the trail of the objective: when following the trail of the objective in picture too greatly or too hour, the multiplying power of promptly adjusting ball-type video camera 10 is can meet the form of picture; Via object 20 become follow the trail of the objective after; The orientation that this PTZ camera lens 11 promptly can and then follow the trail of the objective, displacement, speed move and adjustment, and follow the trail of the objective in picture when too big when discovery, and this ball-type video camera 10 can be adjusted reduction magnification automatically; Can be positioned at the centre of PTZ camera lens 11 so that follow the trail of the objective; Otherwise when finding to follow the trail of the objective in picture too hour, this ball-type video camera 10 also can be adjusted enlargement ratio automatically; Can be positioned at the centre of PTZ camera lens 11 so that follow the trail of the objective; No matter make to follow the trail of the objective far away or near, the PTZ camera lens 11 of its ball-type video camera 10 all can track this and follow the trail of the objective, and carries out next step when being arranged in PTZ camera lens 11 when following the trail of the objective; Whether step S163 judges and to follow the trail of the objective near the picture edge: when following the trail of the objective too near the picture edge, promptly adjust ball-type video camera 10 speed and follow and follow the trail of the objective; And the PTZ camera lens 11 of ball-type video camera 10 is being shot with video-corder when following the trail of the objective, sometimes because itself the speed of following the trail of the objective neglect fast neglect slow so that PTZ camera lens 11 in tracing process its follow the trail of the objective can be because of soon running out of outside the PTZ camera lens 11 near the picture edge very much; Need to accelerate the speed of ball-type video camera 10 this moment through adjustment; Make PTZ camera lens 11 can accelerate to move, follow the trail of the objective can catch up with this toward the direction that follows the trail of the objective, otherwise; As follow the trail of the objective not near the fast words in picture edge; As long as the speed of fine setting ball-type video camera 10 makes PTZ camera lens 11 can follow the direction that follows the trail of the objective and moves, until follows the trail of the objective not in 10 scopes that can shoot with video-corder of ball-type video camera; Follow the trail of the objective and lost, and then finish the tracking that this follows the trail of the objective.
Know that by above the method for tracking of the present invention has following advantage:
Even 1, the object followed the trail of of this desire quickens suddenly, slow down, covered or staggered with another object by barrier, very excellent tracking effect person can be arranged.
2, method of the present invention is fit to be implemented in the relatively poor embedded platform of hardware usefulness, and does not need extra signal processing integrated circuit (DSP) auxiliary, the more smooth person that just can let the ball-type video camera move.
Through above detailed description, can make and know the clear the present invention of this art and can reach aforementioned purpose really, met the regulation of Patent Law, the whence proposes patented claim.
Only the above person is merely preferred embodiment of the present invention, when not limiting the scope that the present invention implements with this; So all simple equivalent of doing according to claim of the present invention and description change and modify, all should still belong in the scope that patent of the present invention contains.
Claims (7)
1. the ball-type video camera method of following the trail of automatically is characterized in that the method key step of this tracking comprises:
Obtain an image: from the picture that the PTZ camera lens is kept watch on, capture a static image;
Obtain the motion vector of image: image is cut into a plurality of huge collection blocks, and obtain a motion vector by each huge collection block in the image;
Judge the motion state of present ball-type video camera: when the ball-type video camera is mobile status, can cooperate present direction of motion of ball-type video camera and speed, with the motion vector reduction of moving scene;
Scene image carries out binaryzation: after the motion vector reduction, promptly carry out the scene image binaryzation, can find out the embryo of object in the scene;
Cut apart the connection object in the motion vector: from the embryo of scene image binaryzation, be partitioned into connection object with motion vector;
Analysis is also followed the trail of the connection object that is partitioned into: will be communicated with object again and exclude noise with image analysing computer, to form the object that desire is followed the trail of; And
Object is followed the trail of: when object become follow the trail of the objective after, promptly follow the bat object automatically by direction, speed and the focal length of control ball-type video camera.
2. the method that ball-type video camera as claimed in claim 1 is followed the trail of automatically is characterized in that, this object is followed the trail of and further contained the following step:
Judge that whether object is for following the trail of the objective: confirm that object follows the trail of the objective for desiring, and begin to follow the trail of;
Judge the size in picture that follows the trail of the objective: when following the trail of the objective in picture too greatly or too hour, the multiplying power of promptly adjusting the ball-type video camera is can meet the form of picture; And
Judge and to follow the trail of the objective: when following the trail of the objective, promptly adjust ball-type video camera speed and follow and follow the trail of the objective near the picture edge whether near the picture edge.
3. the method that ball-type video camera as claimed in claim 2 is followed the trail of automatically; It is characterized in that; When judging that object is not when following the trail of the objective; Promptly carry out next step: judge whether lose the time of following the trail of the objective surpasses critical value: when surpassing critical value, follow the trail of the objective and lose and finish to follow the trail of, otherwise when surpassing critical value, reduce when occurring when following the trail of the objective ball-type video camera speed follow follow the trail of the objective mobile.
4. the method that ball-type video camera as claimed in claim 1 is followed the trail of automatically is characterized in that this image analysing computer further comprises position, size, direct of travel, color or the texture analysis of this object.
5. the method that ball-type video camera as claimed in claim 1 is followed the trail of automatically is characterized in that this each huge collection block is the NxN pixel, and N is 8,16 or 32.
6. the method that ball-type video camera as claimed in claim 1 is followed the trail of automatically is characterized in that this motion vector further includes position vector, motion vector, velocity vector and vector acceleration.
7. the method that ball-type video camera as claimed in claim 1 is followed the trail of automatically is characterized in that, this judges that the scene image that promptly carries out next step carried out binaryzation when the motion state of present ball-type video camera further was stationary state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100876619A CN102737384A (en) | 2011-04-08 | 2011-04-08 | Automatic spherical camera tracing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100876619A CN102737384A (en) | 2011-04-08 | 2011-04-08 | Automatic spherical camera tracing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102737384A true CN102737384A (en) | 2012-10-17 |
Family
ID=46992808
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011100876619A Pending CN102737384A (en) | 2011-04-08 | 2011-04-08 | Automatic spherical camera tracing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102737384A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102932598A (en) * | 2012-11-06 | 2013-02-13 | 苏州科达科技股份有限公司 | Method for intelligently tracking image on screen by camera |
| CN106780674A (en) * | 2016-11-28 | 2017-05-31 | 网易（杭州）网络有限公司 | Camera lens moving method and device |
| CN107431761A (en) * | 2015-04-14 | 2017-12-01 | 索尼公司 | Image processing equipment, image processing method and image processing system |
| CN110290351A (en) * | 2019-06-26 | 2019-09-27 | 广东康云科技有限公司 | A kind of video target tracking method, system, device and storage medium |
| CN111611837A (en) * | 2019-02-23 | 2020-09-01 | 和硕联合科技股份有限公司 | Tracking system and method thereof |
| CN109782604B (en) * | 2017-07-18 | 2022-05-06 | 北京神州启航科技发展有限公司 | 4D film seat control system based on neural network |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1980384A (en) * | 2006-09-27 | 2007-06-13 | 林进灯 | Space mobile-object locking aim-searching device and method |
| CN101043616A (en) * | 2007-02-07 | 2007-09-26 | 北京航空航天大学 | Method for controlling automatically video camera moving according image position of objects |
| CN101699862A (en) * | 2009-11-16 | 2010-04-28 | 上海交通大学 | High-resolution region-of-interest image acquisition method of PTZ camera |
-
2011
- 2011-04-08 CN CN2011100876619A patent/CN102737384A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1980384A (en) * | 2006-09-27 | 2007-06-13 | 林进灯 | Space mobile-object locking aim-searching device and method |
| CN101043616A (en) * | 2007-02-07 | 2007-09-26 | 北京航空航天大学 | Method for controlling automatically video camera moving according image position of objects |
| CN101699862A (en) * | 2009-11-16 | 2010-04-28 | 上海交通大学 | High-resolution region-of-interest image acquisition method of PTZ camera |
Non-Patent Citations (3)
| Title |
|---|
| 吴思 等: "基于动态背景构造的视频运动对象自动分割", 《计算机学报》 * |
| 王洪斌 等: "基于全局运动补偿的多运动目标检测方法研究", 《计算机技术与应用》 * |
| 陆军 等: "摄像机运动下的动态目标检测与跟踪", 《哈尔滨工程大学学报》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102932598A (en) * | 2012-11-06 | 2013-02-13 | 苏州科达科技股份有限公司 | Method for intelligently tracking image on screen by camera |
| CN102932598B (en) * | 2012-11-06 | 2015-04-08 | 苏州科达科技股份有限公司 | Method for intelligently tracking image on screen by camera |
| CN107431761A (en) * | 2015-04-14 | 2017-12-01 | 索尼公司 | Image processing equipment, image processing method and image processing system |
| CN107431761B (en) * | 2015-04-14 | 2021-02-19 | 索尼公司 | Image processing apparatus, image processing method, and image processing system |
| CN106780674A (en) * | 2016-11-28 | 2017-05-31 | 网易（杭州）网络有限公司 | Camera lens moving method and device |
| CN106780674B (en) * | 2016-11-28 | 2020-08-25 | 网易（杭州）网络有限公司 | Lens moving method and device |
| CN109782604B (en) * | 2017-07-18 | 2022-05-06 | 北京神州启航科技发展有限公司 | 4D film seat control system based on neural network |
| CN111611837A (en) * | 2019-02-23 | 2020-09-01 | 和硕联合科技股份有限公司 | Tracking system and method thereof |
| CN110290351A (en) * | 2019-06-26 | 2019-09-27 | 广东康云科技有限公司 | A kind of video target tracking method, system, device and storage medium |
| CN110290351B (en) * | 2019-06-26 | 2021-03-23 | 广东康云科技有限公司 | Video target tracking method, system, device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108111818B (en) | Moving target actively perceive method and apparatus based on multiple-camera collaboration | |
| CN103325112B (en) | Moving target method for quick in dynamic scene | |
| CN106375706B (en) | method and device for measuring speed of moving object by using double cameras and mobile terminal | |
| CN102737384A (en) | Automatic spherical camera tracing method | |
| CN107563347A (en) | A kind of passenger flow counting method and apparatus based on TOF camera | |
| US20100141806A1 (en) | Moving Object Noise Elimination Processing Device and Moving Object Noise Elimination Processing Program | |
| US20060285724A1 (en) | Salient motion detection system, method and program product therefor | |
| CN109886995B (en) | Multi-target tracking method in complex environment | |
| CN109981972B (en) | Target tracking method of robot, robot and storage medium | |
| US20110074927A1 (en) | Method for determining ego-motion of moving platform and detection system | |
| CN106033614B (en) | A kind of mobile camera motion object detection method under strong parallax | |
| CN107547865A (en) | Trans-regional human body video frequency object tracking intelligent control method | |
| KR101202642B1 (en) | Method and apparatus for estimating global motion using the background feature points | |
| CN114724094A (en) | System for measuring number of people in gateway vehicle based on three-dimensional image and radar technology | |
| Tang et al. | A fast video stabilization algorithm based on block matching and edge completion | |
| JP2017092839A5 (en) | ||
| Zhou et al. | Rain detection and removal of sequential images | |
| CN105607218A (en) | Image auto-focusing method measurement data transmission device and method based on fuzzy entropy | |
| TWI453698B (en) | The method of automatic tracking of ball camera | |
| KR101161557B1 (en) | The apparatus and method of moving object tracking with shadow removal moudule in camera position and time | |
| Sincan et al. | Moving object detection by a mounted moving camera | |
| CN115690190B (en) | Moving target detection and positioning method based on optical flow image and pinhole imaging | |
| CN106556958A (en) | The auto focusing method of Range-gated Imager | |
| TWI479857B (en) | PTZ camera automatic tracking method | |
| JP2009068935A (en) | Stereo measurement apparatus, stereo measurement method, and stereo measurement processing program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20121017 |